From c0b149ae96debd78718ad86af44673d53f126128 Mon Sep 17 00:00:00 2001 From: Mark Goble Date: Sun, 6 Nov 2016 13:11:49 +0000 Subject: [PATCH] fix to ensure LaTeX can be knitted to PDF Changed to use align rather than eqnarray --- manuscript/02_probability.md | 934 +++++----- manuscript/03_conditional.md | 4 +- manuscript/LittleInferenceBook.Rmd | 7 +- manuscript/LittleInferenceBook.html | 2647 +++++++++++++++++++++++++++ manuscript/LittleInferenceBook.pdf | Bin 0 -> 620923 bytes 5 files changed, 3121 insertions(+), 471 deletions(-) create mode 100644 manuscript/LittleInferenceBook.html create mode 100644 manuscript/LittleInferenceBook.pdf diff --git a/manuscript/02_probability.md b/manuscript/02_probability.md index 6426788..4e7f769 100644 --- a/manuscript/02_probability.md +++ b/manuscript/02_probability.md @@ -1,467 +1,467 @@ -# Probability -[Watch this video before beginning.](http://youtu.be/oTERv_vrmJM?list=PLpl-gQkQivXiBmGyzLrUjzsblmQsLtkzJ) - -Probability forms the foundation for almost all treatments of statistical inference. -In our treatment, probability is a law that assigns numbers to the long run occurrence of random phenomena after repeated unrelated realizations. - -Before we begin discussing probability, let's dispense with some deep philosophical questions, such as "What is randomness?" and "What is the fundamental interpretation of probability?". One could spend a lifetime studying these questions (and some have). For our purposes, randomness is any process occurring without apparent deterministic patterns. Thus we will treat many things as if they were random when, in fact they are completely deterministic. In my field, biostatistics, we often model disease outcomes as if they were random when they are the result of many mechanistic components whose aggregate behavior appears random. Probability for us will be the long run proportion of times some occurs in repeated unrelated realizations. So, think of the proportion of times that you get a head when flipping a coin. - -For the interested student, I would recommend the books and work by Ian Hacking to learn more about these deep philosophical issues. For us data scientists, the above definitions will work fine. - - -## Where to get a more thorough treatment of probability - -In this lecture, we will cover the fundamentals of probability at low enough of -a level to have a basic understanding for the rest of the series. For a more -complete treatment see the class Mathematical Biostatistics Boot Camp 1, which -can be viewed on YouTube [here](Youtube: -www.youtube.com/playlist?list=PLpl-gQkQivXhk6qSyiNj51qamjAtZISJ-). In addition, -there's the actual [Coursera course](Coursera: -www.coursera.org/course/biostats) that I run periodically (this is the first -Coursera class that I ever taught). Also there are a set of [notes on -GitHub](http://github.com/bcaffo/Caffo-Coursera). Finally, there's a follow up -class, uninspiringly named Mathematical Biostatistics Boot Camp 2, that is more -devoted to biostatistical topics that has an associated [YouTube -playlist](http://www.youtube.com/playlist?list=PLpl-gQkQivXhwOsKPQ4fbCBYOWjvdzrSM), -[Coursera Class](https://www.coursera.org/course/biostats2) and [GitHub -notes](https://github.com/bcaffo/MathematicsBiostatisticsBootCamp2). - -## Kolmogorov's Three Rules - -[Watch this lecture before -beginning.](http://youtu.be/Shzt9uZ8BII?list=PLpl-gQkQivXiBmGyzLrUjzsblmQsLtkzJ) - -Given a random experiment (say rolling a die) a probability measure is a -population quantity - that summarizes the randomness. The brilliant discovery of -the father of probability, the [Russian mathematician -Kolmogorov](http://en.wikipedia.org/wiki/Andrey_Kolmogorov), was that - to -satisfy our intuition about how probability should behave, only three rules were -needed. - -Consider an experiment with a random outcome. Probability takes a possible -outcome from an experiment and: - -1. assigns it a number between 0 and 1 -2. requires that the probability that -something occurs is 1 -3. required that the probability of the union of any two -sets of outcomes that have nothing in common (mutually exclusive) - is the sum of -their respective probabilities. - -From these simple rules all of the familiar rules of probability can be -developed. This all might seem a little odd at first and so we'll build up our -intuition with some simple examples based on coin flipping and die rolling. - -I would like to reiterate the important definition that we wrote out: *mutually -exclusive*. Two events are mutually exclusive if they cannot both -simultaneously occur. For example, we cannot simultaneously get a 1 and a 2 on a -die. Rule 3 says that since the event of getting a 1 and 2 on a die are -mutually exclusive, the probability of getting at least one (the union) is the -sum of their probabilities. So if we know that the probability of getting a 1 is -1/6 and the probability of getting a 2 is 1/6, then the probability of getting a -1 or a 2 is 2/6, the sum of the two probabilities since they are mutually -exclusive. - -## Consequences of The Three Rules - -Let's cover some consequences of our three simple rules. Take, for example, the -probability that something occurs is 1 minus the probability of the opposite -occurring. Let {$$}A{/$$} be the event that we get a 1 or a 2 on a rolled die. -Then {$$}A^c{/$$} is the opposite, getting a 3, 4, 5 or 6. Since {$$}A{/$$} and -{$$}A^c{/$$} -cannot both simultaneously occur, they are mutually exclusive. So -the probability that either {$$}A{/$$} or {$$}A^c{/$$} is {$$}P(A) + P(A^c){/$$}. -Notice, that the probability that either occurs is the probability -of getting a 1, 2, 3, 4, 5 or 6, or in other words, the probability that -something occurs, which is 1 by rule number 2. So we have that -{$$}1 = P(A) + P(A^c){/$$} or that {$$}P(A) = 1 - P(A^c){/$$}. - -We won't go through this tedious exercise (since Kolmogorov already did it for -us). Instead here's a list of some of the consequences of Kolmogorov's rules -that are often useful. - -1. The probability that nothing occurs is 0 -2. The probability that something occurs is 1 -3. The probability of something is 1 minus the probability that the opposite occurs -4. The probability of at least one of two (or more) things that can not simultaneously occur (mutually exclusive) is the sum of their respective probabilities -5. For any two events the probability that at least one occurs is the sum of their probabilities minus their intersection. - -This last rules states that {$$}P(A \cup B) = P(A) + P(B) - P(A \cap B){/$$} -shows what is the issue with adding probabilities that are not mutually -exclusive. If we do this, we've added the probability that both occur in twice! -(Watch the video where I draw a Venn diagram to illustrate this). - -### Example of Implementing Probability Calculus - -The National Sleep Foundation -([www.sleepfoundation.org](http://www.sleepfoundation.org/)) reports that around -3% of the American population has sleep apnea. They also report that around 10% -of the North American and European population has restless leg syndrome. Does -this imply that 13% of people will have at least one sleep problems of these -sorts? In other words, can we simply add these two - probabilities? - -Answer: No, the events can simultaneously occur and so - are not mutually -exclusive. To elaborate let: - -{$$} - \begin{eqnarray*} - A_1 & = & \{\mbox{Person has sleep apnea}\} \\ -A_2 & = & \{\mbox{Person has RLS}\} - \end{eqnarray*} - {/$$} - -Then - -{$$} - \begin{eqnarray*} - P(A_1 \cup A_2 ) & = & P(A_1) + P(A_2) - P(A_1 \cap -A_2) \\ - & = & 0.13 - \mbox{Probability of having both} - \end{eqnarray*} - {/$$} - -Given the scenario, it's likely that some fraction of the population has both. -This example serves as a reminder *don't add probabilities unless the events are -mutually exclusive*. We'll have a similar rule for multiplying probabilities and -independence. - -## Random variables -[Watch this video before reading this section](http://youtu.be/Shzt9uZ8BII?list=PLpl-gQkQivXiBmGyzLrUjzsblmQsLtkzJ) - -Probability calculus is useful for understanding the rules that probabilities must follow. However, we need ways to model and think about probabilities for -numeric outcomes of experiments (broadly defined). - Densities and mass functions -for random variables are the best starting point for this. You've already - heard -of a density since you've heard of the famous "bell curve", or Gaussian -density. In this section - you'll learn exactly what the bell curve is and how to -work with it. - -Remember, everything we're talking about up to at this point is a population -quantity, - not a statement about what occurs in our data. Think about the fact that -50% probability for head is a statement about - the coin and how we're flipping -it, not a statement about the percentage of heads we obtained in a particular -set of flips. This is an important distinction that we will emphasize over and -over in this course. Statistical - inference is about describing populations -using data. Probability density functions are a way to mathematically -characterize the population. In this course, we'll assume that our sample is a -random draw from the population. - -So our definition is that a **random variable** is a numerical outcome of an -experiment. - The random variables that we study will come in two varieties, -**discrete** or **continuous**. Discrete random variables are random variables -that take on only a - countable number of possibilities. Mass functions will -assign probabilities that they - take specific values. Continuous random -variable can conceptually take any value on the real line or - some subset of the -real line and we talk about the probability that they lie within some range. -Densities - will characterize these probabilities. - -Let's consider some examples of measurements that could be considered random variables. -First, familiar gambling experiments like the tossing of a coin and the rolling -of a die produce random variables. For the coin, we typically code a tail as -a 0 and a head as a 1. (For the die, the number facing up would be the random -variable.) We will use these examples a lot to help us build intuition. However, -they aren't interesting in the sense of seeming very contrived. Nonetheless, -the coin example is particularly useful since many of the experiments we -consider will be modeled as if tossing a biased coin. Modeling -any binary characteristic from a random sample of a population can be -thought of as a coin toss, with the random sampling performing the roll of the -toss and the population percentage of individuals with the characteristic -is the probability of a head. Consider, for example, logging whether or -not subjects were hypertensive in a random sample. Each subject's -outcome can be modeled as a coin toss. In a similar sense the die roll serves -as our model for phenomena with more than one level, such as hair color or -rating scales. - -Consider also the random variable of the number of web hits for a site each day. -This variable is a count, but is largely unbounded (or at least we couldn't -put a specific reasonable upper limit). Random variables like this are often modeled -with the so called Poisson distribution. - -Finally, consider some continuous random variables. Think of things like -lengths or weights. It is mathematically convenient to model these -as if they were continuous (even if measurements were truncated liberally). -In fact, even discrete random variables with lots of levels are often -treated as continuous for convenience. - -For all of these kinds of random variables, we need convenient mathematical -functions to model the probabilities of collections of realizations. These -functions, called mass functions and densities, -take possible values of the random variables, and assign the associated -probabilities. These entities describe the population of interest. So, consider -the most famous density, the normal distribution. Saying that body mass -indices follow a normal distribution is a statement about the population of -interest. The goal is to use our data to figure out things about that normal -distribution, where it's centered, how spread out it is and even -whether our assumption of normality is warranted! - -## Probability mass functions - -A probability mass function evaluated at a value corresponds to the - probability -that a random variable takes that value. To be a valid pmf a function, {$$}p{/$$}, -must satisfy: - -1. It must always be larger than or equal to 0. -2. The sum of the possible values that the random variable can take has to add up to one. - - -### Example - -Let {$$}X{/$$} be the result of a coin flip where {$$}X=0{/$$} -represents tails and {$$}X = 1{/$$} -represents heads. {$$}p(x) = (1/2)^{x} (1/2)^{1-x}{/$$} -for {$$}x = 0,1{/$$}. -Suppose that we do not know whether or not the coin is fair; Let - {$$}\theta{/$$} be -the probability of a head expressed as a proportion - (between 0 and 1). -{$$}p(x) = \theta^{x} (1 - \theta)^{1-x} {/$$} -for {$$}x = 0,1 {/$$} - -## Probability density functions - -[Watch this video before beginning.](http://youtu.be/mPe0Us4VYDM?list=PLpl-gQkQivXiBmGyzLrUjzsblmQsLtkzJ) - -A probability density function (pdf), is a function associated with a continuous -random variable. Because of the peculiarities -of treating measurements as having been recorded to infinite -decimal expansions, we need a different set of rules. This -leads us to the central dogma of probability density functions: - -*Areas under PDFs correspond to probabilities for that random variable* - -Therefore, when one says that intelligence quotients (IQ) in population follows a bell -curve, they are saying that the probability of a randomly -selected person from this population having an IQ between -two values is given by the area under the bell curve. - -Not every function can be a valid probability density -function. For example, if the function dips below zero, -then we could have negative probabilities. If the function -contains too much area underneath it, we could have -probabilities larger than one. The following two -rules tell us when a function is a valid probability -density function. - -Specifically, to be a valid pdf, a function must satisfy - -1. It must be larger than or equal to zero everywhere. -2. The total area under it must be one. - -### Example - -Suppose that the proportion of help calls that get addressed in -a random day by a help line is given by {$$} f(x) = 2 x {/$$} -for {$$}0< x < 1{/$$}. The R code for plotting this density is - -{title="Code for plotting the density", line-numbers=off,lang=r} -~~~~~~ -x <- c(-0.5, 0, 1, 1, 1.5) -y <- c(0, 0, 2, 0, 0) -plot(x, y, lwd = 3,frame = FALSE, type = "l") -~~~~~~ - -The result of the code is given below. - -![Help call density](images/triangleDensity.png) - -Is this a mathematically valid density? To answer this -we need to make sure it satisfies our two conditions. -First it's clearly nonnegative (it's at or above -the horizontal axis everywhere). The area is similarly -easy. Being a right triangle in the only section of the -density that is above zero, we can calculate it as -1/2 the area of the base times the height. This -is {$$}\frac{1}{2} \times 1 \times 2 = 1{/$$} - - -Now consider answering the following question. -What is the probability that 75% or fewer of calls get addressed? -Remember, for continuous random variables, probabilities are represented -by areas underneath the density function. So, we want the area from -0.75 and below, as illustrated by the figure below. - -![Help call density](images/triangleDensityArea.png) - -This again is a right triangle, with length of the base as 0.75 -and height 1.5. The R code below shows the calculation. - -{line-numbers=off,lang=r} -~~~~~~ -> 1.5 * 0.75/2 - -[1] 0.5625 -~~~~~~ - -Thus, the probability of 75% or fewer calls getting addressed in a random -day for this help line is 56%. We'll do this a lot throughout this class -and work with more useful densities. It should be noted that this specific -density is a special case of the so called *beta* density. Below I show how -to use R's built in evaluation function for the beta density to get the -probability. - -{line-numbers=off,lang=r} -~~~~~~ - -> pbeta(0.75, 2, 1) - -[1] 0.5625 -~~~~~~ - -Notice the syntax `pbeta`. In R, a prefix of `p` returns probabilities, -`d` returns the density, `q` returns the quantile and `r` returns generated -random variables. (You'll learn what each of these does in subsequent sections.) - -## CDF and survival function - -Certain areas of PDFs and PMFs are so useful, we give them names. -The **cumulative distribution function** (CDF) of a random variable, {$$}X{/$$}, -returns the probability that the random variable is less than or equal to the -value {$$}x{/$$}. Notice the (slightly annoying) convention that we use an upper -case {$$}X{/$$} to denote a random, unrealized, version of the random variable -and a lowercase {$$}x{/$$} to denote a specific number that we plug into. -(This notation, as odd as it may seem, dates back to Fisher and isn't going -anywhere, so you might as well get used to it. Uppercase for unrealized random -variables and lowercase as placeholders for numbers to plug into.) So we -could write the following to describe the distribution function -{$$}F{/$$}: - - -{$$} -F(x) = P(X \leq x) -{/$$} - -This definition applies regardless of -whether the random variable is discrete or continuous. The **survival function** -of a random variable {$$}X{/$$} is defined as the -probability that the random variable is greater than the value {$$}x{/$$}. - -{$$} -S(x) = P(X > x) -{/$$} - - Notice that {$$}S(x) = 1 - F(x){/$$}, since the survival function evaluated - at a particular value of {$$}x{/$$} is calculating the probability of the - opposite event (greater than as opposed to less than or equal to). The - survival function is often preferred in biostatistical applications while - the distribution function is more generally used (though both convey the - same information.) - - -### Example - -What are the survival function and CDF from the density considered before? - -{$$} -F(x) = P(X \leq x) = \frac{1}{2} Base \times Height = \frac{1}{2} (x) \times (2 x) = x^2, -{/$$} - -for {$$}1 \geq x \geq 0{/$$}. Notice that calculating the survival function -is now trivial given that we've already calculated the distribution function. - -{$$} - S(x) = 1 = F(x) = 1 - x^2 -{/$$} - -Again, R has a function that calculates the distribution function for us -in this case, `pbeta`. Let's try calculating {$$}F(.4){/$$}, {$$}F(.5){/$$} -and {$$}F(.6){/$$} - -{line-numbers=off,lang=r} -~~~~~~ -> pbeta(c(0.4, 0.5, 0.6), 2, 1) - -[1] 0.16 0.25 0.36 - ~~~~~~ - -Notice, of course, these are simply the numbers squared. By default the prefix -`p` in front of a density in R gives the distribution function (`pbeta`, `pnorm`, - `pgamma`). If you want the survival function values, you could always subtract - by one, or give the argument `lower.tail = FALSE` as an argument to the function, - which asks R to calculate the upper area instead of the lower. - -## Quantiles - -You've heard of sample quantiles. If you were the 95th percentile on an exam, you know -that 95% of people scored worse than you and 5% scored better. -These are sample quantities. But you might have wondered, what are my sample -quantiles estimating? In fact, they are estimating the population quantiles. -Here we define these population analogs. - - -The {$$}\alpha^{th}{/$$} **quantile** of a distribution -with distribution function {$$}F{/$$} is the point {$$}x_\alpha{/$$} so that - -{$$} -F(x_\alpha) = \alpha -{/$$} - -So the 0.95 quantile of a distribution is the point so that 95% of the mass -of the density lies below it. Or, in other words, the point so that the -probability of getting a randomly sampled point below it is 0.95. This is -analogous to the sample -quantiles where the 0.95 sample quantile is the value so that 95% of the data -lies below it. - - A **percentile** is simply a quantile with {$$}\alpha{/$$} expressed as a percent - rather than a proportion. The (population) -**median** is the {$$}50^{th}{/$$} percentile. Remember that percentiles -are not probabilities! Remember that quantiles have units. So the population -median height is the height (in inches say) so that the probability that a randomly selected -person from the population is shorter is 50%. The sample, or empirical, -median would be the height so in a sample so that 50% of the people in the -sample were shorter. - -### Example -What is the median of the distribution that we were working with before? -We want to solve {$$}0.5 = F(x) = x^2{/$$}, resulting in the solution - -{line-numbers=off,lang=r} -~~~~~~ -> sqrt(0.5) - -[1] 0.7071 -~~~~~~ - -Therefore, 0.7071 of calls being answered on a random day is the median. -Or, the probability that 70% or fewer calls get answered is 50%. - -R can approximate quantiles for you for common distributions with the -prefix `q` in front of the distribution name - -{line-numbers=off,lang=r} -~~~~~~ -> qbeta(0.5, 2, 1) - -[1] 0.7071 -~~~~~~ - -## Exercises - -1. Can you add the probabilities of any two events to get the probability of at least one occurring? -2. I define a PMF, {$$}p{/$$} so that for {$$}x = 0{/$$} and {$$}x=1{/$$} we have -{$$}p(0) = -0.1{/$$} and {$$}p(1) = 1.1{/$$}. Is this a valid PMF? -3. What is the probability that 75% or fewer calls get answered in a randomly sampled day from the population distribution from this chapter? -4. The 97.5th percentile of a distribution is? -5. Consider influenza epidemics for two parent heterosexual families. Suppose that the probability is 15% that at least one of the parents has contracted the disease. The probability that the father has contracted influenza is 10% while that the mother contracted the disease is 9%. What is the probability that both contracted influenza expressed as a whole number percentage? -[Watch a video solution to this problem.](http://youtu.be/CvnmoCuIN08?list=PLpl-gQkQivXhHOcVeU3bSJg78zaDYbP9L) and -[see a written out solution.](http://bcaffo.github.io/courses/06_StatisticalInference/homework/hw1.html#3) -6. A random variable, {$$}X{/$$}, is uniform, a box from 0 to 1 of height 1. (So that it's density is {$$}f(x) = 1{/$$} for {$$}0\leq x \leq 1{/$$}.) -What is it's median expressed to two decimal places? [Watch a video solution to this problem -here](http://youtu.be/UXcarD-1xAM?list=PLpl-gQkQivXhHOcVeU3bSJg78zaDYbP9L) and [see written solutions here](http://bcaffo.github.io/courses/06_StatisticalInference/homework/hw1.html#4). -7. If a continuous density that never touches the horizontal axis is symmetric about zero, can we say that its associated median is zero? [Watch a worked out solution to this problem here](http://youtu.be/sn48CGH_TXI?list=PLpl-gQkQivXhHOcVeU3bSJg78zaDYbP9L) and [see the question and a typed up answer here](http://bcaffo.github.io/courses/06_StatisticalInference/homework/hw1.html#9) +# Probability +[Watch this video before beginning.](http://youtu.be/oTERv_vrmJM?list=PLpl-gQkQivXiBmGyzLrUjzsblmQsLtkzJ) + +Probability forms the foundation for almost all treatments of statistical inference. +In our treatment, probability is a law that assigns numbers to the long run occurrence of random phenomena after repeated unrelated realizations. + +Before we begin discussing probability, let's dispense with some deep philosophical questions, such as "What is randomness?" and "What is the fundamental interpretation of probability?". One could spend a lifetime studying these questions (and some have). For our purposes, randomness is any process occurring without apparent deterministic patterns. Thus we will treat many things as if they were random when, in fact they are completely deterministic. In my field, biostatistics, we often model disease outcomes as if they were random when they are the result of many mechanistic components whose aggregate behavior appears random. Probability for us will be the long run proportion of times some occurs in repeated unrelated realizations. So, think of the proportion of times that you get a head when flipping a coin. + +For the interested student, I would recommend the books and work by Ian Hacking to learn more about these deep philosophical issues. For us data scientists, the above definitions will work fine. + + +## Where to get a more thorough treatment of probability + +In this lecture, we will cover the fundamentals of probability at low enough of +a level to have a basic understanding for the rest of the series. For a more +complete treatment see the class Mathematical Biostatistics Boot Camp 1, which +can be viewed on YouTube [here](Youtube: +www.youtube.com/playlist?list=PLpl-gQkQivXhk6qSyiNj51qamjAtZISJ-). In addition, +there's the actual [Coursera course](Coursera: +www.coursera.org/course/biostats) that I run periodically (this is the first +Coursera class that I ever taught). Also there are a set of [notes on +GitHub](http://github.com/bcaffo/Caffo-Coursera). Finally, there's a follow up +class, uninspiringly named Mathematical Biostatistics Boot Camp 2, that is more +devoted to biostatistical topics that has an associated [YouTube +playlist](http://www.youtube.com/playlist?list=PLpl-gQkQivXhwOsKPQ4fbCBYOWjvdzrSM), +[Coursera Class](https://www.coursera.org/course/biostats2) and [GitHub +notes](https://github.com/bcaffo/MathematicsBiostatisticsBootCamp2). + +## Kolmogorov's Three Rules + +[Watch this lecture before +beginning.](http://youtu.be/Shzt9uZ8BII?list=PLpl-gQkQivXiBmGyzLrUjzsblmQsLtkzJ) + +Given a random experiment (say rolling a die) a probability measure is a +population quantity + that summarizes the randomness. The brilliant discovery of +the father of probability, the [Russian mathematician +Kolmogorov](http://en.wikipedia.org/wiki/Andrey_Kolmogorov), was that + to +satisfy our intuition about how probability should behave, only three rules were +needed. + +Consider an experiment with a random outcome. Probability takes a possible +outcome from an experiment and: + +1. assigns it a number between 0 and 1 +2. requires that the probability that +something occurs is 1 +3. required that the probability of the union of any two +sets of outcomes that have nothing in common (mutually exclusive) + is the sum of +their respective probabilities. + +From these simple rules all of the familiar rules of probability can be +developed. This all might seem a little odd at first and so we'll build up our +intuition with some simple examples based on coin flipping and die rolling. + +I would like to reiterate the important definition that we wrote out: *mutually +exclusive*. Two events are mutually exclusive if they cannot both +simultaneously occur. For example, we cannot simultaneously get a 1 and a 2 on a +die. Rule 3 says that since the event of getting a 1 and 2 on a die are +mutually exclusive, the probability of getting at least one (the union) is the +sum of their probabilities. So if we know that the probability of getting a 1 is +1/6 and the probability of getting a 2 is 1/6, then the probability of getting a +1 or a 2 is 2/6, the sum of the two probabilities since they are mutually +exclusive. + +## Consequences of The Three Rules + +Let's cover some consequences of our three simple rules. Take, for example, the +probability that something occurs is 1 minus the probability of the opposite +occurring. Let {$$}A{/$$} be the event that we get a 1 or a 2 on a rolled die. +Then {$$}A^c{/$$} is the opposite, getting a 3, 4, 5 or 6. Since {$$}A{/$$} and +{$$}A^c{/$$} +cannot both simultaneously occur, they are mutually exclusive. So +the probability that either {$$}A{/$$} or {$$}A^c{/$$} is {$$}P(A) + P(A^c){/$$}. +Notice, that the probability that either occurs is the probability +of getting a 1, 2, 3, 4, 5 or 6, or in other words, the probability that +something occurs, which is 1 by rule number 2. So we have that +{$$}1 = P(A) + P(A^c){/$$} or that {$$}P(A) = 1 - P(A^c){/$$}. + +We won't go through this tedious exercise (since Kolmogorov already did it for +us). Instead here's a list of some of the consequences of Kolmogorov's rules +that are often useful. + +1. The probability that nothing occurs is 0 +2. The probability that something occurs is 1 +3. The probability of something is 1 minus the probability that the opposite occurs +4. The probability of at least one of two (or more) things that can not simultaneously occur (mutually exclusive) is the sum of their respective probabilities +5. For any two events the probability that at least one occurs is the sum of their probabilities minus their intersection. + +This last rules states that {$$}P(A \cup B) = P(A) + P(B) - P(A \cap B){/$$} +shows what is the issue with adding probabilities that are not mutually +exclusive. If we do this, we've added the probability that both occur in twice! +(Watch the video where I draw a Venn diagram to illustrate this). + +### Example of Implementing Probability Calculus + +The National Sleep Foundation +([www.sleepfoundation.org](http://www.sleepfoundation.org/)) reports that around +3% of the American population has sleep apnea. They also report that around 10% +of the North American and European population has restless leg syndrome. Does +this imply that 13% of people will have at least one sleep problems of these +sorts? In other words, can we simply add these two + probabilities? + +Answer: No, the events can simultaneously occur and so + are not mutually +exclusive. To elaborate let: + +{$$} + \begin{aligned} + A_1 &= \{\mbox{Person has sleep apnea}\} \\ +A_2 &= \{\mbox{Person has RLS}\} + \end{aligned} + {/$$} + +Then + +{$$} + \begin{aligned} + P(A_1 \cup A_2 ) &= P(A_1) + P(A_2) - P(A_1 \cap +A_2) \\ + &= 0.13 - \mbox{Probability of having both} + \end{aligned} + {/$$} + +Given the scenario, it's likely that some fraction of the population has both. +This example serves as a reminder *don't add probabilities unless the events are +mutually exclusive*. We'll have a similar rule for multiplying probabilities and +independence. + +## Random variables +[Watch this video before reading this section](http://youtu.be/Shzt9uZ8BII?list=PLpl-gQkQivXiBmGyzLrUjzsblmQsLtkzJ) + +Probability calculus is useful for understanding the rules that probabilities must follow. However, we need ways to model and think about probabilities for +numeric outcomes of experiments (broadly defined). + Densities and mass functions +for random variables are the best starting point for this. You've already + heard +of a density since you've heard of the famous "bell curve", or Gaussian +density. In this section + you'll learn exactly what the bell curve is and how to +work with it. + +Remember, everything we're talking about up to at this point is a population +quantity, + not a statement about what occurs in our data. Think about the fact that +50% probability for head is a statement about + the coin and how we're flipping +it, not a statement about the percentage of heads we obtained in a particular +set of flips. This is an important distinction that we will emphasize over and +over in this course. Statistical + inference is about describing populations +using data. Probability density functions are a way to mathematically +characterize the population. In this course, we'll assume that our sample is a +random draw from the population. + +So our definition is that a **random variable** is a numerical outcome of an +experiment. + The random variables that we study will come in two varieties, +**discrete** or **continuous**. Discrete random variables are random variables +that take on only a + countable number of possibilities. Mass functions will +assign probabilities that they + take specific values. Continuous random +variable can conceptually take any value on the real line or + some subset of the +real line and we talk about the probability that they lie within some range. +Densities + will characterize these probabilities. + +Let's consider some examples of measurements that could be considered random variables. +First, familiar gambling experiments like the tossing of a coin and the rolling +of a die produce random variables. For the coin, we typically code a tail as +a 0 and a head as a 1. (For the die, the number facing up would be the random +variable.) We will use these examples a lot to help us build intuition. However, +they aren't interesting in the sense of seeming very contrived. Nonetheless, +the coin example is particularly useful since many of the experiments we +consider will be modeled as if tossing a biased coin. Modeling +any binary characteristic from a random sample of a population can be +thought of as a coin toss, with the random sampling performing the roll of the +toss and the population percentage of individuals with the characteristic +is the probability of a head. Consider, for example, logging whether or +not subjects were hypertensive in a random sample. Each subject's +outcome can be modeled as a coin toss. In a similar sense the die roll serves +as our model for phenomena with more than one level, such as hair color or +rating scales. + +Consider also the random variable of the number of web hits for a site each day. +This variable is a count, but is largely unbounded (or at least we couldn't +put a specific reasonable upper limit). Random variables like this are often modeled +with the so called Poisson distribution. + +Finally, consider some continuous random variables. Think of things like +lengths or weights. It is mathematically convenient to model these +as if they were continuous (even if measurements were truncated liberally). +In fact, even discrete random variables with lots of levels are often +treated as continuous for convenience. + +For all of these kinds of random variables, we need convenient mathematical +functions to model the probabilities of collections of realizations. These +functions, called mass functions and densities, +take possible values of the random variables, and assign the associated +probabilities. These entities describe the population of interest. So, consider +the most famous density, the normal distribution. Saying that body mass +indices follow a normal distribution is a statement about the population of +interest. The goal is to use our data to figure out things about that normal +distribution, where it's centered, how spread out it is and even +whether our assumption of normality is warranted! + +## Probability mass functions + +A probability mass function evaluated at a value corresponds to the + probability +that a random variable takes that value. To be a valid pmf a function, {$$}p{/$$}, +must satisfy: + +1. It must always be larger than or equal to 0. +2. The sum of the possible values that the random variable can take has to add up to one. + + +### Example + +Let {$$}X{/$$} be the result of a coin flip where {$$}X=0{/$$} +represents tails and {$$}X = 1{/$$} +represents heads. {$$}p(x) = (1/2)^{x} (1/2)^{1-x}{/$$} +for {$$}x = 0,1{/$$}. +Suppose that we do not know whether or not the coin is fair; Let + {$$}\theta{/$$} be +the probability of a head expressed as a proportion + (between 0 and 1). +{$$}p(x) = \theta^{x} (1 - \theta)^{1-x} {/$$} +for {$$}x = 0,1 {/$$} + +## Probability density functions + +[Watch this video before beginning.](http://youtu.be/mPe0Us4VYDM?list=PLpl-gQkQivXiBmGyzLrUjzsblmQsLtkzJ) + +A probability density function (pdf), is a function associated with a continuous +random variable. Because of the peculiarities +of treating measurements as having been recorded to infinite +decimal expansions, we need a different set of rules. This +leads us to the central dogma of probability density functions: + +*Areas under PDFs correspond to probabilities for that random variable* + +Therefore, when one says that intelligence quotients (IQ) in population follows a bell +curve, they are saying that the probability of a randomly +selected person from this population having an IQ between +two values is given by the area under the bell curve. + +Not every function can be a valid probability density +function. For example, if the function dips below zero, +then we could have negative probabilities. If the function +contains too much area underneath it, we could have +probabilities larger than one. The following two +rules tell us when a function is a valid probability +density function. + +Specifically, to be a valid pdf, a function must satisfy + +1. It must be larger than or equal to zero everywhere. +2. The total area under it must be one. + +### Example + +Suppose that the proportion of help calls that get addressed in +a random day by a help line is given by {$$} f(x) = 2 x {/$$} +for {$$}0< x < 1{/$$}. The R code for plotting this density is + +{title="Code for plotting the density", line-numbers=off,lang=r} +~~~~~~ +x <- c(-0.5, 0, 1, 1, 1.5) +y <- c(0, 0, 2, 0, 0) +plot(x, y, lwd = 3,frame = FALSE, type = "l") +~~~~~~ + +The result of the code is given below. + +![Help call density](images/triangleDensity.png) + +Is this a mathematically valid density? To answer this +we need to make sure it satisfies our two conditions. +First it's clearly nonnegative (it's at or above +the horizontal axis everywhere). The area is similarly +easy. Being a right triangle in the only section of the +density that is above zero, we can calculate it as +1/2 the area of the base times the height. This +is {$$}\frac{1}{2} \times 1 \times 2 = 1{/$$} + + +Now consider answering the following question. +What is the probability that 75% or fewer of calls get addressed? +Remember, for continuous random variables, probabilities are represented +by areas underneath the density function. So, we want the area from +0.75 and below, as illustrated by the figure below. + +![Help call density](images/triangleDensityArea.png) + +This again is a right triangle, with length of the base as 0.75 +and height 1.5. The R code below shows the calculation. + +{line-numbers=off,lang=r} +~~~~~~ +> 1.5 * 0.75/2 + +[1] 0.5625 +~~~~~~ + +Thus, the probability of 75% or fewer calls getting addressed in a random +day for this help line is 56%. We'll do this a lot throughout this class +and work with more useful densities. It should be noted that this specific +density is a special case of the so called *beta* density. Below I show how +to use R's built in evaluation function for the beta density to get the +probability. + +{line-numbers=off,lang=r} +~~~~~~ + +> pbeta(0.75, 2, 1) + +[1] 0.5625 +~~~~~~ + +Notice the syntax `pbeta`. In R, a prefix of `p` returns probabilities, +`d` returns the density, `q` returns the quantile and `r` returns generated +random variables. (You'll learn what each of these does in subsequent sections.) + +## CDF and survival function + +Certain areas of PDFs and PMFs are so useful, we give them names. +The **cumulative distribution function** (CDF) of a random variable, {$$}X{/$$}, +returns the probability that the random variable is less than or equal to the +value {$$}x{/$$}. Notice the (slightly annoying) convention that we use an upper +case {$$}X{/$$} to denote a random, unrealized, version of the random variable +and a lowercase {$$}x{/$$} to denote a specific number that we plug into. +(This notation, as odd as it may seem, dates back to Fisher and isn't going +anywhere, so you might as well get used to it. Uppercase for unrealized random +variables and lowercase as placeholders for numbers to plug into.) So we +could write the following to describe the distribution function +{$$}F{/$$}: + + +{$$} +F(x) = P(X \leq x) +{/$$} + +This definition applies regardless of +whether the random variable is discrete or continuous. The **survival function** +of a random variable {$$}X{/$$} is defined as the +probability that the random variable is greater than the value {$$}x{/$$}. + +{$$} +S(x) = P(X > x) +{/$$} + + Notice that {$$}S(x) = 1 - F(x){/$$}, since the survival function evaluated + at a particular value of {$$}x{/$$} is calculating the probability of the + opposite event (greater than as opposed to less than or equal to). The + survival function is often preferred in biostatistical applications while + the distribution function is more generally used (though both convey the + same information.) + + +### Example + +What are the survival function and CDF from the density considered before? + +{$$} +F(x) = P(X \leq x) = \frac{1}{2} Base \times Height = \frac{1}{2} (x) \times (2 x) = x^2, +{/$$} + +for {$$}1 \geq x \geq 0{/$$}. Notice that calculating the survival function +is now trivial given that we've already calculated the distribution function. + +{$$} + S(x) = 1 = F(x) = 1 - x^2 +{/$$} + +Again, R has a function that calculates the distribution function for us +in this case, `pbeta`. Let's try calculating {$$}F(.4){/$$}, {$$}F(.5){/$$} +and {$$}F(.6){/$$} + +{line-numbers=off,lang=r} +~~~~~~ +> pbeta(c(0.4, 0.5, 0.6), 2, 1) + +[1] 0.16 0.25 0.36 + ~~~~~~ + +Notice, of course, these are simply the numbers squared. By default the prefix +`p` in front of a density in R gives the distribution function (`pbeta`, `pnorm`, + `pgamma`). If you want the survival function values, you could always subtract + by one, or give the argument `lower.tail = FALSE` as an argument to the function, + which asks R to calculate the upper area instead of the lower. + +## Quantiles + +You've heard of sample quantiles. If you were the 95th percentile on an exam, you know +that 95% of people scored worse than you and 5% scored better. +These are sample quantities. But you might have wondered, what are my sample +quantiles estimating? In fact, they are estimating the population quantiles. +Here we define these population analogs. + + +The {$$}\alpha^{th}{/$$} **quantile** of a distribution +with distribution function {$$}F{/$$} is the point {$$}x_\alpha{/$$} so that + +{$$} +F(x_\alpha) = \alpha +{/$$} + +So the 0.95 quantile of a distribution is the point so that 95% of the mass +of the density lies below it. Or, in other words, the point so that the +probability of getting a randomly sampled point below it is 0.95. This is +analogous to the sample +quantiles where the 0.95 sample quantile is the value so that 95% of the data +lies below it. + + A **percentile** is simply a quantile with {$$}\alpha{/$$} expressed as a percent + rather than a proportion. The (population) +**median** is the {$$}50^{th}{/$$} percentile. Remember that percentiles +are not probabilities! Remember that quantiles have units. So the population +median height is the height (in inches say) so that the probability that a randomly selected +person from the population is shorter is 50%. The sample, or empirical, +median would be the height so in a sample so that 50% of the people in the +sample were shorter. + +### Example +What is the median of the distribution that we were working with before? +We want to solve {$$}0.5 = F(x) = x^2{/$$}, resulting in the solution + +{line-numbers=off,lang=r} +~~~~~~ +> sqrt(0.5) + +[1] 0.7071 +~~~~~~ + +Therefore, 0.7071 of calls being answered on a random day is the median. +Or, the probability that 70% or fewer calls get answered is 50%. + +R can approximate quantiles for you for common distributions with the +prefix `q` in front of the distribution name + +{line-numbers=off,lang=r} +~~~~~~ +> qbeta(0.5, 2, 1) + +[1] 0.7071 +~~~~~~ + +## Exercises + +1. Can you add the probabilities of any two events to get the probability of at least one occurring? +2. I define a PMF, {$$}p{/$$} so that for {$$}x = 0{/$$} and {$$}x=1{/$$} we have +{$$}p(0) = -0.1{/$$} and {$$}p(1) = 1.1{/$$}. Is this a valid PMF? +3. What is the probability that 75% or fewer calls get answered in a randomly sampled day from the population distribution from this chapter? +4. The 97.5th percentile of a distribution is? +5. Consider influenza epidemics for two parent heterosexual families. Suppose that the probability is 15% that at least one of the parents has contracted the disease. The probability that the father has contracted influenza is 10% while that the mother contracted the disease is 9%. What is the probability that both contracted influenza expressed as a whole number percentage? +[Watch a video solution to this problem.](http://youtu.be/CvnmoCuIN08?list=PLpl-gQkQivXhHOcVeU3bSJg78zaDYbP9L) and +[see a written out solution.](http://bcaffo.github.io/courses/06_StatisticalInference/homework/hw1.html#3) +6. A random variable, {$$}X{/$$}, is uniform, a box from 0 to 1 of height 1. (So that it's density is {$$}f(x) = 1{/$$} for {$$}0\leq x \leq 1{/$$}.) +What is it's median expressed to two decimal places? [Watch a video solution to this problem +here](http://youtu.be/UXcarD-1xAM?list=PLpl-gQkQivXhHOcVeU3bSJg78zaDYbP9L) and [see written solutions here](http://bcaffo.github.io/courses/06_StatisticalInference/homework/hw1.html#4). +7. If a continuous density that never touches the horizontal axis is symmetric about zero, can we say that its associated median is zero? [Watch a worked out solution to this problem here](http://youtu.be/sn48CGH_TXI?list=PLpl-gQkQivXhHOcVeU3bSJg78zaDYbP9L) and [see the question and a typed up answer here](http://bcaffo.github.io/courses/06_StatisticalInference/homework/hw1.html#9) diff --git a/manuscript/03_conditional.md b/manuscript/03_conditional.md index 967c881..3ef15ef 100644 --- a/manuscript/03_conditional.md +++ b/manuscript/03_conditional.md @@ -132,12 +132,12 @@ the specificity, {$$}P(- ~|~ D^c) =.985{/$$} and the prevalence {$$}P(D) = .001{/$$}. {$$} -\begin{eqnarray*} +\begin{aligned} P(D ~|~ +) & = &\frac{P(+~|~D)P(D)}{P(+~|~D)P(D) + P(+~|~D^c)P(D^c)}\\ & = & \frac{P(+~|~D)P(D)}{P(+~|~D)P(D) + \{1-P(-~|~D^c)\}\{1 - P(D)\}} \\ & = & \frac{.997\times .001}{.997 \times .001 + .015 \times .999}\\ & = & .062 -\end{eqnarray*} +\end{aligned} {/$$} In this population a positive test result only suggests a 6% probability that diff --git a/manuscript/LittleInferenceBook.Rmd b/manuscript/LittleInferenceBook.Rmd index 92b223f..a3474f4 100644 --- a/manuscript/LittleInferenceBook.Rmd +++ b/manuscript/LittleInferenceBook.Rmd @@ -1,10 +1,13 @@ --- output: + pdf_document: + toc: yes + toc_depth: '2' html_document: - toc: true - toc_depth: 2 includes: before_body: frontmatter.html + toc: yes + toc_depth: 2 --- diff --git a/manuscript/LittleInferenceBook.html b/manuscript/LittleInferenceBook.html new file mode 100644 index 0000000..37bdaee --- /dev/null +++ b/manuscript/LittleInferenceBook.html @@ -0,0 +1,2647 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
+ + + + + + + + + + +

Go to table of contents

+

+
+

Statistical inference for data science

+
+

A companion to the Coursera Statistical Inference Course

+
+

Brian Caffo

+

This book is for sale at http://leanpub.com/LittleInferenceBook

+

This book was produced from the source found in Brian Caffo's repository or a derivative.

+

+

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

+

+

To Kerri, Penelope and Scarlett

+

+
+
+
+
+

Contents

+
+ + + +
+ +
+ +
+

About this book

+

This book is written as a companion book to the Statistical Inference Coursera class as part of the Data Science Specialization. However, if you do not take the class, the book mostly stands on its own. A useful component of the book is a series of YouTube videos that comprise the Coursera class.

+

The book is intended to be a low cost introduction to the important field of statistical inference. The intended audience are students who are numerically and computationally literate, who would like to put those skills to use in Data Science or Statistics. The book is offered for free as a series of markdown documents on github and in more convenient forms (epub, mobi) on LeanPub and retail outlets.

+

This book is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License, which requires author attribution for derivative works, non-commercial use of derivative works and that changes are shared in the same way as the original work.

+
+
+

About the picture on the cover

+

The picture on the cover is a public domain image taken from Wikipedia’s article on Francis Galton’s quincunx. Francis Galton was an 19th century polymath who invented many of key concepts of statistics. The quincunx was an ingenious invention for illustrating the central limit theorem using a pinball setup.

+
+
+

Introduction

+
+

Before beginning

+

This book is designed as a companion to the Statistical Inference Coursera class as part of the Data Science Specialization, a ten course program offered by three faculty, Jeff Leek, Roger Peng and Brian Caffo, at the Johns Hopkins University Department of Biostatistics.

+

The videos associated with this book can be watched in full here, though the relevant links to specific videos are placed at the appropriate locations throughout.

+

Before beginning, we assume that you have a working knowledge of the R programming language. If not, there is a wonderful Coursera class by Roger Peng, that can be found here.

+

The entirety of the book is on GitHub here. Please submit pull requests if you find errata! In addition the course notes can be found also on GitHub here. While most code is in the book, all of the code for every figure and analysis in the book is in the R markdown files files (.Rmd) for the respective lectures.

+

Finally, we should mention swirl (statistics with interactive R programming). swirl is an intelligent tutoring system developed by Nick Carchedi, with contributions by Sean Kross and Bill and Gina Croft. It offers a way to learn R in R. Download swirl here. There’s a swirl module for this course!. Try it out, it’s probably the most effective way to learn.

+
+
+

Statistical inference defined

+

Watch this video before beginning.

+

We’ll define statistical inference as the process of generating conclusions about a population from a noisy sample. Without statistical inference we’re simply living within our data. With statistical inference, we’re trying to generate new knowledge.

+

Knowledge and parsimony, (using simplest reasonable models to explain complex phenomena), go hand in hand. Probability models will serve as our parsimonious description of the world. The use of probability models as the connection between our data and a populations represents the most effective way to obtain inference.

+
+

Motivating example: who’s going to win the election?

+

In every major election, pollsters would like to know, ahead of the actual election, who’s going to win. Here, the target of estimation (the estimand) is clear, the percentage of people in a particular group (city, state, county, country or other electoral grouping) who will vote for each candidate.

+

We can not poll everyone. Even if we could, some polled may change their vote by the time the election occurs. How do we collect a reasonable subset of data and quantify the uncertainty in the process to produce a good guess at who will win?

+
+
+

Motivating example, predicting the weather

+

When a weatherman tells you the probability that it will rain tomorrow is 70%, they’re trying to use historical data to predict tomorrow’s weather - and to actually attach a probability to it. That probability refers to population.

+
+
+

Motivating example, brain activation

+

An example that’s very close to the research I do is trying to predict what areas of the brain activate when a person is put in the fMRI scanner. In that case, people are doing a task while in the scanner. For example, they might be tapping their finger. We’d like to compare when they are tapping their finger to when they are not tapping their finger and try to figure out what areas of the brain are associated with the finger tapping.

+
+
+
+

Summary notes

+

These examples illustrate many of the difficulties of trying to use data to create general conclusions about a population.

+

Paramount among our concerns are:

+
    +
  • Is the sample representative of the population that we’d like to draw inferences about?
    • +
  • Are there known and observed, known and unobserved or unknown and unobserved variables that contaminate our conclusions?
    • +
  • Is there systematic bias created by missing data or the design or conduct of the study?
    • +
  • What randomness exists in the data and how do we use or adjust for it? Here randomness can either be explicit via randomization or random sampling, or implicit as the aggregation of many complex unknown processes.
    • +
  • Are we trying to estimate an underlying mechanistic model of phenomena under study?
    • +
+

Statistical inference requires navigating the set of assumptions and tools and subsequently thinking about how to draw conclusions from data.

+
+
+

The goals of inference

+

You should recognize the goals of inference. Here we list five examples of inferential goals.

+
    +
  1. Estimate and quantify the uncertainty of an estimate of a population quantity (the proportion of people who will vote for a candidate).
    2. +
  2. Determine whether a population quantity is a benchmark value (“is the treatment effective?”).
    3. +
  3. Infer a mechanistic relationship when quantities are measured with noise (“What is the slope for Hooke’s law?”)
    4. +
  4. Determine the impact of a policy? (“If we reduce pollution levels, will asthma rates decline?”)
    5. +
  5. Talk about the probability that something occurs.
    6. +
+
+
+

The tools of the trade

+

Several tools are key to the use of statistical inference. We’ll only be able to cover a few in this class, but you should recognize them anyway.

+
    +
  1. Randomization: concerned with balancing unobserved variables that may confound inferences of interest.
    2. +
  2. Random sampling: concerned with obtaining data that is representative of the population of interest.
    3. +
  3. Sampling models: concerned with creating a model for the sampling process, the most common is so called “iid”.
    4. +
  4. Hypothesis testing: concerned with decision making in the presence of uncertainty.
    5. +
  5. Confidence intervals: concerned with quantifying uncertainty in estimation.
    6. +
  6. Probability models: a formal connection between the data and a population of interest. Often probability models are assumed or are approximated.
    7. +
  7. Study design: the process of designing an experiment to minimize biases and variability.
    8. +
  8. Nonparametric bootstrapping: the process of using the data to, with minimal probability model assumptions, create inferences.
    9. +
  9. Permutation, randomization and exchangeability testing: the process of using data permutations to perform inferences.
    10. +
+
+
+

Different thinking about probability leads to different styles of inference

+

We won’t spend too much time talking about this, but there are several different styles of inference. Two broad categories that get discussed a lot are:

+
    +
  1. Frequency probability: is the long run proportion of times an event occurs in independent, identically distributed repetitions.
    2. +
  2. Frequency style inference: uses frequency interpretations of probabilities to control error rates. Answers questions like “What should I decide given my data controlling the long run proportion of mistakes I make at a tolerable level.”
    3. +
  3. Bayesian probability: is the probability calculus of beliefs, given that beliefs follow certain rules.
    4. +
  4. Bayesian style inference: the use of Bayesian probability representation of beliefs to perform inference. Answers questions like “Given my subjective beliefs and the objective information from the data, what should I believe now?”
    5. +
+

Data scientists tend to fall within shades of gray of these and various other schools of inference. Furthermore, there are so many shades of gray between the styles of inferences that it is hard to pin down most modern statisticians as either Bayesian or frequentist. In this class, we will primarily focus on basic sampling models, basic probability models and frequency style analyses to create standard inferences. This is the most popular style of inference by far.

+

Being data scientists, we will also consider some inferential strategies that
+rely heavily on the observed data, such as permutation testing and bootstrapping. As probability modeling will be our starting point, we first build up basic probability as our first task.

+
+
+

Exercises

+
    +
  1. The goal of statistical inference is to?
    2. +
+
    +
  • Infer facts about a population from a sample.
    • +
  • Infer facts about the sample from a population.
    • +
  • Calculate sample quantities to understand your data.
    • +
  • To torture Data Science students.
    • +
+
    +
  1. The goal of randomization of a treatment in a randomized trial is to?
    2. +
+
    +
  • It doesn’t really do anything.
    • +
  • To obtain a representative sample of subjects from the population of interest.
    • +
  • Balance unobserved covariates that may contaminate the comparison between the treated and control groups.
    • +
  • To add variation to our conclusions.
    • +
+
    +
  1. Probability is a?
    2. +
+
    +
  • Population quantity that we can potentially estimate from data.
    • +
  • A data quantity that does not require the idea of a population.
    • +
+
+
+
+

Probability

+

Watch this video before beginning.

+

Probability forms the foundation for almost all treatments of statistical inference. In our treatment, probability is a law that assigns numbers to the long run occurrence of random phenomena after repeated unrelated realizations.

+

Before we begin discussing probability, let’s dispense with some deep philosophical questions, such as “What is randomness?” and “What is the fundamental interpretation of probability?”. One could spend a lifetime studying these questions (and some have). For our purposes, randomness is any process occurring without apparent deterministic patterns. Thus we will treat many things as if they were random when, in fact they are completely deterministic. In my field, biostatistics, we often model disease outcomes as if they were random when they are the result of many mechanistic components whose aggregate behavior appears random. Probability for us will be the long run proportion of times some occurs in repeated unrelated realizations. So, think of the proportion of times that you get a head when flipping a coin.

+

For the interested student, I would recommend the books and work by Ian Hacking to learn more about these deep philosophical issues. For us data scientists, the above definitions will work fine.

+
+

Where to get a more thorough treatment of probability

+

In this lecture, we will cover the fundamentals of probability at low enough of a level to have a basic understanding for the rest of the series. For a more complete treatment see the class Mathematical Biostatistics Boot Camp 1, which can be viewed on YouTube here. In addition, there’s the actual Coursera course that I run periodically (this is the first Coursera class that I ever taught). Also there are a set of notes on GitHub. Finally, there’s a follow up class, uninspiringly named Mathematical Biostatistics Boot Camp 2, that is more devoted to biostatistical topics that has an associated YouTube playlist, Coursera Class and GitHub notes.

+
+
+

Kolmogorov’s Three Rules

+

Watch this lecture before beginning.

+

Given a random experiment (say rolling a die) a probability measure is a population quantity that summarizes the randomness. The brilliant discovery of the father of probability, the Russian mathematician Kolmogorov, was that to satisfy our intuition about how probability should behave, only three rules were needed.

+

Consider an experiment with a random outcome. Probability takes a possible outcome from an experiment and:

+
    +
  1. assigns it a number between 0 and 1
    2. +
  2. requires that the probability that something occurs is 1
    3. +
  3. required that the probability of the union of any two sets of outcomes that have nothing in common (mutually exclusive) is the sum of their respective probabilities.
    4. +
+

From these simple rules all of the familiar rules of probability can be developed. This all might seem a little odd at first and so we’ll build up our intuition with some simple examples based on coin flipping and die rolling.

+

I would like to reiterate the important definition that we wrote out: mutually exclusive. Two events are mutually exclusive if they cannot both simultaneously occur. For example, we cannot simultaneously get a 1 and a 2 on a die. Rule 3 says that since the event of getting a 1 and 2 on a die are mutually exclusive, the probability of getting at least one (the union) is the sum of their probabilities. So if we know that the probability of getting a 1 is 1/6 and the probability of getting a 2 is 1/6, then the probability of getting a 1 or a 2 is 2/6, the sum of the two probabilities since they are mutually exclusive.

+
+
+

Consequences of The Three Rules

+

Let’s cover some consequences of our three simple rules. Take, for example, the probability that something occurs is 1 minus the probability of the opposite occurring. Let \(A\) be the event that we get a 1 or a 2 on a rolled die. Then \(A^c\) is the opposite, getting a 3, 4, 5 or 6. Since \(A\) and \(A^c\) cannot both simultaneously occur, they are mutually exclusive. So the probability that either \(A\) or \(A^c\) is \(P(A) + P(A^c)\). Notice, that the probability that either occurs is the probability of getting a 1, 2, 3, 4, 5 or 6, or in other words, the probability that something occurs, which is 1 by rule number 2. So we have that \(1 = P(A) + P(A^c)\) or that \(P(A) = 1 - P(A^c)\).

+

We won’t go through this tedious exercise (since Kolmogorov already did it for us). Instead here’s a list of some of the consequences of Kolmogorov’s rules that are often useful.

+
    +
  1. The probability that nothing occurs is 0
    2. +
  2. The probability that something occurs is 1
    3. +
  3. The probability of something is 1 minus the probability that the opposite occurs
    4. +
  4. The probability of at least one of two (or more) things that can not simultaneously occur (mutually exclusive) is the sum of their respective probabilities
    5. +
  5. For any two events the probability that at least one occurs is the sum of their probabilities minus their intersection.
    6. +
+

This last rules states that \(P(A \cup B) = P(A) + P(B) - P(A \cap B)\) shows what is the issue with adding probabilities that are not mutually exclusive. If we do this, we’ve added the probability that both occur in twice! (Watch the video where I draw a Venn diagram to illustrate this).

+
+

Example of Implementing Probability Calculus

+

The National Sleep Foundation (www.sleepfoundation.org) reports that around 3% of the American population has sleep apnea. They also report that around 10% of the North American and European population has restless leg syndrome. Does this imply that 13% of people will have at least one sleep problems of these sorts? In other words, can we simply add these two probabilities?

+

Answer: No, the events can simultaneously occur and so are not mutually exclusive. To elaborate let:

+

\[ + \begin{aligned} + A_1 &= \{\mbox{Person has sleep apnea}\} \\ +A_2 &= \{\mbox{Person has RLS}\} + \end{aligned} +\]

+

Then

+

\[ + \begin{aligned} + P(A_1 \cup A_2 ) &= P(A_1) + P(A_2) - P(A_1 \cap +A_2) \\ + &= 0.13 - \mbox{Probability of having both} + \end{aligned} +\]

+

Given the scenario, it’s likely that some fraction of the population has both. This example serves as a reminder don’t add probabilities unless the events are mutually exclusive. We’ll have a similar rule for multiplying probabilities and independence.

+
+
+
+

Random variables

+

Watch this video before reading this section

+

Probability calculus is useful for understanding the rules that probabilities must follow. However, we need ways to model and think about probabilities for numeric outcomes of experiments (broadly defined). Densities and mass functions for random variables are the best starting point for this. You’ve already heard of a density since you’ve heard of the famous “bell curve”, or Gaussian density. In this section you’ll learn exactly what the bell curve is and how to work with it.

+

Remember, everything we’re talking about up to at this point is a population quantity, not a statement about what occurs in our data. Think about the fact that 50% probability for head is a statement about the coin and how we’re flipping it, not a statement about the percentage of heads we obtained in a particular set of flips. This is an important distinction that we will emphasize over and over in this course. Statistical inference is about describing populations using data. Probability density functions are a way to mathematically characterize the population. In this course, we’ll assume that our sample is a random draw from the population.

+

So our definition is that a random variable is a numerical outcome of an experiment. The random variables that we study will come in two varieties, discrete or continuous. Discrete random variables are random variables that take on only a countable number of possibilities. Mass functions will assign probabilities that they take specific values. Continuous random variable can conceptually take any value on the real line or some subset of the real line and we talk about the probability that they lie within some range. Densities will characterize these probabilities.

+

Let’s consider some examples of measurements that could be considered random variables. First, familiar gambling experiments like the tossing of a coin and the rolling of a die produce random variables. For the coin, we typically code a tail as a 0 and a head as a 1. (For the die, the number facing up would be the random variable.) We will use these examples a lot to help us build intuition. However, they aren’t interesting in the sense of seeming very contrived. Nonetheless, the coin example is particularly useful since many of the experiments we consider will be modeled as if tossing a biased coin. Modeling any binary characteristic from a random sample of a population can be thought of as a coin toss, with the random sampling performing the roll of the toss and the population percentage of individuals with the characteristic is the probability of a head. Consider, for example, logging whether or not subjects were hypertensive in a random sample. Each subject’s outcome can be modeled as a coin toss. In a similar sense the die roll serves as our model for phenomena with more than one level, such as hair color or rating scales.

+

Consider also the random variable of the number of web hits for a site each day. This variable is a count, but is largely unbounded (or at least we couldn’t put a specific reasonable upper limit). Random variables like this are often modeled with the so called Poisson distribution.

+

Finally, consider some continuous random variables. Think of things like lengths or weights. It is mathematically convenient to model these as if they were continuous (even if measurements were truncated liberally). In fact, even discrete random variables with lots of levels are often treated as continuous for convenience.

+

For all of these kinds of random variables, we need convenient mathematical functions to model the probabilities of collections of realizations. These functions, called mass functions and densities, take possible values of the random variables, and assign the associated probabilities. These entities describe the population of interest. So, consider the most famous density, the normal distribution. Saying that body mass indices follow a normal distribution is a statement about the population of interest. The goal is to use our data to figure out things about that normal distribution, where it’s centered, how spread out it is and even whether our assumption of normality is warranted!

+
+
+

Probability mass functions

+

A probability mass function evaluated at a value corresponds to the probability that a random variable takes that value. To be a valid pmf a function, \(p\), must satisfy:

+
    +
  1. It must always be larger than or equal to 0.
    2. +
  2. The sum of the possible values that the random variable can take has to add up to one.
    3. +
+
+

Example

+

Let \(X\) be the result of a coin flip where \(X=0\) represents tails and \(X = 1\) represents heads. \(p(x) = (1/2)^{x} (1/2)^{1-x}\) for \(x = 0,1\). Suppose that we do not know whether or not the coin is fair; Let \(\theta\) be the probability of a head expressed as a proportion (between 0 and 1). \(p(x) = \theta^{x} (1 - \theta)^{1-x}\) for \(x = 0,1\)

+
+
+
+

Probability density functions

+

Watch this video before beginning.

+

A probability density function (pdf), is a function associated with a continuous random variable. Because of the peculiarities of treating measurements as having been recorded to infinite decimal expansions, we need a different set of rules. This leads us to the central dogma of probability density functions:

+

Areas under PDFs correspond to probabilities for that random variable

+

Therefore, when one says that intelligence quotients (IQ) in population follows a bell curve, they are saying that the probability of a randomly selected person from this population having an IQ between two values is given by the area under the bell curve.

+

Not every function can be a valid probability density function. For example, if the function dips below zero, then we could have negative probabilities. If the function contains too much area underneath it, we could have probabilities larger than one. The following two rules tell us when a function is a valid probability density function.

+

Specifically, to be a valid pdf, a function must satisfy

+
    +
  1. It must be larger than or equal to zero everywhere.
    2. +
  2. The total area under it must be one.
    3. +
+
+

Example

+

Suppose that the proportion of help calls that get addressed in a random day by a help line is given by \(f(x) = 2 x\) for \(0< x < 1\). The R code for plotting this density is

+

Code for plotting the density

+
x <- c(-0.5, 0, 1, 1, 1.5)
+y <- c(0, 0, 2, 0, 0)
+plot(x, y, lwd = 3,frame = FALSE, type = "l")
+
+

The result of the code is given below.

+
+Help call density +

Help call density

+
+

Is this a mathematically valid density? To answer this we need to make sure it satisfies our two conditions. First it’s clearly nonnegative (it’s at or above the horizontal axis everywhere). The area is similarly easy. Being a right triangle in the only section of the density that is above zero, we can calculate it as 1/2 the area of the base times the height. This is \(\frac{1}{2} \times 1 \times 2 = 1\)

+

Now consider answering the following question. What is the probability that 75% or fewer of calls get addressed? Remember, for continuous random variables, probabilities are represented by areas underneath the density function. So, we want the area from 0.75 and below, as illustrated by the figure below.

+
+Help call density +

Help call density

+
+

This again is a right triangle, with length of the base as 0.75 and height 1.5. The R code below shows the calculation.

+
+
> 1.5 * 0.75/2
+
+[1] 0.5625
+
+

Thus, the probability of 75% or fewer calls getting addressed in a random day for this help line is 56%. We’ll do this a lot throughout this class and work with more useful densities. It should be noted that this specific density is a special case of the so called beta density. Below I show how to use R’s built in evaluation function for the beta density to get the probability.

+
+
> pbeta(0.75, 2, 1)
+
+[1] 0.5625
+
+

Notice the syntax pbeta. In R, a prefix of p returns probabilities, d returns the density, q returns the quantile and r returns generated random variables. (You’ll learn what each of these does in subsequent sections.)

+
+
+
+

CDF and survival function

+

Certain areas of PDFs and PMFs are so useful, we give them names. The cumulative distribution function (CDF) of a random variable, \(X\), returns the probability that the random variable is less than or equal to the value \(x\). Notice the (slightly annoying) convention that we use an upper case \(X\) to denote a random, unrealized, version of the random variable and a lowercase \(x\) to denote a specific number that we plug into. (This notation, as odd as it may seem, dates back to Fisher and isn’t going anywhere, so you might as well get used to it. Uppercase for unrealized random variables and lowercase as placeholders for numbers to plug into.) So we could write the following to describe the distribution function \(F\):

+

\[ +F(x) = P(X \leq x) +\]

+

This definition applies regardless of
+whether the random variable is discrete or continuous. The survival function of a random variable \(X\) is defined as the probability that the random variable is greater than the value \(x\).

+

\[ +S(x) = P(X > x) +\]

+

Notice that \(S(x) = 1 - F(x)\), since the survival function evaluated at a particular value of \(x\) is calculating the probability of the opposite event (greater than as opposed to less than or equal to). The survival function is often preferred in biostatistical applications while the distribution function is more generally used (though both convey the same information.)

+
+

Example

+

What are the survival function and CDF from the density considered before?

+

\[ +F(x) = P(X \leq x) = \frac{1}{2} Base \times Height = \frac{1}{2} (x) \times (2 x) = x^2, +\]

+

for \(1 \geq x \geq 0\). Notice that calculating the survival function is now trivial given that we’ve already calculated the distribution function.

+

\[ + S(x) = 1 = F(x) = 1 - x^2 +\]

+

Again, R has a function that calculates the distribution function for us in this case, pbeta. Let’s try calculating \(F(.4)\), \(F(.5)\) and \(F(.6)\)

+
+
> pbeta(c(0.4, 0.5, 0.6), 2, 1)
+
+[1] 0.16 0.25 0.36
+
+

Notice, of course, these are simply the numbers squared. By default the prefix p in front of a density in R gives the distribution function (pbeta, pnorm, pgamma). If you want the survival function values, you could always subtract by one, or give the argument lower.tail = FALSE as an argument to the function, which asks R to calculate the upper area instead of the lower.

+
+
+
+

Quantiles

+

You’ve heard of sample quantiles. If you were the 95th percentile on an exam, you know that 95% of people scored worse than you and 5% scored better. These are sample quantities. But you might have wondered, what are my sample quantiles estimating? In fact, they are estimating the population quantiles. Here we define these population analogs.

+

The \(\alpha^{th}\) quantile of a distribution with distribution function \(F\) is the point \(x_\alpha\) so that

+

\[ +F(x_\alpha) = \alpha +\]

+

So the 0.95 quantile of a distribution is the point so that 95% of the mass of the density lies below it. Or, in other words, the point so that the probability of getting a randomly sampled point below it is 0.95. This is analogous to the sample quantiles where the 0.95 sample quantile is the value so that 95% of the data lies below it.

+

A percentile is simply a quantile with \(\alpha\) expressed as a percent rather than a proportion. The (population) median is the \(50^{th}\) percentile. Remember that percentiles are not probabilities! Remember that quantiles have units. So the population median height is the height (in inches say) so that the probability that a randomly selected person from the population is shorter is 50%. The sample, or empirical, median would be the height so in a sample so that 50% of the people in the sample were shorter.

+
+

Example

+

What is the median of the distribution that we were working with before? We want to solve \(0.5 = F(x) = x^2\), resulting in the solution

+
+
> sqrt(0.5)
+
+[1] 0.7071
+
+

Therefore, 0.7071 of calls being answered on a random day is the median. Or, the probability that 70% or fewer calls get answered is 50%.

+

R can approximate quantiles for you for common distributions with the prefix q in front of the distribution name

+
+
> qbeta(0.5, 2, 1)
+
+[1] 0.7071
+
+
+
+
+

Exercises

+
    +
  1. Can you add the probabilities of any two events to get the probability of at least one occurring?
    2. +
  2. I define a PMF, \(p\) so that for \(x = 0\) and \(x=1\) we have \(p(0) = -0.1\) and \(p(1) = 1.1\). Is this a valid PMF?
    3. +
  3. What is the probability that 75% or fewer calls get answered in a randomly sampled day from the population distribution from this chapter?
    4. +
  4. The 97.5th percentile of a distribution is?
    5. +
  5. Consider influenza epidemics for two parent heterosexual families. Suppose that the probability is 15% that at least one of the parents has contracted the disease. The probability that the father has contracted influenza is 10% while that the mother contracted the disease is 9%. What is the probability that both contracted influenza expressed as a whole number percentage? Watch a video solution to this problem. and see a written out solution.
    6. +
  6. A random variable, \(X\), is uniform, a box from 0 to 1 of height 1. (So that it’s density is \(f(x) = 1\) for \(0\leq x \leq 1\).) What is it’s median expressed to two decimal places? Watch a video solution to this problem here and see written solutions here.
    7. +
  7. If a continuous density that never touches the horizontal axis is symmetric about zero, can we say that its associated median is zero? Watch a worked out solution to this problem here and see the question and a typed up answer here
    8. +
+
+
+
+

Conditional probability

+
+

Conditional probability, motivation

+

Watch this video before beginning.

+

Conditioning is a central subject in statistics. If we are given information about a random variable, it changes the probabilities associated with it. For example, the probability of getting a one when rolling a (standard) die is usually assumed to be one sixth. If you were given the extra information that the die roll was an odd number (hence 1, 3 or 5) then conditional on this new information, the probability of a one is now one third.

+

This is the idea of conditioning, taking away the randomness that we know to have occurred. Consider another example, such as the result of a diagnostic imaging test for lung cancer. What’s the probability that a person has cancer given a positive test? How does that probability change under the knowledge that a patient has been a lifetime heavy smoker and both of their parents had lung cancer? Conditional on this new information, the probability has increased dramatically.

+
+
+

Conditional probability, definition

+

We can formalize the definition of conditional probability so that the mathematics matches our intuition.

+

Let \(B\) be an event so that \(P(B) > 0\). Then the conditional probability of an event \(A\) given that \(B\) has occurred is:

+

\[ +P(A ~|~ B) = \frac{P(A \cap B)}{P(B)}. +\]

+

If \(A\) and \(B\) are unrelated in any way, or in other words independent, (discussed more later in the lecture), then

+

\[ +P(A ~|~ B) = \frac{P(A) P(B)}{P(B)} = P(A) +\]

+

That is, if the occurrence of \(B\) offers no information about the occurrence of \(A\) - the probability conditional on the information is the same as the probability without the information, we say that the two events are independent.

+
+

Example

+

Consider our die roll example again. Here we have that \(B = \{1, 3, 5\}\) and \(A = \{1\}\)

+

\[ +P(\mbox{one given that roll is odd}) = P(A ~|~ B) += \frac{P(A \cap B)}{P(B)} += \frac{P(A)}{P(B)} += \frac{1/6}{3/6} = \frac{1}{3} +\]

+

Which exactly mirrors our intuition.

+
+
+
+

Bayes’ rule

+

Watch this video before beginning

+

Bayes’ rule is a famous result in statistics and probability. It forms the foundation for large branches of statistical thinking. Bayes’ rule allows us to reverse the conditioning set provided that we know some marginal probabilities.

+

Why is this useful? Consider our lung cancer example again. It would be relatively easy for physicians to calculate the probability that the diagnostic method is positive for people with lung cancer and negative for people without. They could take several people who are already known to have the disease and apply the test and conversely take people known not to have the disease. However, for the collection of people with a positive test result, the reverse probability is more of interest, “given a positive test what is the probability of having the disease?”, and “given a given a negative test what is the probability of not having the disease?”.

+

Bayes’ rule allows us to switch the conditioning event, provided a little bit of extra information. Formally Bayes’ rule is:

+

\[ +P(B ~|~ A) = \frac{P(A ~|~ B) P(B)}{P(A ~|~ B) P(B) + P(A ~|~ B^c)P(B^c)}. +\]

+
+

Diagnostic tests

+

Since diagnostic tests are a really good example of Bayes’ rule in practice, let’s go over them in greater detail. (In addition, understanding Bayes’ rule will be helpful for your own ability to understand medical tests that you see in your daily life). We require a few definitions first.

+

Let \(+\) and \(-\) be the events that the result of a diagnostic test is positive or negative respectively Let \(D\) and \(D^c\) be the event that the subject of the test has or does not have the disease respectively

+

The sensitivity is the probability that the test is positive given that the subject actually has the disease, \(P(+ ~|~ D)\)

+

The specificity is the probability that the test is negative given that the subject does not have the disease, \(P(- ~|~ D^c)\)

+

So, conceptually at least, the sensitivity and specificity are straightforward to estimate. Take people known to have and not have the disease and apply the diagnostic test to them. However, the reality of estimating these quantities is quite challenging. For example, are the people known to have the disease in its later stages, while the diagnostic will be used on people in the early stages where it’s harder to detect? Let’s put these subtleties to the side and assume that they are known well.

+

The quantities that we’d like to know are the predictive values.

+

The positive predictive value is the probability that the subject has the disease given that the test is positive, \(P(D ~|~ +)\)

+

The negative predictive value is the probability that the subject does not have the disease given that the test is negative, \(P(D^c ~|~ -)\)

+

Finally, we need one last thing, the prevalence of the disease - which is the marginal probability of disease, \(P(D)\). Let’s now try to figure out a PPV in a specific setting.

+
+
+

Example

+

A study comparing the efficacy of HIV tests, reports on an experiment which concluded that HIV antibody tests have a sensitivity of 99.7% and a specificity of 98.5% Suppose that a subject, from a population with a .1% prevalence of HIV, receives a positive test result. What is the positive predictive value?

+

Mathematically, we want \(P(D ~|~ +)\) given the sensitivity, \(P(+ ~|~ D) = .997\), the specificity, \(P(- ~|~ D^c) =.985\) and the prevalence \(P(D) = .001\).

+

\[ +\begin{aligned} +P(D ~|~ +) & = &\frac{P(+~|~D)P(D)}{P(+~|~D)P(D) + P(+~|~D^c)P(D^c)}\\ + & = & \frac{P(+~|~D)P(D)}{P(+~|~D)P(D) + \{1-P(-~|~D^c)\}\{1 - P(D)\}} \\ + & = & \frac{.997\times .001}{.997 \times .001 + .015 \times .999}\\ + & = & .062 +\end{aligned} +\]

+

In this population a positive test result only suggests a 6% probability that the subject has the disease, (the positive predictive value is 6% for this test). If you were wondering how it could be so low for this test, the low positive predictive value is due to low prevalence of disease and the somewhat modest specificity

+

Suppose it was known that the subject was an intravenous drug user and routinely had intercourse with an HIV infected partner? Our prevalence would change dramatically, thus increasing the PPV. You might wonder if there’s a way to summarize the evidence without appealing to an often unknowable prevalence? Diagnostic likelihood ratios provide this for us.

+
+
+
+

Diagnostic Likelihood Ratios

+

The diagnostic likelihood ratios summarize the evidence of disease given a positive or negative test. They are defined as:

+

The diagnostic likelihood ratio of a positive test, labeled \(DLR_+\), is \(P(+ ~|~ D) / P(+ ~|~ D^c)\), which is the \(sensitivity / (1 - specificity)\).

+

The diagnostic likelihood ratio of a negative test, labeled \(DLR_-\), is \(P(- ~|~ D) / P(- ~|~ D^c)\), which is the \((1 - sensitivity) / specificity\).

+

How do we interpret the DLRs? This is easiest when looking at so called odds ratios. Remember that if \(p\) is a probability, then \(p / (1 - p)\) is the odds. Consider now the odds in our setting:

+

Using Bayes rule, we have

+

\[ +P(D ~|~ +) = \frac{P(+~|~D)P(D)}{P(+~|~D)P(D) + P(+~|~D^c)P(D^c)} +\]

+

and

+

\[ +P(D^c ~|~ +) = \frac{P(+~|~D^c)P(D^c)}{P(+~|~D)P(D) + P(+~|~D^c)P(D^c)}. +\]

+

Therefore, dividing these two equations we have:

+

\[ +\frac{P(D ~|~ +)}{P(D^c ~|~ +)} = \frac{P(+~|~D)}{P(+~|~D^c)}\times \frac{P(D)}{P(D^c)} +\]

+

In other words, the post test odds of disease is the pretest odds of disease times the \(DLR_+\). Similarly, \(DLR_-\) relates the decrease in the odds of the disease after a negative test result to the odds of disease prior to the test.

+

So, the DLRs are the factors by which you multiply your pretest odds to get your post test odds. Thus, if a test has a \(DLR_+\) of 6, regardless of the prevalence of disease, the post test odds is six times that of the pretest odds.

+
+

HIV example revisited

+

Let’s reconsider our HIV antibody test again.
+Suppose a subject has a positive HIV test

+

\[DLR_+ = .997 / (1 - .985) = 66\]

+

The result of the positive test is that the odds of disease is now 66 times the pretest odds. Or, equivalently, the hypothesis of disease is 66 times more supported by the data than the hypothesis of no disease

+

Suppose instead that a subject has a negative test result

+

\[DLR_- = (1 - .997) / .985 =.003\]

+

Therefore, the post-test odds of disease is now 0.3% of the pretest odds given the negative test. Or, the hypothesis of disease is supported \(.003\) times that of the hypothesis of absence of disease given the negative test result

+
+
+
+

Independence

+

Watch this video before beginning.

+

Statistical independence of events is the idea that the events are unrelated. Consider successive coin flips. Knowledge of the result of the first coin flip tells us nothing about the second. We can formalize this into a definition.

+

Two events \(A\) and \(B\) are independent if

+

\[P(A \cap B) = P(A)P(B)\]

+

Equivalently if \(P(A ~|~ B) = P(A)\). Note that since \(A\) is independent of \(B\) we know that \(A^c\) is independent of \(B\) \(A\) is independent of \(B^c\) \(A^c\) is independent of \(B^c\).

+

While this definition works for sets, remember that random variables are really the things that we are interested in. Two random variables, \(X\) and \(Y\) are independent if for any two sets \(A\) and \(B\) \(P([X \in A] \cap [Y \in B]) = P(X\in A)P(Y\in B)\)

+

We will almost never work with these definitions. Instead, the important principle is that probabilities of independent things multiply! This has numerous consequences, including the idea that we shouldn’t multiply non-independent probabilities.

+
+

Example

+

Let’s cover a very simple example: “What is the probability of getting two consecutive heads?”. Then we have that \(A\) is the event of getting a head on flip 1 \(P(A) = 0.5\) \(B\) is the event of getting a head on flip 2 \(P(B) = 0.5\) \(A \cap B\) is the event of getting heads on flips 1 and 2. Then independence would tell us that:

+

\[P(A \cap B) = P(A)P(B) = 0.5 \times 0.5 = 0.25\]

+

This is exactly what we would have intuited of course. But, it’s nice that the mathematics mirrors our intuition. In more complex settings, it’s easy to get tripped up. Consider the following famous (among statisticians at least) case study.

+
+
+

Case Study

+

Volume 309 of Science reports on a physician who was on trial for expert testimony in a criminal trial. Based on an estimated prevalence of sudden infant death syndrome (SIDS) of 1 out of 8,543, a physician testified that that the probability of a mother having two children with SIDS was \((1 / 8,543)^2\). The mother on trial was convicted of murder.

+

Relevant to this discussion, the principal mistake was to assume that the events of having SIDs within a family are independent. That is, \(P(A_1 \cap A_2)\) is not necessarily equal to \(P(A_1)P(A_2)\). This is because biological processes that have a believed genetic or familiar environmental component, of course, tend to be dependent within families. Thus, we can’t just multiply the probabilities to obtain the result.

+

There are many other interesting aspects to the case. For example, the idea of a low probability of an event representing evidence against a plaintiff. (Could we convict all lottery winners of fixing the lotter since the chance that they would win is so small.)

+
+
+
+

IID random variables

+

Now that we’ve introduced random variables and independence, we can introduce a central modeling assumption made in statistics. Specifically the idea of a random sample. Random variables are said to be independent and identically distributed (iid) if they are independent and all are drawn from the same population. The reason iid samples are so important is that they are a model for random samples. This is a default starting point for most statistical inferences.

+

The idea of having a random sample is powerful for a variety of reasons. Consider that in some study designs, such as in election polling, great pains are made to make sure that the sample is randomly drawn from a population of interest. The idea is to expend a lot of effort on design to get robust inferences. In these settings assuming that the data is iid is both natural and warranted.

+

In other settings, the study design is far more opaque, and statistical inferences are conducted under the assumption that the data arose from a random sample, since it serves as a useful benchmark. Most studies in the fields of epidemiology and economics fall under this category. Take, for example, studying how policies impact countries gross domestic product by looking at countries before and after enacting the policies. The countries are not a random sample from the set of countries. Instead, conclusions must be made under the assumption that the countries are a random sample and the interpretation of the strength of the inferences adapted in kind.

+
+
+

Exercises

+
    +
  1. I pull a card from a deck and do not show you the result. I say that the resulting card is a heart. What is the probability that it is the queen of hearts?
    2. +
  2. The odds associated with a probability, \(p\), are defined as?
    3. +
  3. The probability of getting two sixes when rolling a pair of dice is?
    4. +
  4. The probability that a manuscript gets accepted to a journal is 12% (say). However, given that a revision is asked for, the probability that it gets accepted is 90%. Is it possible that the probability that a manuscript has a revision asked for is 20%? Watch a video of this problem getting solved and see the worked out solutions here.
    5. +
  5. Suppose 5% of housing projects have issues with asbestos. The sensitivity of a test for asbestos is 93% and the specificity is 88%. What is the probability that a housing project has no asbestos given a negative test expressed as a percentage to the nearest percentage point? Watch a video solution here and see the worked out problem here.
    6. +
+
+
+
+

Expected values

+

Watch this video before beginning.

+

Expected values characterize a distribution. The most useful expected value, the mean, characterizes the center of a density or mass function. Another expected value summary, the variance, characterizes how spread out a density is. Yet another expected value calculation is the skewness, which considers how much a density is pulled toward high or low values.

+

Remember, in this lecture we are discussing population quantities. It is convenient (and of course by design) that the names for all of the sample analogs estimate the associated population quantity. So, for example, the sample or empirical mean estimates the population mean; the sample variance estimates the population variance and the sample skewness estimates the population skewness.

+
+

The population mean for discrete random variables

+

The expected value or (population) mean of a random variable is the center of its distribution. For discrete random variable \(X\) with PMF \(p(x)\), it is defined as follows:

+

\[ +E[X] = \sum_x xp(x). +\]

+

where the sum is taken over the possible values of \(x\). Where did they get this idea from? It’s taken from the physical idea of the center of mass. Specifically, \(E[X]\) represents the center of mass of a collection of locations and weights, \(\{x, p(x)\}\). We can exploit this fact to quickly calculate population means for distributions where the center of mass is obvious.

+
+
+

The sample mean

+

It is important to contrast the population mean (the estimand) with the sample mean (the estimator). The sample mean estimates the population mean. Not coincidentally, since the population mean is the center of mass of the population distribution, the sample mean is the center of mass of the data. In fact, it’s exactly the same equation:

+

\[ +\bar X = \sum_{i=1}^n x_i p(x_i), +\]

+

where \(p(x_i) = 1/n\).

+
+

Example Find the center of mass of the bars

+

Let’s go through an example of illustrating how the sample mean is the center of mass of observed data. Below we plot Galton’s fathers and sons data:

+

Loading in and displaying the Galton data

+
library(UsingR); data(galton); library(ggplot2); library(reshape2)
+longGalton <- melt(galton, measure.vars = c("child", "parent"))
+g <- ggplot(longGalton, aes(x = value)) + geom_histogram(aes(y = ..density..,  fill = variable), binwidth=1, color = "black") + geom_density(size = 2)
+g <- g + facet_grid(. ~ variable)
+g
+
+
+Galton’s Data +

Galton’s Data

+
+

Using rStudio’s manipulate package, you can try moving the histogram around and see what value balances it out. Be sure to watch the video to see this in action.

+

Using manipulate to explore the mean

+
library(manipulate)
+myHist <- function(mu){
+    g <- ggplot(galton, aes(x = child))
+    g <- g + geom_histogram(fill = "salmon",
+      binwidth=1, aes(y = ..density..), color = "black")
+    g <- g + geom_density(size = 2)
+    g <- g + geom_vline(xintercept = mu, size = 2)
+    mse <- round(mean((galton$child - mu)^2), 3)  
+    g <- g + labs(title = paste('mu = ', mu, ' MSE = ', mse))
+    g
+}
+manipulate(myHist(mu), mu = slider(62, 74, step = 0.5))
+
+

Going through this exercise, you find that the point that balances out the histogram is the empirical mean. (Note there’s a small distinction here that comes about from rounding with the histogram bar widths, but ignore that for the time being.) If the bars of the histogram are from the observed data, the point that balances it out is the empirical mean; if the bars are the true population probabilities (which we don’t know of course) then the point is the population mean. Let’s now go through some examples of mathematically calculating the population mean.

+
+
+

The center of mass is the empirical mean

+
+Histogram illustration +

Histogram illustration

+
+
+
+

Example of a population mean, a fair coin

+

Watch the video before beginning here.

+

Suppose a coin is flipped and \(X\) is declared 0 or 1 corresponding to a head or a tail, respectively. What is the expected value of \(X\)?

+

\[ +E[X] = .5 \times 0 + .5 \times 1 = .5 +\]

+

Note, if thought about geometrically, this answer is obvious; if two equal weights are spaced at 0 and 1, the center of mass will be 0.5.

+
+Fair coin mass function +

Fair coin mass function

+
+
+
+

What about a biased coin?

+

Suppose that a random variable, \(X\) , is so that \(P(X=1) = p\) and \(P(X=0) = (1 - p)\) (This is a biased coin when \(p\neq 0.5\).) What is its expected value?

+

\[ +E[X] = 0 * (1 - p) + 1 * p = p +\]

+

Notice that the expected value isn’t a value that the coin can take in the same way that the sample proportion of heads will also likely be neither 0 nor 1.

+

This coin example is not exactly trivial as it serves as the basis for a random sample of any population for a binary trait. So, we might model the answer from an election polling question as if it were a coin flip.

+
+
+

Example Die Roll

+

Suppose that a die is rolled and \(X\) is the number face up. What is the expected value of \(X\)?

+

\[ +E[X] = 1 \times \frac{1}{6} + 2 \times \frac{1}{6} + +3 \times \frac{1}{6} + 4 \times \frac{1}{6} + +5 \times \frac{1}{6} + 6 \times \frac{1}{6} = 3.5 +\]

+

Again, the geometric argument makes this answer obvious without calculation.

+
+Bar graph of die probabilities +

Bar graph of die probabilities

+
+
+
+
+

Continuous random variables

+

Watch this video before beginning.

+

For a continuous random variable, \(X\), with density, \(f\), the expected value is again exactly the center of mass of the density. Think of it like cutting the continuous density out of a thick piece of wood and trying to find the point where it balances out.

+
+

Example

+

Consider a density where \(f(x) = 1\) for \(x\) between zero and one. Suppose that \(X\) follows this density; what is its expected value?

+
+Uniform Density +

Uniform Density

+
+

The answer is clear since the density looks like a box, it would balance out exactly in the middle, 0.5.

+
+
+

Facts about expected values

+

Recall that expected values are properties of population distributions. The expected value, or mean, height is the center of the population density of heights.

+

Of course, the average of ten randomly sampled people’s height is itself a random variable, in the same way that the average of ten die rolls is itself a random number. Thus, the distribution of heights gives rise to the distribution of averages of ten heights in the same way that distribution associated with a die roll gives rise to the distribution of the average of ten dice.

+

An important question to ask is: “What does the distribution of averages look like?”. This question is important, since it tells us things about averages, the best way to estimate the population mean, when we only get to observe one average.

+

Consider the die rolls again. If wanted to know the distribution of averages of 100 die rolls, you could (at least in principle) roll 100 dice, take the average and repeat that process. Imagine, if you could only roll the 100 dice once. Then we would have direct information about the distribution of die rolls (since we have 100 of them), but we wouldn’t have any direct information about the distribution of the average of 100 die rolls, since we only observed one average.

+

Fortunately, the mathematics tells us about that distribution. Notably, it’s centered at the same spot as the original distribution! Thus, the distribution of the estimator (the sample mean) is centered at the distribution of what it’s estimating (the population mean). When the expected value of an estimator is what its trying to estimate, we say that the estimator is unbiased.

+

Let’s go through several simulation experiments to see this more fully.

+
+
+
+

Simulation experiments

+
+

Standard normals

+

Consider simulating a lot of standard normals and plotting a histogram (the blue density). Now consider simulating lots of averages of 10 standard normals and plotting their histogram (the salmon colored density). Notice that they’re centered in the same spot! It’s also more concentrated around that point. (We’ll discuss that more in the next lectures).

+
+Simulation of normals +

Simulation of normals

+
+
+
+

Averages of x die rolls

+

Consider rolling a die a lot of times and taking a histogram of the results, that’s the left most plot. The bars are equally distributed at the six possible outcomes and thus the histogram is centered around 3.5. Now consider simulating lots of averages of 2 dice. Its histogram is also centered at 3.5. So is it for 3 and 4. Notice also the distribution gets increasing Gaussian looking (like a bell curve) and increasingly concentrated around 3.5.

+
+Simulation of die rolls +

Simulation of die rolls

+
+
+
+

Averages of x coin flips

+

For the coin flip simulation exactly the same occurs. All of the distributions are centered around 0.5.

+
+Simulation of coin flips +

Simulation of coin flips

+
+
+
+
+

Summary notes

+
    +
  • Expected values are properties of distributions.
    • +
  • The population mean is the center of mass of population.
    • +
  • The sample mean is the center of mass of the observed data.
    • +
  • The sample mean is an estimate of the population mean.
    • +
  • The sample mean is unbiased: the population mean of its distribution is the mean that it’s trying to estimate.
    • +
  • The more data that goes into the sample mean, the more. concentrated its density / mass function is around the population mean.
    • +
+
+
+

Exercises

+
    +
  1. A standard die takes the values 1, 2, 3, 4, 5, 6 with equal probability. What is the expected value?
    2. +
  2. Consider a density that is uniform from -1 to 1. (I.e. has height equal to 1/2 and looks like a box starting at -1 and ending at 1). What is the mean of this distribution?
    3. +
  3. If a population has mean \(\mu\), what is the mean of the distribution of averages of 20 observations from this distribution?
    4. +
  4. You are playing a game with a friend where you flip a coin and if it comes up heads you give her \(X\) dollars and if it comes up tails she gives you \(Y\) dollars. The odds that the coin is heads is \(d\). What is your expected earnings? Watch a video of the solution to this problem and look at the problem and the solution here..
    5. +
  5. If you roll ten standard dice, take their average, then repeat this process over and over and construct a histogram what would it be centered at? Watch a video solution here and see the original problem here.
    6. +
+
+
+
+

Variation

+
+

The variance

+

Watch this video before beginning.

+

Recall that the mean of distribution was a measure of its center. The variance, on the other hand, is a measure of spread. To get a sense, the plot below shows a series of increasing variances.

+
+Distributions with increasing variance +

Distributions with increasing variance

+
+

We saw another example of how variances changed in the last chapter when we looked at the distribution of averages; they were always centered at the same spot as the original distribution, but are less spread out. Thus, it is less likely for sample means to be far away from the population mean than it is for individual observations. (This is why the sample mean is a better estimate than the population mean.)

+

If \(X\) is a random variable with mean \(\mu\), the variance of \(X\) is defined as

+

\[ +Var(X) = E[(X - \mu)^2] = E[X^2] - E[X]^2. +\]

+

The rightmost equation is the shortcut formula that is almost always used for calculating variances in practice.
+Thus the variance is the expected (squared) distance from the mean. Densities with a higher variance are more spread out than densities with a lower variance. The square root of the variance is called the standard deviation. The main benefit of working with standard deviations is that they have the same units as the data, whereas the variance has the units squared.

+

In this class, we’ll only cover a few basic examples for calculating a variance. Otherwise, we’re going to use the ideas without the formalism. Also remember, what we’re talking about is the population variance. It measures how spread out the population of interest is, unlike the sample variance which measures how spread out the observed data are. Just like the sample mean estimates the population mean, the sample variance will estimate the population variance.

+
+

Example

+

What’s the variance from the result of a toss of a die? First recall that \(E[X] = 3.5\), as we discussed in the previous lecture. Then let’s calculate the other bit of information that we need, \(E[X^2]\).

+

\[E[X^2] = 1 ^ 2 \times \frac{1}{6} + 2 ^ 2 \times \frac{1}{6} + 3 ^ 2 \times \frac{1}{6} + 4 ^ 2 \times \frac{1}{6} + 5 ^ 2 \times \frac{1}{6} + 6 ^ 2 \times \frac{1}{6} = 15.17\]

+

Thus now we can calculate the variance as:

+

\[Var(X) = E[X^2] - E[X]^2 \approx 2.92.\]

+
+
+

Example

+

What’s the variance from the result of the toss of a (potentially biased) coin with probability of heads (1) of \(p\)? First recall that \(E[X] = 0 \times (1 - p) + 1 \times p = p.\) Secondly, recall that since \(X\) is either 0 or 1, \(X^2 = X\). So we know that:

+

\[E[X^2] = E[X] = p.\]

+

Thus we can now calculate the variance of a coin flip as \(Var(X) = E[X^2] - E[X]^2 = p - p^2 = p(1 - p).\) This is a well known formula, so it’s worth committing to memory. It’s interesting to note that this function is maximized at \(p = 0.5\). The plot below shows this by plotting \(p(1-p)\) by \(p\).

+

Plotting the binomial variance

+
p = seq(0 , 1, length = 1000)
+y = p * (1 - p)
+plot(p, y, type = "l", lwd = 3, frame = FALSE)
+
+
+Plot of the binomial variance +

Plot of the binomial variance

+
+
+
+
+

The sample variance

+

The sample variance is the estimator of the population variance. Recall that the population variance is the expected squared deviation around the population mean. The sample variance is (almost) the average squared deviation of observations around the sample mean. It is given by

+

\[ +S^2 = \frac{\sum_{i=1} (X_i - \bar X)^2}{n-1} +\]

+

The sample standard deviation is the square root of the sample variance. Note again that the sample variance is almost, but not quite, the average squared deviation from the sample mean since we divide by \(n-1\) instead of \(n\). Why do we do this you might ask? To answer that question we have to think in the terms of simulations. Remember that the sample variance is a random variable, thus it has a distribution and that distribution has an associated population mean. That mean is the population variance that we’re trying to estimate if we divide by \((n-1)\) rather than \(n\).

+

It is also nice that as we collect more data the distribution of the sample variance gets more concentrated around the population variance that it’s estimating.

+
+
+

Simulation experiments

+

Watch this video before beginning.

+
+

Simulating from a population with variance 1

+

Let’s try simulating collections of standard normals and taking the variance. If we repeat this over and over, we get a sense of the distribution of sample variances variances.

+
+Simulation of variances of samples of standard normals +

Simulation of variances of samples of standard normals

+
+

Notice that these histograms are always centered in the same spot, 1. In other words, the sample variance is an unbiased estimate of the population variances. Notice also that they get more concentrated around the 1 as more data goes into them. Thus, sample variances comprised of more observations are less variable than sample variances comprised of fewer.

+
+
+

Variances of x die rolls

+

Let’s try the same thing, now only with die rolls instead of simulating standard normals. In this experiment, we simulated samples of die rolls, took the variance and then repeated that process over and over. What is plotted are histograms of the collections of sample variances.

+
+Simulated distributions of variances of dies +

Simulated distributions of variances of dies

+
+

Recall that we calculated the variance of a die roll as 2.92 earlier on in this chapter. Notice each of the histograms are centered there. In addition, they get more concentrated around 2.92 as more the variances are comprised of more dice.

+
+
+
+

The standard error of the mean

+

At last, we finally get to a perhaps very surprising (and useful) fact: how to estimate the variability of the mean of a sample, when we only get to observe one realization. Recall that the average of random sample from a population is itself a random variable having a distribution, which in simulation settings we can explore by repeated sampling averages. We know that this distribution is centered around the population mean, \(E[\bar X] = \mu\). We also know the variance of the distribution of means of random samples.

+

The variance of the sample mean is: \(Var(\bar X) = \sigma^2 / n\) where \(\sigma^2\) is the variance of the population being sampled from.

+

This is very useful, since we don’t have repeat sample means to get its variance directly using the data. We already know a good estimate of \(\sigma^2\) via the sample variance. So, we can get a good estimate of the variability of the mean, even though we only get to observe 1 mean.

+

Notice also this explains why in all of our simulation experiments the variance of the sample mean kept getting smaller as the sample size increased. This is because of the square root of the sample size in the denominator.

+

Often we take the square root of the variance of the mean to get the standard deviation of the mean. We call the standard deviation of a statistic its standard error.

+
+

Summary notes

+
    +
  • The sample variance, \(S^2\), estimates the population variance, \(\sigma^2\).
    • +
  • The distribution of the sample variance is centered around \(\sigma^2\).
    • +
  • The variance of the sample mean is \(\sigma^2 / n\).
    • +
  • Its logical estimate is \(s^2 / n\).
    • +
  • The logical estimate of the standard error is \(S / \sqrt{n}\).
    • +
  • \(S\), the standard deviation, talks about how variable the population is.
    • +
  • \(S/\sqrt{n}\), the standard error, talks about how variable averages of random samples of size \(n\) from the population are.
    • +
+
+
+

Simulation example 1: standard normals

+

Watch this video before beginning.

+

Standard normals have variance 1. Let’s try sampling means of \(n\) standard normals. If our theory is correct, they should have standard deviation \(1/\sqrt{n}\)

+

Simulating means of random normals

+
> nosim <- 1000
+> n <- 10
+## simulate nosim averages of 10 standard normals
+> sd(apply(matrix(rnorm(nosim * n), nosim), 1, mean))
+[1] 0.3156
+## Let's check to make sure that this is sigma / sqrt(n)
+> 1 / sqrt(n)
+[1] 0.3162
+
+

So, in this simulation, we simulated 1000 means of 10 standard normals. Our theory says the standard deviation of averages of 10 standard normals must be \(1/\sqrt{n}\). Taking the standard deviation of the 10000 means yields nearly exactly that. (Note that it’s only close, 0.3156 versus 0.31632. To get it to be exact, we’d have to simulate infinitely many means.)

+
+
+

Simulation example 2: uniform density

+

Standard uniforms have variance \(1/12\). Our theory mandates that means of random samples of \(n\) uniforms have sd \(1/\sqrt{12 \times n}\). Let’s try it with a simulation.

+

Simulating means of uniforms

+
> nosim <- 1000
+> n <- 10
+> sd(apply(matrix(runif(nosim * n), nosim), 1, mean))
+[1] 0.09017
+> 1 / sqrt(12 * n)
+[1] 0.09129
+
+
+
+

Simulation example 3: Poisson

+

Poisson(4) random variables have variance \(4\). Thus means of random samples of \(n\) Poisson(4) should have standard deviation \(2/\sqrt{n}\). Again let’s try it out.

+

Simulating means of Poisson variates

+
> nosim <- 1000
+> n <- 10
+> sd(apply(matrix(rpois(nosim * n, 4), nosim), 1, mean))
+[1] 0.6219
+> 2 / sqrt(n)
+[1] 0.6325
+
+
+
+

Simulation example 4: coin flips

+

Our last example is an important one. Recall that the variance of a coin flip is \(p (1 - p)\). Therefore the standard deviation of the average of \(n\) coin flips should be \(\sqrt{\frac{p(1-p)}{n}}\).

+

Let’s just do the simulation with a fair coin. Such coin flips have variance 0.25. Thus means of random samples of \(n\) coin flips have sd \(1 / (2 \sqrt{n})\). Let’s try it.

+

Simulating means of coin flips

+
> nosim <- 1000
+> n <- 10
+> sd(apply(matrix(sample(0 : 1, nosim * n, replace = TRUE),
+                nosim), 1, mean))
+[1] 0.1587
+> 1 / (2 * sqrt(n))
+[1] 0.1581
+
+
+
+
+

Data example

+

Watch this before beginning.

+

Now let’s work through a data example to show how the standard error of the mean is used in practice. We’ll use the father.son height data from Francis Galton.

+

Loading the data

+
library(UsingR); data(father.son);
+x <- father.son$sheight
+n<-length(x)
+
+

Here’s a histogram of the sons’ heights from the dataset. Let’ calculate different variances and interpret them in this context.

+
+Histogram of the sons’ heights +

Histogram of the sons’ heights

+
+

Loading the data

+
>round(c(var(x), var(x) / n, sd(x), sd(x) / sqrt(n)),2)
+[1] 7.92 0.01 2.81 0.09
+
+

The first number, 7.92, and its square root, 2.81, are the estimated variance and standard deviation of the sons’ heights. Therefore, 7.92 tells us exactly how variable sons’ heights were in the data and estimates how variable sons’ heights are in the population. In contrast 0.01, and the square root 0.09, estimate how variable averages of \(n\) sons’ heights are.

+

Therefore, the smaller numbers discuss the precision of our estimate of the mean of sons’ heights. The larger numbers discuss how variable sons’ heights are in general.

+
+
+

Summary notes

+
    +
  • The sample variance estimates the population variance.
    • +
  • The distribution of the sample variance is centered at what its estimating.
    • +
  • It gets more concentrated around the population variance with larger sample sizes.
    • +
  • The variance of the sample mean is the population variance divided by \(n\).
    • +
  • The square root is the standard error.
    • +
  • It turns out that we can say a lot about the distribution of averages from random samples, even though we only get one to look at in a given data set.
    • +
+
+
+

Exercises

+
    +
  1. If I have a random sample from a population, the sample variance is an estimate of?
    2. +
+
    +
  • The population standard deviation.
    • +
  • The population variance.
    • +
  • The sample variance.
    • +
  • The sample standard deviation.
    • +
+
    +
  1. The distribution of the sample variance of a random sample from a population is centered at what?
    2. +
+
    +
  • The population variance.
    • +
  • The population mean.
    • +
+
    +
  1. I keep drawing samples of size \(n\) from a population with variance \(\sigma^2\) and taking their average. I do this thousands of times. If I were to take the variance of the collection of averages, about what would it be?
    2. +
  2. You get a random sample of \(n\) observations from a population and take their average. You would like to estimate the variability of averages of \(n\) observations from this population to better understand how precise of an estimate it is. Do you need to repeated collect averages to do this?
    3. +
+
    +
  • No, we can multiply our estimate of the population variance by \(1/n\) to get a good estimate of the variability of the average.
    • +
  • Yes, you have to get repeat averages.
    • +
+
    +
  1. A random variable takes the value -4 with probability .2 and 1 with probability .8. What is the variance of this random variable? Watch a video solution to this problem. and look at a version with a worked out solution.
    2. +
  2. If \(\bar X\) and \(\bar Y\) are comprised of n iid random variables arising from distributions having means \(\mu_x\) and \(\mu_y\), respectively and common variance \(\sigma^2\) what is the variance \(\bar X - \bar Y\)? Watch a video solution to this problem here and see a typed up solution here
    3. +
  3. Let \(X\) be a random variable having standard deviation \(\sigma\). What can be said about the variance of \(X /\sigma\)? Watch a video solution to this problem here and typed up solutions here.
    4. +
  4. Consider the following pmf given in R by the code p <- c(.1, .2, .3, .4) and ’x <- 2 : 5`. What is the variance? Watch a video solution to this problem here and here is the problem worked out.
    5. +
  5. If you roll ten standard dice, take their average, then repeat this process over and over and construct a histogram, what would be its variance expressed to 3 decimal places? Watch a video solution here and see the text here.
    6. +
+
+
+
+

Some common distributions

+
+

The Bernoulli distribution

+

The Bernoulli distribution arises as the result of a binary outcome, such as a coin flip. Thus, Bernoulli random variables take (only) the values 1 and 0 with probabilities of (say) \(p\) and \(1-p\), respectively. Recall that the PMF for a Bernoulli random variable \(X\) is \(P(X = x) = p^x (1 - p)^{1 - x}\).

+

The mean of a Bernoulli random variable is \(p\) and the variance is \(p(1 - p)\). If we let \(X\) be a Bernoulli random variable, it is typical to call \(X=1\) as a “success” and \(X=0\) as a “failure”.

+

If a random variable follows a Bernoulli distribution with success probability \(p\) we write that \(X \sim\) Bernoulli\((p)\).

+

Bernoulli random variables are commonly used for modeling any binary trait for a random sample. So, for example, in a random sample whether or not a participant has high blood pressure would be reasonably modeled as Bernoulli.

+
+
+

Binomial trials

+

The binomial random variables are obtained as the sum of iid Bernoulli trials. So if a Bernoulli trial is the result of a coin flip, a binomial random variable is the total number of heads.

+

To write it out as mathematics, let \(X_1,\ldots,X_n\) be iid Bernoulli\((p)\), then \(X = \sum_{i=1}^n X_i\) is a binomial random variable. We write out that \(X \sim\) Binomial\((n,p)\). The binomial mass function is

+

\[ +P(X = x) = +\left( +\begin{array}{c} + n \\ x +\end{array} +\right) +p^x(1 - p)^{n-x} +\]

+

where \(x=0,\ldots,n\). Recall that the notation

+

\[ +\left( + \begin{array}{c} n \\ x \end{array} +\right) = \frac{n!}{x!(n-x)!} +\]

+

(read “\(n\) choose \(x\)”) counts the number of ways of selecting \(x\) items out of \(n\) without replacement disregarding the order of the items. It turns out that \(n\) choose 0 is 1 and \(n\) choose 1 and \(n\) choose \(n-1\) are both \(n\).

+
+

Example

+

Suppose a friend has 8 children, \(7\) of which are girls and none are twins. If each gender has an independent \(50\)% probability for each birth, what’s the probability of getting \(7\) or more girls out of \(8\) births?

+

\[ +\left( +\begin{array}{c} + 8 \\ 7 +\end{array} +\right) .5^{7}(1-.5)^{1} ++ +\left( +\begin{array}{c} + 8 \\ 8 +\end{array} +\right) .5^{8}(1-.5)^{0} \approx 0.04 . +\]

+

Simulating means of coin flips

+
> choose(8, 7) * 0.5^8 + choose(8, 8) * 0.5^8
+[1] 0.03516
+> pbinom(6, size = 8, prob = 0.5, lower.tail = FALSE)
+[1] 0.03516
+
+
+
+
+

The normal distribution

+

Watch this video before beginning

+

The normal distribution is easily the handiest distribution in all of statistics. It can be used in an endless variety of settings. Moreover, as we’ll see later on in the course, sample means follow normal distributions for large sample sizes.

+

Remember the goal of probability modeling. We are assuming a probability distribution for our population as a way of parsimoniously characterizing it. In fact, the normal distribution only requires two numbers to characterize it. Specifically, a random variable is said to follow a normal or Gaussian distribution with mean \(\mu\) and variance \(\sigma^2\) if the associated density is:

+

\[ + (2\pi \sigma^2)^{-1/2}e^{-(x - \mu)^2/2\sigma^2}. +\]

+

If \(X\) is a RV with this density then \(E[X] = \mu\) and \(Var(X) = \sigma^2\). That is, the normal distribution is characterized by the mean and variance. We write \(X\sim N(\mu, \sigma^2)\) to denote a normal random variable. When \(\mu = 0\) and \(\sigma = 1\) the resulting distribution is called the standard normal distribution. Standard normal RVs are often labeled \(Z\)

+

Consider an example, if we say that intelligence quotients are normally distributed with a mean of 100 and a standard deviation of 15. Then, we are saying that if we randomly sample a person from this population, the probability that they have an IQ of say 120 or larger, is governed by a normal distribution with a mean of 100 and a variance of \(15^2\).

+

Taken another way, if we know that the population is normally distributed then to estimate everything about the population, we need only estimate the population mean and variance. (Estimated by the sample mean and the sample variance.)

+
+

Reference quantiles for the standard normal

+

The normal distribution is so important that it is useful to memorize reference probabilities and quantiles. The image below shows reference lines at 0, 1, 2 and 3 standard deviations above and below 0. This is for the standard normal; however, all of the rules apply to non standard normals as 0, 1, 2 and 3 standard deviations above and below \(\mu\), the population mean.

+
+Standard normal reference lines. +

Standard normal reference lines.

+
+

The most relevant probabilities are.

+
    +
  1. Approximately 68%, 95% and 99% of the normal density lies within 1, 2 and 3 standard deviations from the mean, respectively.
    2. +
  2. -1.28, -1.645, -1.96 and -2.33 are the \(10^{th}\), \(5^{th}\), \(2.5^{th}\) and \(1^{st}\) percentiles of the standard normal distribution, respectively.
    3. +
  3. By symmetry, 1.28, 1.645, 1.96 and 2.33 are the \(90^{th}\), \(95^{th}\), \(97.5^{th}\) and \(99^{th}\) percentiles of the standard normal distribution, respectively.
    4. +
+
+
+

Shifting and scaling normals

+

Since the normal distribution is characterized by only the mean and variance, which are a shift and a scale, we can transform normal random variables to be standard normals and vice versa. For example If \(X \sim N(\mu,\sigma^2)\) then:

+

\[Z = \frac{X -\mu}{\sigma} \sim N(0, 1).\]

+

If \(Z\) is standard normal

+

\[X = \mu + \sigma Z \sim N(\mu, \sigma^2)\]

+

then \(X\) is \(X \sim N(\mu,\sigma^2)\). We can use these facts to answer questions about non-standard normals by relating them back to the standard normal.

+
+
+

Example

+

What is the \(95^{th}\) percentile of a \(N(\mu, \sigma^2)\) distribution? Quick answer in R qnorm(.95, mean = mu, sd = sigma). Alternatively, because we have the standard normal quantiles memorized, and we know that 1.645 is its 95th percentile, the answer has to be \(\mu + \sigma 1.645\).

+

In general, \(\mu + \sigma z_0\) where \(z_0\) is the appropriate standard normal quantile.

+

To put some context on our previous setting, population mean BMI for men is reported as 29 \(kg/mg^2\) with a standard deviation of 4.73. Assuming normality of BMI, what is the population \(95^{th}\) percentile? The answer is then:

+

\[ +29 + 4.73 \times 1.645 = 36.78. +\]

+

Or alternatively, we could simply type r qnorm(.95, 29, 4.73) in R.

+

Now let’s reverse the process. Imaging asking what’s the probability that a randomly drawn subject from this population has a BMI less than 24.27? Notice that

+

\[ +\frac{24.27 - 29}{4.73} = -1. +\]

+

Therefore, 24.27 is 1 standard deviation below the mean. We know that 16% lies below or above 1 standard deviation from the mean. Thus 16% lies below. Alternatively, pnorm(24.27, 29, 4.73) yields the result.

+
+
+

Example

+

Assume that the number of daily ad clicks for a company is (approximately) normally distributed with a mean of 1020 and a standard deviation of 50. What’s the probability of getting more than 1,160 clicks in a day? Notice that:

+

\[ +\frac{1160 - 1020}{50} = 2.8 +\]

+

Therefore, 1,160 is 2.8 standard deviations above the mean. We know from our standard normal quantiles that the probability of being larger than 2 standard deviation is 2.5% and 3 standard deviations is far in the tail. Therefore, we know that the probability has to be smaller than 2.5% and should be very small. We can obtain it exactly as r pnorm(1160, 1020, 50, lower.tail = FALSE) which is 0.3%. Note that we can also obtain the probability as r pnorm(2.8, lower.tail = FALSE).

+
+
+

Example

+

Consider the previous example again. What number of daily ad clicks would represent the one where 75% of days have fewer clicks (assuming days are independent and identically distributed)? We can obtain this as:

+

Finding a normal quantile

+
> qnorm(0.75, mean = 1020, sd = 50)
+[1] 1054
+
+
+
+
+

The Poisson distribution

+

Watch this video before beginning.

+

The Poisson distribution is used to model counts. It is perhaps only second to the normal distribution usefulness. In fact, the Bernoulli, binomial and multinomial distributions can all be modeled by clever uses of the Poisson.

+

The Poisson distribution is especially useful for modeling unbounded counts or counts per unit of time (rates). Like the number of clicks on advertisements, or the number of people who show up at a bus stop. (While these are in principle bounded, it would be hard to actually put an upper limit on it.) There is also a deep connection between the Poisson distribution and popular models for so-called event-time data. In addition, the Poisson distribution is the default model for so-called contingency table data, which is simply tabulations of discrete characteristics. Finally, when \(n\) is large and \(p\) is small, the Poisson is an accurate approximation to the binomial distribution.

+

The Poisson mass function is:

+

\[ +P(X = x; \lambda) = \frac{\lambda^x e^{-\lambda}}{x!} +\]

+

for \(x=0,1,\ldots\). The mean of this distribution is \(\lambda\). The variance of this distribution is also \(\lambda\). Notice that \(x\) ranges from 0 to \(\infty\). Therefore, the Poisson distribution is especially useful for modeling unbounded counts.

+
+

Rates and Poisson random variables

+

The Poisson distribution is useful for rates, counts that occur over units of time. Specifically, if \(X \sim Poisson(\lambda t)\) where \(\lambda = E[X / t]\) is the expected count per unit of time and \(t\) is the total monitoring time.

+
+
+

Example

+

The number of people that show up at a bus stop is Poisson with a mean of 2.5 per hour. If watching the bus stop for 4 hours, what is the probability that \(3\) or fewer people show up for the whole time?

+

Finding a normal quantile

+
> ppois(3, lambda = 2.5 * 4)
+[1] 0.01034
+
+

Therefore, there is about a 1% chance that 3 or fewer people show up. Notice the multiplication by four in the function argument. Since lambda is specified as events per hour we have to multiply by four to consider the number of events that occur in 4 hours.

+
+
+

Poisson approximation to the binomial

+

When \(n\) is large and \(p\) is small the Poisson distribution is an accurate approximation to the binomial distribution. Formally, if \(X \sim \mbox{Binomial}(n, p)\) then \(X\) is approximately Poisson where \(\lambda = n p\) provided that \(n\) is large \(p\) is small.

+
+

Example, Poisson approximation to the binomial

+

We flip a coin with success probability 0.01 five hundred times. What’s the probability of 2 or fewer successes?

+

Finding a normal quantile

+
> pbinom(2, size = 500, prob = 0.01)
+[1] 0.1234
+> ppois(2, lambda = 500 * 0.01)
+[1] 0.1247
+
+

So we can see that the probabilities agree quite well. This approximation is often done as the Poisson model is a more convenient model in many respects.

+
+
+
+
+

Exercises

+
    +
  1. Your friend claims that changing the font to comic sans will result in more ad revenue on your web sites. When presented in random order, 9 pages out of 10 had more revenue when the font was set to comic sans. If it was really a coin flip for these 10 sites, what’s the probability of getting 9 or 10 out of 10 with more revenue for the new font?
    2. +
  2. A software company is doing an analysis of documentation errors of their products. They sampled their very large codebase in chunks and found that the number of errors per chunk was approximately normally distributed with a mean of 11 errors and a standard deviation of 2. When randomly selecting a chunk from their codebase, whats the probability of fewer than 5 documentation errors?
    3. +
  3. The number of search entries entered at a web site is Poisson at a rate of 9 searches per minute. The site is monitored for 5 minutes. What is the probability of 40 or fewer searches in that time frame?
    4. +
  4. Suppose that the number of web hits to a particular site are approximately normally distributed with a mean of 100 hits per day and a standard deviation of 10 hits per day. What’s the probability that a given day has fewer than 93 hits per day expressed as a percentage to the nearest percentage point? Watch a video solution and see the problem.
    5. +
  5. Suppose that the number of web hits to a particular site are approximately normally distributed with a mean of 100 hits per day and a standard deviation of 10 hits per day. What number of web hits per day represents the number so that only 5% of days have more hits? Watch a video solution and see the problem and solution.
    6. +
  6. Suppose that the number of web hits to a particular site are approximately normally distributed with a mean of 100 hits per day and a standard deviation of 10 hits per day. Imagine taking a random sample of 50 days. What number of web hits would be the point so that only 5% of averages of 50 days of web traffic have more hits? Watch a video solution and see the problem and solution.
    7. +
  7. You don’t believe that your friend can discern good wine from cheap. Assuming that you’re right, in a blind test where you randomize 6 paired varieties (Merlot, Chianti, …) of cheap and expensive wines. What is the change that she gets 5 or 6 right? Watch a video solution and see the original problem.
    8. +
  8. The number of web hits to a site is Poisson with mean 16.5 per day. What is the probability of getting 20 or fewer in 2 days? Watch a video solution and see a written solution.
    9. +
+
+
+
+

Asymptopia

+
+

Asymptotics

+

Watch this video before beginning.

+

Asymptotics is the term for the behavior of statistics as the sample size limits to infinity. Asymptotics are incredibly useful for simple statistical inference and approximations. Asymptotics often make hard problems easy and difficult calculations simple. We will not cover the philosophical considerations in this book, but is true nonetheless, that asymptotics often lead to nice understanding of procedures. In fact, the ideas of asymptotics are so important form the basis for frequency interpretation of probabilities by considering the long run proportion of times an event occurs.

+

Some things to bear in mind about the seemingly magical nature of asymptotics. There’s no free lunch and unfortunately, asymptotics generally give no assurances about finite sample performance.

+
+
+

Limits of random variables

+

We’ll only talk about the limiting behavior of one statistic, the sample mean. Fortunately, for the sample mean there’s a set of powerful results. These results allow us to talk about the large sample distribution of sample means of a collection of iid observations.

+

The first of these results we intuitively already know. It says that the average limits to what its estimating, the population mean. This result is called the Law of Large Numbers. It simply says that if you go to the trouble of collecting an infinite amount of data, you estimate the population mean perfectly. Note there’s sampling assumptions that have to hold for this result to be true. The data have to be iid.

+

A great example of this comes from coin flipping. Imagine if \(\bar X_n\) is the average of the result of \(n\) coin flips (i.e. the sample proportion of heads). The Law of Large Numbers states that as we flip a coin over and over, it eventually converges to the true probability of a head.

+
+

Law of large numbers in action

+

Let’s try using simulation to investigate the law of large numbers in action. Let’s simulate a lot of standard normals and plot the cumulative means. If the LLN is correct, the line should converge to 0, the mean of the standard normal distribution.

+

Finding a normal quantile

+
n <- 10000
+means <- cumsum(rnorm(n))/(1:n)
+library(ggplot2)
+g <- ggplot(data.frame(x = 1:n, y = means), aes(x = x, y = y))
+g <- g + geom_hline(yintercept = 0) + geom_line(size = 2)
+g <- g + labs(x = "Number of obs", y = "Cumulative mean")
+g
+
+
+Cumulative average from standard normal simulations. +

Cumulative average from standard normal simulations.

+
+
+
+

Law of large numbers in action, coin flip

+

Let’s try the same thing, but for a fair coin flip. We’ll simulate a lot of coin flips and plot the cumulative proportion of heads.

+

Finding a normal quantile

+
means <- cumsum(sample(0:1, n, replace = TRUE))/(1:n)
+g <- ggplot(data.frame(x = 1:n, y = means), aes(x = x, y = y))
+g <- g + geom_hline(yintercept = 0.5) + geom_line(size = 2)
+g <- g + labs(x = "Number of obs", y = "Cumulative mean")
+g
+
+
+Cumulative proportion of heads from a sequence of coin flips. +

Cumulative proportion of heads from a sequence of coin flips.

+
+
+
+

Discussion

+

An estimator is called consistent if it converges to what you want to estimate. Thus, the LLN says that the sample mean of iid sample is consistent for the population mean. Typically, good estimators are consistent; it’s not too much to ask that if we go to the trouble of collecting an infinite amount of data that we get the right answer. The sample variance and the sample standard deviation of iid random variables are consistent as well.

+
+
+
+

The Central Limit Theorem

+

Watch this video before beginning.

+

The Central Limit Theorem (CLT) is one of the most important theorems in statistics. For our purposes, the CLT states that the distribution of averages of iid variables becomes that of a standard normal as the sample size increases. Consider this fact for a second. We already know the mean and standard deviation of the distribution of averages from iid samples. The CLT gives us an approximation to the full distribution! Thus, for iid samples, we have a good sense of distribution of the average event though: (1) we only observed one average and (2) we don’t know what the population distribution is. Because of this, the CLT applies in an endless variety of settings and is one of the most important theorems ever discovered.

+

The formal result is that

+

\[ +\frac{\bar X_n - \mu}{\sigma / \sqrt{n}}= +\frac{\sqrt n (\bar X_n - \mu)}{\sigma} += \frac{\mbox{Estimate} - \mbox{Mean of estimate}}{\mbox{Std. Err. of estimate}} +\]

+

has a distribution like that of a standard normal for large \(n\). Replacing the standard error by its estimated value doesn’t change the CLT.

+

The useful way to think about the CLT is that \(\bar X_n\) is approximately \(N(\mu, \sigma^2 / n)\).

+
+
+

CLT simulation experiments

+

Let’s try simulating lots of averages from various distributions and showing that the resulting distribution looks like a bell curve.

+
+

Die rolling

+
    +
  • Simulate a standard normal random variable by rolling \(n\) (six sided) dice.
    • +
  • Let \(X_i\) be the outcome for die \(i\).
    • +
  • Then note that \(\mu = E[X_i] = 3.5\).
    • +
  • Recall also that \(Var(X_i) = 2.92\).
    • +
  • SE \(\sqrt{2.92 / n} = 1.71 / \sqrt{n}\).
    • +
  • Lets roll \(n\) dice, take their mean, subtract off 3.5, and divide by \(1.71 / \sqrt{n}\) and repeat this over and over.
    • +
+
+Result of coin CLT simulation. +

Result of coin CLT simulation.

+
+

It’s pretty remarkable that the approximation works so well with so few rolls of the die. So, if you’re stranded on an island, and need to simulate a standard normal without a computer, but you do have a die, you can get a pretty good approximation with 10 rolls even.

+
+
+

Coin CLT

+

In fact the oldest application of the CLT is to the idea of flipping coins (by de Moivre). Let \(X_i\) be the 0 or 1 result of the \(i^{th}\) flip of a possibly unfair coin. The sample proportion, say \(\hat p\), is the average of the coin flips. We know that:

+
    +
  • \(E[X_i] = p\),
    • +
  • \(Var(X_i) = p(1-p)\),
    • +
  • \(\sqrt{Var(\hat p)} = \sqrt{p(1-p)/n}\).
    • +
+

Furthermore, because of the CLT, we also know that:

+

\[ +\frac{\hat p - p}{\sqrt{p(1-p)/n}} +\]

+

will be approximately normally distributed.

+

Let’s test this by flipping a coin \(n\) times, taking the sample proportion of heads, subtract off 0.5 and multiply the result by \(2 \sqrt{n}\) (divide by \(1/(2 \sqrt{n})\)).

+
+Results of the coin CLT simulation. +

Results of the coin CLT simulation.

+
+

This convergence doesn’t look quite as good as the die, since the coin has fewer possible outcomes. In fact, among coins of various degrees of bias, the convergence to normality is governed by how far from 0.5 \(p\) is. Let’s redo the simulation, now using \(p=0.9\) instead of \(p=0.5\) like we did before.

+
+Results of the simulation when p=0.9 +

Results of the simulation when p=0.9

+
+

Notice that the convergence to normality is quite poor. Thus, be careful when using CLT approximations for sample proportions when your proportion is very close to 0 or 1.

+
+
+
+

Confidence intervals

+

Watch this video before beginning.

+

Confidence intervals are methods for quantifying uncertainty in our estimates. The fact that the interval has width characterizes that there is randomness that prevents us from getting a perfect estimate. Let’s go through how a confidence interval using the CLT is constructed.

+

According to the CLT, the sample mean, \(\bar X\), is approximately normal with mean \(\mu\) and standard deviation \(\sigma / \sqrt{n}\). Furthermore,

+

\[\mu + 2 \sigma /\sqrt{n}\]

+

is pretty far out in the tail (only 2.5% of a normal being larger than 2 sds in the tail). Similarly,

+

\[\mu - 2 \sigma /\sqrt{n}\]

+

is pretty far in the left tail (only 2.5% chance of a normal being smaller than 2 standard deviations in the tail). So the probability \(\bar X\) is bigger than \(\mu + 2 \sigma / \sqrt{n}\) or smaller than \(\mu - 2 \sigma / \sqrt{n}\) is 5%. Or equivalently, the probability that these limits contain \(\mu\) is 95%. The quantity:

+

\[\bar X \pm 2 \sigma /\sqrt{n}\]

+

is called a 95% interval for \(\mu\). The 95% refers to the fact that if one were to repeatedly get samples of size \(n\), about 95% of the intervals obtained would contain \(\mu\). The 97.5th quantile is 1.96 (so I rounded to 2 above). If instead of a 95% interval, you wanted a 90% interval, then you want (100 - 90) / 2 = 5% in each tail. Thus your replace the 2 with the 95th percentile, which is 1.645.

+
+

Example CI

+

Give a confidence interval for the average height of sons in Galton’s data.

+

Finding a confidence interval.

+
> library(UsingR)
+> data(father.son)
+> x <- father.son$sheight
+> (mean(x) + c(-1, 1) * qnorm(0.975) * sd(x)/sqrt(length(x)))/12
+[1] 5.710 5.738
+
+

Here we divided by 12 to get our interval in feet instead of inches. So we estimate the average height of the sons as 5.71 to 5.74 with 95% confidence.

+
+
+

Example using sample proportions

+

In the event that each \(X_i\) is 0 or 1 with common success probability \(p\) then \(\sigma^2 = p(1 - p)\). The interval takes the form:

+

\[ +\hat p \pm z_{1 - \alpha/2} \sqrt{\frac{p(1 - p)}{n}}. +\]

+

Replacing \(p\) by \(\hat p\) in the standard error results in what is called a Wald confidence interval for \(p\). Remember also that \(p(1 - p)\) is maximized at 1/4. Plugging this in and setting our \(Z\) quantile as 2 (which is about a 95% interval) we find that a quick and dirty confidence interval is:

+

\[\hat p \pm \frac{1}{\sqrt{n}}.\]

+

This is useful for doing quick confidence intervals for binomial proportions in your head.

+
+
+

Example

+

Your campaign advisor told you that in a random sample of 100 likely voters, 56 intent to vote for you. Can you relax? Do you have this race in the bag? Without access to a computer or calculator, how precise is this estimate?

+
+
> 1/sqrt(100)
+[1] 0.1
+
+

so a back of the envelope calculation gives an approximate 95% interval of (0.46, 0.66).

+

Thus, since the interval contains 0.5 and numbers below it, there’s not enough votes for you to relax; better go do more campaigning!

+

The basic rule of thumb is then, \(1/\sqrt{n}\) gives you a good estimate for the margin of error of a proportion. Thus, \(n=100\)
+for about 1 decimal place, 10,000 for 2, 1,000,000 for 3.

+
+
> round(1/sqrt(10^(1:6)), 3)
+[1] 0.316 0.100 0.032 0.010 0.003 0.001
+
+

We could very easily do the full Wald interval, which is less conservative (may provide a narrower interval). Remember the Wald interval for a binomial proportion is:

+

\[ +\hat p \pm Z_{1-\alpha/2} \sqrt{\frac{\hat p (1 - \hat p)}{n}}. +\]

+

Here’s the R code for our election setting, both coding it directly and using binom.test.

+
+
> 0.56 + c(-1, 1) * qnorm(0.975) * sqrt(0.56 * 0.44/100)
+[1] 0.4627 0.6573
+> binom.test(56, 100)$conf.int
+[1] 0.4572 0.6592
+
+
+
+
+

Simulation of confidence intervals

+

It is interesting to note that the coverage of confidence intervals describes an aggregate behavior. In other words the confidence interval describes the percentage of intervals that would cover the parameter being estimated if we were to repeat the experiment over and over. So, one can not technically say that the interval contains the parameter with probability 95%, say. So called Bayesian credible intervals address this issue at the expense (or benefit depending on who you ask) of adopting a Bayesian framework.

+

For our purposes, we’re using confidence intervals and so will investigate their frequency performance over repeated realizations of the experiment. We can do this via simulation. Let’s consider different values of \(p\) and look at the Wald interval’s coverage when we repeatedly create confidence intervals.

+

Code for investigating Wald interval coverage

+
n <- 20
+pvals <- seq(0.1, 0.9, by = 0.05)
+nosim <- 1000
+coverage <- sapply(pvals, function(p) {
+    phats <- rbinom(nosim, prob = p, size = n)/n
+    ll <- phats - qnorm(0.975) * sqrt(phats * (1 - phats)/n)
+    ul <- phats + qnorm(0.975) * sqrt(phats * (1 - phats)/n)
+    mean(ll < p & ul > p)
+})
+
+
+Plot of Wald interval coverage. +

Plot of Wald interval coverage.

+
+

The figure shows that if we were to repeatedly try experiments for any fixed value of \(p\), it’s rarely the case that our intervals will cover the value that they’re trying to estimate in 95% of them. This is bad, since covering the parameter that its estimating 95% of the time is the confidence interval’s only job!

+

So what’s happening? Recall that the CLT is an approximation. In this case \(n\) isn’t large enough for the CLT to be applicable for many of the values of \(p\). Let’s see if the coverage improves for larger \(n\).

+

Code for investigating Wald interval coverage

+
n <- 100
+pvals <- seq(0.1, 0.9, by = 0.05)
+nosim <- 1000
+coverage2 <- sapply(pvals, function(p) {
+    phats <- rbinom(nosim, prob = p, size = n)/n
+    ll <- phats - qnorm(0.975) * sqrt(phats * (1 - phats)/n)
+    ul <- phats + qnorm(0.975) * sqrt(phats * (1 - phats)/n)
+    mean(ll < p & ul > p)
+})
+
+
+Output of simulation with n=100. +

Output of simulation with \(n=100\).

+
+

Now it looks much better. Of course, increasing our sample size is rarely an option. There’s exact fixes to make this interval work better for small sample sizes.

+

However, for a quick fix is to take your data and add two successes and two failures. So, for example, in our election example, we would form our interval with 58 votes out of 104 sampled (disregarding that the actual numbers were 56 and 100). This interval is called the Agresti/Coull interval. This interval has much better coverage. Let’s show it via a simulation.

+

Code for investigating Agresti/Coull interval coverage when n=20.

+
n <- 20
+pvals <- seq(0.1, 0.9, by = 0.05)
+nosim <- 1000
+coverage <- sapply(pvals, function(p) {
+    phats <- (rbinom(nosim, prob = p, size = n) + 2)/(n + 4)
+    ll <- phats - qnorm(0.975) * sqrt(phats * (1 - phats)/n)
+    ul <- phats + qnorm(0.975) * sqrt(phats * (1 - phats)/n)
+    mean(ll < p & ul > p)
+})
+
+
+Coverage of the Agresti/Coull interval with n=20 +

Coverage of the Agresti/Coull interval with \(n=20\)

+
+

The coverage is better, if maybe a little conservative in the sense of being over the 95% line most of the time. If the interval is too conservative, it’s likely a little too wide. To see this clearly, imagine if we made our interval \(-\infty\) to \(\infty\). Then we would always have 100% coverage in any setting, but the interval wouldn’t be useful. Nonetheless, the Agrestic/Coull interval gives a much better trade off between coverage and width than the Wald interval.

+

In general, one should use the add two successes and failures method for binomial confidence intervals with smaller \(n\). For very small \(n\) consider using an exact interval (not covered in this class).

+
+
+

Poisson interval

+

Since the Poisson distribution is so central for data science, let’s do a Poisson confidence interval. Remember that if \(X \sim \mbox{Poisson}(\lambda t)\) then our estimate of \(\lambda\) is \(\hat \lambda = X/t\). Furthermore, we know that \(Var(\hat \lambda) = \lambda / t\) and so the natural estimate is \(\hat \lambda / t\). While it’s not immediate how the CLT applies in this case, the interval is of the familiar form

+

\[ +\mbox{Estimate} \pm Z_{1-\alpha/2} \mbox{SE}. +\]

+

So our Poisson interval is:

+

\[ +\hat \lambda \pm Z_{1-\alpha/2} \sqrt{\frac{\hat \lambda}{t}} +\]

+
+

Example

+

A nuclear pump failed 5 times out of 94.32 days. Give a 95% confidence interval for the failure rate per day.

+

Code for asymptotic Poisson confidence interval

+
> x <- 5
+> t <- 94.32
+> lambda <- x/t
+> round(lambda + c(-1, 1) * qnorm(0.975) * sqrt(lambda/t), 3)
+[1] 0.007 0.099
+
+

A non-asymptotic test, one that guarantees coverage, is also available. But, it has to be evaluated numerically.

+

Code for exact Poisson confidence interval

+
> poisson.test(x, T = 94.32)$conf
+[1] 0.01721 0.12371
+
+
+
+

Simulating the Poisson coverage rate

+

Let’s see how the asymptotic interval performs for lambda values near what we’re estimating.

+

Code for evaluating the coverage of the asymptotic Poisson confidence interval

+
lambdavals <- seq(0.005, 0.1, by = 0.01)
+nosim <- 1000
+t <- 100
+coverage <- sapply(lambdavals, function(lambda) {
+    lhats <- rpois(nosim, lambda = lambda * t)/t
+    ll <- lhats - qnorm(0.975) * sqrt(lhats/t)
+    ul <- lhats + qnorm(0.975) * sqrt(lhats/t)
+    mean(ll < lambda & ul > lambda)
+})
+
+
+Coverage of Poisson intervals for various values of lambda +

Coverage of Poisson intervals for various values of lambda

+
+

The coverage can be low for low values of lambda. In this case the asymptotics works as we increase the monitoring time, t. Here’s the coverage if we increase \(t\) to 1,000.

+
+Coverage of Poisson intervals for various values of lambda and t=1000 +

Coverage of Poisson intervals for various values of lambda and t=1000

+
+
+
+
+

Summary notes

+
    +
  • The LLN states that averages of iid samples. converge to the population means that they are estimating.
    • +
  • The CLT states that averages are approximately normal, with distributions.
    • +
  • centered at the population mean.
    • +
  • with standard deviation equal to the standard error of the mean.
    • +
  • CLT gives no guarantee that \(n\) is large enough.
    • +
  • Taking the mean and adding and subtracting the relevant. normal quantile times the SE yields a confidence interval for the mean.
    • +
  • Adding and subtracting 2 SEs works for 95% intervals.
    • +
  • Confidence intervals get wider as the coverage increases.
    • +
  • Confidence intervals get narrower with less variability or larger sample sizes.
    • +
  • The Poisson and binomial case have exact intervals that don’t require the CLT.
    • +
  • But a quick fix for small sample size binomial calculations is to add 2 successes and failures.
    • +
+
+
+

Exercises

+
    +
  1. I simulate 1,000,000 standard normals. The LLN says that their sample average must be close to?
    2. +
  2. About what is the probability of getting 45 or fewer heads out 100 flips of a fair coin? (Use the CLT, not the exact binomial calculation).
    3. +
  3. Consider the father.son data. Using the CLT and assuming that the fathers are a random sample from a population of interest, what is a 95% confidence mean height in inches?
    4. +
  4. The goal of a a confidence interval having coverage 95% is to imply that:
    5. +
+
    +
  • If one were to repeated collect samples and reconstruct the intervals, around 95% percent of them would contain the true mean being estimated.
    • +
  • The probability that the sample mean is in the interval is 95%.
    • +
+
    +
  1. The rate of search entries into a web site was 10 per minute when monitoring for an hour. Use R to calculate the exact Poisson interval for the rate of events per minute?
    2. +
  2. Consider a uniform distribution. If we were to sample 100 draws from a a uniform distribution (which has mean 0.5, and variance 1/12) and take their mean, \(\bar X\). What is the approximate probability of getting as large as 0.51 or larger? Watch this video solution and see the problem and solution here..
    3. +
+
+
+
+

t Confidence intervals

+
+

Small sample confidence intervals

+

Watch this video before beginning.

+

In the previous lecture, we discussed creating a confidence interval using the CLT. Our intervals took the form:

+

\[Est \pm Z \times SE_{Est}.\]

+

In this lecture, we discuss some methods for small samples, notably Gosset’s t distribution and t confidence intervals.

+

These intervals are of the form:

+

\[Est \pm t \times SE_{Est}.\]

+

So the only change is that we’ve replaced the Z quantile now with a t quantile. These are some of the handiest of intervals in all of statistics. If you want a rule between whether to use a t interval or normal interval, just always use the t interval.

+
+
+

Gosset’s t distribution

+

The t distribution was invented by William Gosset (under the pseudonym “Student”) in 1908. Fisher provided further mathematical details about the distribution later. This distribution has thicker tails than the normal. It’s indexed by a degrees of freedom and it gets more like a standard normal as the degrees of freedom get larger. It assumes that the underlying data are iid Gaussian with the result that

+

\[ +\frac{\bar X - \mu}{S/\sqrt{n}} +\]

+

follows Gosset’s t distribution with \(n-1\) degrees of freedom. (If we replaced \(s\) by \(\sigma\) the statistic would be exactly standard normal.) The interval is

+

\[\bar X \pm t_{n-1} S/\sqrt{n},\]

+

where \(t_{n-1}\) is the relevant quantile from the t distribution.

+
+

Code for manipulate

+

You can use rStudio’s manipulate function to to compare the t and Z distributions.

+

Code for investigating t and Z densities.

+
k <- 1000
+xvals <- seq(-5, 5, length = k)
+myplot <- function(df){
+  d <- data.frame(y = c(dnorm(xvals), dt(xvals, df)),
+                  x = xvals,
+                  dist = factor(rep(c("Normal", "T"), c(k,k))))
+  g <- ggplot(d, aes(x = x, y = y))
+  g <- g + geom_line(size = 2, aes(color = dist))
+  g
+}
+manipulate(myplot(mu), mu = slider(1, 20, step = 1))  
+
+

The difference is perhaps easier to see in the tails. Therefore, the following code plots the upper quantiles of the Z distribution by those of the t distribution.

+

Code for investigating the upper quantiles of the t and Z densities.

+
pvals <- seq(.5, .99, by = .01)
+myplot2 <- function(df){
+  d <- data.frame(n= qnorm(pvals),t=qt(pvals, df),
+                  p = pvals)
+  g <- ggplot(d, aes(x= n, y = t))
+  g <- g + geom_abline(size = 2, col = "lightblue")
+  g <- g + geom_line(size = 2, col = "black")
+  g <- g + geom_vline(xintercept = qnorm(0.975))
+  g <- g + geom_hline(yintercept = qt(0.975, df))
+  g
+}
+manipulate(myplot2(df), df = slider(1, 20, step = 1))
+
+
+
+

Summary notes

+

In this section, we give an overview of important facts about the t distribution.

+
    +
  • The t interval technically assumes that the data are iid normal, though it is robust to this assumption.
    • +
  • It works well whenever the distribution of the data is roughly symmetric and mound shaped.
    • +
  • Paired observations are often analyzed using the t interval by taking differences.
    • +
  • For large degrees of freedom, t quantiles become the same as standard normal quantiles; therefore this interval converges to the same interval as the CLT yielded.
    • +
  • For skewed distributions, the spirit of the t interval assumptions are violated.
    • +
  • Also, for skewed distributions, it doesn’t make a lot of sense to center the interval at the mean.
    • +
  • In this case, consider taking logs or using a different summary like the median.
    • +
  • For highly discrete data, like binary, other intervals are available.
    • +
+
+
+

Example of the t interval, Gosset’s sleep data

+

Watch this video before beginning.

+

In R typing r data(sleep) brings up the sleep data originally analyzed in Gosset’s Biometrika paper, which shows the increase in hours for 10 patients on two soporific drugs. R treats the data as two groups rather than paired.

+
+
+
+

The data

+

Loading Galton’s data.

+
> data(sleep)
+> head(sleep)
+   extra group ID
+ 1   0.7     1  1
+ 2  -1.6     1  2
+ 3  -0.2     1  3
+ 4  -1.2     1  4
+ 5  -0.1     1  5
+ 6   3.4     1  6
+
+

Here’s a plot of the data. In this plot paired observations are connected with a line.

+
+A plot of the pairs of observations from Galton’s sleep data. +

A plot of the pairs of observations from Galton’s sleep data.

+
+

Now let’s calculate the t interval for the differences from baseline to follow up. Below we give four different ways for calculating the interval.

+

Loading Galton’s data.

+
g1 <- sleep$extra[1 : 10]; g2 <- sleep$extra[11 : 20]
+difference <- g2 - g1
+mn <- mean(difference); s <- sd(difference); n <- 10
+## Calculating directly
+mn + c(-1, 1) * qt(.975, n-1) * s / sqrt(n)
+## using R's built in function
+t.test(difference)
+## using R's built in function, another format
+t.test(g2, g1, paired = TRUE)
+## using R's built in function, another format
+t.test(extra ~ I(relevel(group, 2)), paired = TRUE, data = sleep)
+## Below are the results (after a little formatting)
+        [,1] [,2]
+ [1,] 0.7001 2.46
+ [2,] 0.7001 2.46
+ [3,] 0.7001 2.46
+ [4,] 0.7001 2.46
+
+

Therefore, since our interval doesn’t include 0, our 95% confidence interval estimate for the mean change (follow up - baseline) is 0.70 to 2.45.

+
+
+

Independent group t confidence intervals

+

Watch this video before beginning.

+

Suppose that we want to compare the mean blood pressure between two groups in a randomized trial; those who received the treatment to those who received a placebo. The randomization is useful for attempting to balance unobserved covariates that might contaminate our results. Because of the randomization, it would be reasonable to compare the two groups without considering further variables.

+

We cannot use the paired t interval that we just used for Galton’s data, because the groups are independent. Person 1 from the treated group has no relationship with person 1 from the control group. Moreover, the groups may have different sample sizes, so taking paired differences may not even be possible even if it isn’t advisable in this setting.

+

We now present methods for creating confidence intervals for comparing independent groups.

+
+
+

Confidence interval

+

A \((1 - \alpha)\times 100\%\) confidence interval for the mean difference between the groups, \(\mu_y - \mu_x\) is:

+

\[ +\bar Y - \bar X \pm t_{n_x + n_y - 2, 1 - \alpha/2}S_p +\left( \frac{1}{n_x} + \frac{1}{n_y} \right)^{1/2}. +\]

+

The notation \(t_{n_x + n_y - 2, 1 - \alpha/2}\) means a t quantile with \(n_x + n_y - 2\) degrees of freedom. The pooled variance estimator is:

+

\[ +S_p^2 = \{(n_x - 1) S_x^2 + (n_y - 1) S_y^2\}/(n_x + n_y - 2). +\]

+

This variance estimate is used if one is willing to assume a constant variance across the groups. It is a weighted average of the group-specific variances, with greater weight given to whichever group has the larger sample size.

+

If there is some doubt about the constant variance assumption, assume a different variance per group, which we will discuss later.

+
+
+

Mistakenly treating the sleep data as grouped

+

Let’s first go through an example where we treat paired data as if it were independent. Consider Galton’s sleep data from before. In the code below, we do the R code for grouped data directly, and using the r t.test function.

+

Galton’s data treated as grouped and independent.

+
n1 <- length(g1); n2 <- length(g2)
+sp <- sqrt( ((n1 - 1) * sd(x1)^2 + (n2-1) * sd(x2)^2) / (n1 + n2-2))
+md <- mean(g2) - mean(g1)
+semd <- sp * sqrt(1 / n1 + 1/n2)
+rbind(
+  md + c(-1, 1) * qt(.975, n1 + n2 - 2) * semd,  
+  t.test(g2, g1, paired = FALSE, var.equal = TRUE)$conf,
+  t.test(g2, g1, paired = TRUE)$conf
+)
+
+

The results are:

+
+
[,1]  [,2]
+[1,] -0.2039 3.364
+[2,] -0.2039 3.364
+[3,]  0.7001 2.460
+
+

Notice that the paired interval (the last row) is entirely above zero. The grouped interval (first two rows) contains zero. Thus, acknowledging the pairing explains variation that would otherwise be absorbed into the variation for the group means. As a result, treating the groups as independent results in wider intervals. Even if it didn’t result in a shorter interval, the paired interval would be correct as the groups are not statistically independent!

+
+

ChickWeight data in R

+

Now let’s try an example with actual independent groups. Load in the ChickWeight data in R. We are also going to manipulate the dataset to have a total weight gain variable using dplyr.

+
+
library(datasets); data(ChickWeight); library(reshape2)
+##define weight gain or loss
+wideCW <- dcast(ChickWeight, Diet + Chick ~ Time, value.var = "weight")
+names(wideCW)[-(1 : 2)] <- paste("time", names(wideCW)[-(1 : 2)], sep = "")
+library(dplyr)
+wideCW <- mutate(wideCW,
+  gain = time21 - time0
+)
+
+

Here’s a plot of the data.

+
+Chickweight data over time. +

Chickweight data over time.

+
+

Here’s a plot only of the weight gain for the diets.

+
+Violin plots of chickweight data by weight gain (final minus baseline) by diet. +

Violin plots of chickweight data by weight gain (final minus baseline) by diet.

+
+

Now let’s do a t interval comparing groups 1 and 4. We’ll show the two intervals, one assuming that the variances are equal and one assuming otherwise.

+

Code for t interval of the chickWeight data

+
wideCW14 <- subset(wideCW, Diet %in% c(1, 4))
+rbind(
+  t.test(gain ~ Diet, paired = FALSE, var.equal = TRUE, data = wideCW14)$conf,
+  t.test(gain ~ Diet, paired = FALSE, var.equal = FALSE, data = wideCW14)$conf
+)
+
+
+
[,1]   [,2]
+[1,] -108.1 -14.81
+[2,] -104.7 -18.30
+
+

For the time being, let’s interpret the equal variance interval. Since the interval is entirely below zero it suggest that group 1 had less weight gain than group 4 (at 95% confidence).

+
+
+
+

Unequal variances

+

Watch this video before beginning.

+

Under unequal variances our t interval becomes:

+

\[ +\bar Y - \bar X \pm t_{df} \times \left(\frac{s_x^2}{n_x} + \frac{s_y^2}{n_y}\right)^{1/2} +\]

+

where \(t_{df}\) is the t quantile calculated with degrees of freedom:

+

\[ +df= \frac{\left(S_x^2 / n_x + S_y^2/n_y\right)^2} + {\left(\frac{S_x^2}{n_x}\right)^2 / (n_x - 1) + + \left(\frac{S_y^2}{n_y}\right)^2 / (n_y - 1)} +\]

+

which will be approximately a 95% interval. This works really well. So when in doubt, just assume unequal variances. Also, we present the formula for completeness. In practice, it’s easy to mess up, so make sure to do t.test.

+

Referring back to the previous ChickWeight example, the violin plots suggest that considering unequal variances would be wise. Recall the code is

+
+
> t.test(gain ~ Diet, paired = FALSE, var.equal = FALSE, data = wideCW14)$conf
+[2,] -104.7 -18.30
+
+

This interval is remains entirely below zero. However, it is wider than the equal variance interval.

+
+
+

Summary notes

+
    +
  • The t distribution is useful for small sample size comparisons.
    • +
  • It technically assumes normality, but is robust to this assumption within limits.
    • +
  • The t distribution gives rise to t confidence intervals (and tests, which we will see later)
    • +
+

For other kinds of data, there are preferable small and large sample intervals and tests.

+
    +
  • For binomial data, there’s lots of ways to compare two groups.
    • +
  • Relative risk, risk difference, odds ratio.
    • +
  • Chi-squared tests, normal approximations, exact tests.
    • +
  • For count data, there’s also Chi-squared tests and exact tests.
    • +
  • We’ll leave the discussions for comparing groups of data for binary and count data until covering glms in the regression class.
    • +
  • In addition, Mathematical Biostatistics Boot Camp 2 covers many special cases relevant to biostatistics.
    • +
+
+
+

Exercises

+
    +
  1. For iid Gaussian data, the statistic \(\frac{\bar X - \mu}{s / \sqrt{n}}\) must follow a:
    2. +
+
    +
  • Z distribution
    • +
  • t distribution
    • +
+
    +
  1. Paired differences T confidence intervals are useful when:
    2. +
+
    +
  • Pairs of observations are linked, such as when there is subject level matching or in a study with baseline and follow up measurements on all participants.
    • +
  • When there was randomization of a treatment between two independent groups.
    • +
+
    +
  1. The assumption that the variances are equal for the independent group T interval means that:
    2. +
+
    +
  • The sample variances have to be nearly identical.
    • +
  • The population variances are identical, but the sample variances may be different.
    • +
+
    +
  1. Load the data set mtcars in the datasets R package. Calculate a 95% confidence interval to the nearest MPG for the variable mpg. Watch a video solution and see written solutions.
    2. +
  2. Suppose that standard deviation of 9 paired differences is \(1\). What value would the average difference have to be so that the lower endpoint of a 95% students t confidence interval touches zero? Watch a video solution here and see the text here.
    3. +
  3. An independent group Student’s T interval is used instead of a paired T interval when:
    4. +
+
    +
  • The observations are paired between the groups.
    • +
  • The observations between the groups are naturally assumed to be statistically independent.
    • +
  • As long as you do it correctly, either is fine.
    • +
  • More details are needed to answer this question. watch a discussion of this problem and see the text.
    • +
+
    +
  1. Consider the mtcars dataset. Construct a 95% T interval for MPG comparing 4 to 6 cylinder cars (subtracting in the order of 4 - 6) assume a constant variance. Watch a video solution and see the text.
    2. +
  2. +
  3. If someone put a gun to your head and said “Your confidence interval must contain what it’s estimating or I’ll pull the trigger”, what would be the smart thing to do?
    4. +
+ +
    +
  1. Refer back to comparing MPG for 4 versus 6 cylinders (question 7). What do you conclude?
    2. +
+
    +
  • The interval is above zero, suggesting 6 is better than 4 in the terms of MPG.
    • +
  • The interval is above zero, suggesting 4 is better than 6 in the terms of MPG.
    • +
  • The interval does not tell you anything about the hypothesis test; you have to do the test.
    • +
  • The interval contains 0 suggesting no difference. Watch a video solution and see the text.
    • +
+
    +
  1. Suppose that 18 obese subjects were randomized, 9 each, to a new diet pill and a placebo. Subjects’ body mass indices (BMIs) were measured at a baseline and again after having received the treatment or placebo for four weeks. The average difference from follow-up to the baseline (followup - baseline) was 3 kg/m2 for the treated group and 1 kg/m2 for the placebo group. The corresponding standard deviations of the differences was 1.5 kg/m2 for the treatment group and 1.8 kg/m2 for the placebo group. The study aims to answer whether the change in BMI over the four week period appear to differ between the treated and placebo groups. What is the pooled variance estimate? Watch a video solution here and see the text here.
    2. +
+
+
+
+

Hypothesis testing

+

Hypothesis testing is concerned with making decisions using data.

+
+

Hypothesis testing

+

Watch this video before beginning.

+

To make decisions using data, we need to characterize the kinds of conclusions we can make. Classical hypothesis testing is concerned with deciding between two decisions (things get much harder if there’s more than two). The first, a null hypothesis is specified that represents the status quo. This hypothesis is usually labeled, \(H_0\). This is what we assume by default. The alternative or research hypothesis is what we require evidence to conclude. This hypothesis is usually labeled, \(H_a\) or sometimes \(H_1\) (or some other number other than 0).

+

So to reiterate, the null hypothesis is assumed true and statistical evidence is required to reject it in favor of a research or alternative hypothesis

+
+

Example

+

A respiratory disturbance index (RDI) of more than 30 events / hour, say, is considered evidence of severe sleep disordered breathing (SDB). Suppose that in a sample of 100 overweight subjects with other risk factors for sleep disordered breathing at a sleep clinic, the mean RDI was 32 events / hour with a standard deviation of 10 events / hour.

+

We might want to test the hypothesis that

+

\[H_0 : \mu = 30\]

+

versus the hypothesis

+

\[H_a : \mu > 30\]

+

where \(\mu\) is the population mean RDI. Clearly, somehow we must figure out a way to decide between these hypotheses using the observed data, particularly the sample mean.

+

Before we go through the specifics, let’s set up the central ideas.

+
+
+
+

Types of errors in hypothesis testing

+

The alternative hypotheses are typically of the form of the true mean being \(<\), \(>\) or \(\neq\) to the hypothesized mean, such as \(H_a : \mu > 30\) from our example. The null typically sharply specifies the mean, such as \(H_0 : \mu = 30\) in our example. More complex null hypotheses are possible, but are typically covered in later courses.

+

Note that there are four possible outcomes of our statistical decision process:

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
TruthDecideResult
\(H_0\)\(H_0\)Correctly accept null
\(H_0\)\(H_a\)Type I error
\(H_a\)\(H_a\)Correctly reject null
\(H_a\)\(H_0\)Type II error
————-————–———————–
+

We will perform hypothesis testing by forcing the probability of a Type I error to be small. This approach consequences, which we can discuss with an analogy to court cases.

+
+
+

Discussion relative to court cases

+

Consider a court of law and a criminal case. The null hypothesis is that the defendant is innocent. The rules requires a standard on the available evidence to reject the null hypothesis (and the jury to convict). The standard is specified loosely in this case, such as convict if the defendant appears guilty “Beyond reasonable doubt”. In statistics, we can be mathematically specific about our standard of evidence.

+

Note the consequences of setting a standard. If we set a low standard, say convicting only if there circumstantial or greater evidence, then we would increase the percentage of innocent people convicted (type I errors). However, we would also increase the percentage of guilty people convicted (correctly rejecting the null).

+

If we set a high standard, say the standard of convicting if the jury has “No doubts whatsoever”, then we increase the the percentage of innocent people let free (correctly accepting the null) while we would also increase the percentage of guilty people let free (type II errors).

+
+
+

Building up a standard of evidence

+

Watch this video before beginning.

+

Consider our sleep example again. A reasonable strategy would reject the null hypothesis if the sample mean, \(\bar X\), was larger than some constant, say \(C\). Typically, \(C\) is chosen so that the probability of a Type I error, labeled \(\alpha\), is \(0.05\) (or some other relevant constant) To reiterate, \(\alpha =\) Type I error rate = Probability of rejecting the null hypothesis when, in fact, the null hypothesis is correct

+

Let’s see if we can figure out what \(C\) has to be. The standard error of the mean is \(10 / \sqrt{100} = 1\). Furthermore, under \(H_0\) we know that \(\bar X \sim N(30, 1)\) (at least approximately) via the CLT. We want to chose \(C\) so that:

+

\[P(\bar X > C; H_0)=0.05.\]

+

The 95th percentile of a normal distribution is 1.645 standard deviations from the mean. So, if \(C\) is set 1.645 standard deviations from the mean, we should be set since the probability of getting a sample mean that large is only 5%. The 95th percentile from a \(N(30, 1)\) is:

+

\[ +C = 30 + 1 \times 1.645 = 31.645. +\]

+

So the rule “Reject \(H_0\) when \(\bar X \geq 31.645\)” has the property that the probability of rejection is 5% when \(H_0\) is true.

+

In general, however, we don’t convert \(C\) back to the original scale. Instead, we calculate how many standard errors the observed mean is from the hypothesized mean

+

\[ +Z = \frac{\bar X - \mu_0}{s / \sqrt{n}}. +\]

+

This is called a Z-score. We can compare this statistic to standard normal quantiles.

+

To reiterate, the Z-score is how many standard errors the sample mean is above the hypothesized mean. In our example:

+

\[ +\frac{32 - 30}{10 / \sqrt{100}} = 2 +\]

+

Since 2 is greater than 1.645 we would reject. Setting the rule “We reject if our Z-score is larger than 1.645” controls the Type I error rate at 5%. We could write out a general rule for this alternative hypothesis as reject whenever \(\sqrt{n} (\bar X - \mu_0) / s > Z_{1-\alpha}\) where \(\alpha\) is the desired Type I error rate.

+

Because the Type I error rate was controlled to be small, if we reject we know that one of the following occurred:

+
    +
  1. the null hypothesis is false,
    2. +
  2. we observed an unlikely event in support of the alternative even though the null is true,
    3. +
  3. our modeling assumptions are wrong.
    4. +
+

The third option can be difficult to check and at some level all bets are off depending on how wrong we are about our basic assumptions. So for this course, we speak of our conclusions under the assumption that our modeling choices (such as the iid sampling model) are correct, but do so wide eyed acknowledging the limitations of our approach.

+
+
+

General rules

+

We developed our test for one possible alternatives. Here’s some general rules for the three most important alternatives.

+

Consider the \(Z\) test for \(H_0:\mu = \mu_0\) versus: \(H_1: \mu < \mu_0\), \(H_2: \mu \neq \mu_0\), \(H_3: \mu > \mu_0\). Our test statistic

+

\[TS = \frac{\bar{X} - \mu_0}{S / \sqrt{n}}.\]

+

We reject the null hypothesis when:

+

\[H_1 : ~~~ TS \leq Z_{\alpha} = -Z_{1 - \alpha}\],

+

\[H_2 : ~~~ |TS| \geq Z_{1 - \alpha / 2}\]

+

\[H_3 : ~~~ TS \geq Z_{1 - \alpha}\],

+

respectively.

+
+

Summary notes

+
    +
  • We have fixed \(\alpha\) to be low, so if we reject \(H_0\) (either our model is wrong) or there is a low probability that we have made an error.
    • +
  • We have not fixed the probability of a type II error, \(\beta\); therefore we tend to say “Fail to reject \(H_0\)” rather than accepting \(H_0\).
    • +
  • Statistical significance is no the same as scientific significance.
    • +
  • The region of TS values for which you reject \(H_0\) is called the rejection region.
    • +
  • The \(Z\) test requires the assumptions of the CLT and for \(n\) to be large enough for it to apply.
    • +
  • If \(n\) is small, then a Gosset’s t test is performed exactly in the same way, with the normal quantiles replaced by the appropriate Student’s t quantiles and \(n-1\) df.
    • +
  • The probability of rejecting the null hypothesis when it is false is called power
    • +
  • Power is a used a lot to calculate sample sizes for experiments.
    • +
+
+
+

Example reconsidered

+

Watch this video before beginning.

+

Consider our example again. Suppose that \(n= 16\) (rather than \(100\)). The statistic

+

\[ +\frac{\bar X - 30}{s / \sqrt{16}}, +\]

+

follows a t distribution with 15 df under \(H_0\).

+

Under \(H_0\), the probability that it is larger that the 95th percentile of the t distribution is 5%. The 95th percentile of the T distribution with 15 df is 1.7531 (obtained via r qt(.95, 15)).

+

Assuming that everything but the sample size is the same, our test statistic is now \(\sqrt{16}(32 - 30) / 10 = 0.8\). Since 0.8 is not larger than 1.75, we now fail to reject.

+
+
+
+

Two sided tests

+

In many settings, we would like to reject if the true mean is different than the hypothesized, not just larger or smaller. I other words, we would reject the null hypothesis if in fact the sample mean was much larger or smaller than the hypothesized mean. In our example, we want to test the alternative \(H_a : \mu \neq 30\).

+

We will reject if the test statistic, \(0.8\), is either too large or too small. Then we want the probability of rejecting under the null to be 5%, split equally as 2.5% in the upper tail and 2.5% in the lower tail.

+

Thus we reject if our test statistic is larger than qt(.975, 15) or smaller than qt(.025, 15). This is the same as saying: reject if the absolute value of our statistic is larger than qt(0.975, 15) = 2.1314.

+

In this case, since our test statistic is 0.8, which is smaller than 2.1314, we fail to reject the two sided test (as well as the one sided test).

+

If you fail to reject the one sided test, then you would fail to reject the two sided test. Because of its larger rejection region, two sided tests are the norm (even in settings where a one sided test makes more sense).

+
+
+

T test in R

+

Let’s try the t test on the pairs of fathers and sons in Galton’s data.

+

Example of using the t test in R.

+
> library(UsingR); data(father.son)
+> t.test(father.son$sheight - father.son$fheight)
+
+    One Sample t-test
+
+ data:  father.son$sheight - father.son$fheight
+ t = 11.79, df = 1077, p-value < 2.2e-16
+ alternative hypothesis: true mean is not equal to 0
+ 95 percent confidence interval:
+  0.831 1.163
+ sample estimates:
+ mean of x
+     0.997
+
+
+
+

Connections with confidence intervals

+

Consider testing \(H_0: \mu = \mu_0\) versus \(H_a: \mu \neq \mu_0\). Take the set of all possible values for which you fail to reject \(H_0\), this set is a \((1-\alpha)100\%\) confidence interval for \(\mu\).

+

The same works in reverse; if a \((1-\alpha)100\%\) interval contains \(\mu_0\), then we fail to reject \(H_0\).

+

In other words, two sided tests and confidence intervals agree.

+
+
+

Two group intervals

+

Doing group tests is now straightforward given that we’ve already covered independent group T intervals. Our rejection rules are the same, the only change is how the statistic is calculated. However, the form is familiar:

+

\[ +\frac{\mbox{Estimate} - \mbox{Hypothesized Value}}{\mbox{Standard Error}}. +\]

+

Consider now testing \(H_0 : \mu_1 = \mu_2\). Our statistic is

+

\[ +\frac{\bar X_1 - \bar X_2 - (\mu_1 - \mu_0)}{S_p\sqrt{\frac{1}{n1} + \frac{1}{n_2}}}. +\]

+

For the equal variance case and and

+

\[ +\frac{\bar X_1 - \bar X_2 - (\mu_1 - \mu_0)}{\sqrt{\frac{S_1^2}{n1} + \frac{S_2^2}{n_2}}}. +\]

+

Let’s just go through an example.

+
+

Example chickWeight data

+

Recall that we reformatted this data as follows

+

Reformatting the data.

+
library(datasets); data(ChickWeight); library(reshape2)
+##define weight gain or loss
+wideCW <- dcast(ChickWeight, Diet + Chick ~ Time, value.var = "weight")
+names(wideCW)[-(1 : 2)] <- paste("time", names(wideCW)[-(1 : 2)], sep = "")
+library(dplyr)
+wideCW <- mutate(wideCW,
+  gain = time21 - time0
+)
+
+

Unequal variance T test comparing diets 1 and 4.

+
> wideCW14 <- subset(wideCW, Diet %in% c(1, 4))
+> t.test(gain ~ Diet, paired = FALSE,
++       var.equal = TRUE, data = wideCW14)
+
+    Two Sample t-test
+
+  data:  gain by Diet
+  t = -2.725, df = 23, p-value = 0.01207
+  alternative hypothesis: true difference in means is not equal to 0
+  95 percent confidence interval:
+   -108.15  -14.81
+  sample estimates:
+  mean in group 1 mean in group 4
+            136.2           197.7
+
+
+
+
+

Exact binomial test

+

Recall this problem. Suppose a friend has 8 children, 7 of which are girls and none are twins.

+

Perform the relevant hypothesis test. \(H_0 : p = 0.5\) versus \(H_a : p > 0.5\).

+

What is the relevant rejection region so that the probability of rejecting is (less than) 5%?

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Rejection regionType I error rate
[0 : 8]1
[1 : 8]0.9961
[2 : 8]0.9648
[3 : 8]0.8555
[4 : 8]0.6367
[5 : 8]0.3633
[6 : 8]0.1445
[7 : 8]0.0352
[8 : 8]0.0039
—————–——————-
+

Thus if we reject under 7 or 8 girls, we will have a less than 5% chance of rejecting under the null hypothesis.

+

It’s impossible to get an exact 5% level test for this case due to the discreteness of the binomial. The closest is the rejection region [7 : 8]. Further note that an alpha level lower than 0.0039 is not attainable. For larger sample sizes, we could do a normal approximation.

+

Extended this test to two sided test isn’t obvious. Given a way to do two sided tests, we could take the set of values of \(p_0\) for which we fail to reject to get an exact binomial confidence interval (called the Clopper/Pearson interval, by the way). We’ll cover two sided versions of this test when we cover P-values.

+
+
+

Exercises

+
    +
  1. Which hypothesis is typically assumed to be true in hypothesis testing?
    2. +
+
    +
  • The null.
    • +
  • The alternative.
    • +
+
    +
  1. The type I error rate controls what?
    2. +
  2. Load the data set mtcars in the datasets R package. Assume that the data set mtcars is a random sample. Compute the mean MPG, \(\bar x,\) of this sample. You want to test whether the true MPG is \(\mu_0\) or smaller using a one sided 5% level test. (\(H_0 : \mu = \mu_0\) versus \(H_a : \mu < \mu_0\)). Using that data set and a Z test: Based on the mean MPG of the sample \(\bar x,\) and by using a Z test: what is the smallest value of \(\mu_0\) that you would reject for (to two decimal places)? Watch a video solution here and see the text here.
    3. +
  3. Consider again the mtcars dataset. Use a two group t-test to test the hypothesis that the 4 and 6 cyl cars have the same mpg. Use a two sided test with unequal variances. Do you reject? Watch the video here and see the text here
    4. +
  4. A sample of 100 men yielded an average PSA level of 3.0 with a sd of 1.1. What are the complete set of values that a 5% two sided Z test of \(H_0 : \mu = \mu_0\) would fail to reject the null hypothesis for? Watch the video solution and see the text.
    5. +
  5. You believe the coin that you’re flipping is biased towards heads. You get 55 heads out of 100 flips. Do you reject at the 5% level that the coin is fair? Watch a video solution and see the text.
    6. +
  6. Suppose that in an AB test, one advertising scheme led to an average of 10 purchases per day for a sample of 100 days, while the other led to 11 purchases per day, also for a sample of 100 days. Assuming a common standard deviation of 4 purchases per day. Assuming that the groups are independent and that they days are iid, perform a Z test of equivalence. Do you reject at the 5% level? Watch a video solution and see the text.
    7. +
  7. A confidence interval for the mean contains:
    8. +
+
    +
  • All of the values of the hypothesized mean for which we would fail to reject with \(\alpha = 1 - Conf. Level\).
    • +
  • All of the values of the hypothesized mean for which we would fail to reject with \(2 \alpha = 1 - Conf. Level\).
    • +
  • All of the values of the hypothesized mean for which we would reject with \(\alpha = 1 - Conf. Level\).
    • +
  • All of the values of the hypothesized mean for which we would reject with \(2 \alpha = 1 - Conf. Level\). Watch a video solution and see the text.
    • +
+
    +
  1. In a court of law, all things being equal, if via policy you require a lower standard of evidence to convict people then
    2. +
+ +
+
+
+

P-values

+
+

Introduction to P-values

+

Watch this video before beginning.

+

P-values are the most common measure of statistical significance. Their ubiquity, along with concern over their interpretation and use makes them controversial among statisticians. The following manuscripts are interesting reads about P-values.

+ +
+
+

What is a P-value?

+

The central idea of a P-value is to assume that the null hypothesis is true and calculate how unusual it would be to see data (in the form of a test statistic) as extreme as was seen in favor of the alternative hypothesis. The formal definition is:

+

A P-value is the probability of observing a test statistic as or more extreme in favor of the alternative than was actually obtained, where the probability is calculated assuming that the null hypothesis is true.

+

A P-value then requires a few steps. 1. Decide on a statistic that evaluates support of the null or alternative hypothesis. 2. Decide on a distribution of that statistic under the null hypothesis (null distribution). 3. Calculate the probability of obtaining a statistic as or more extreme as was observed using the distribution in 2.

+

The way to interpret P-values is as follows. If the P-value is small, then either \(H_0\) is true and we have observed a rare event or \(H_0\) is false (or possibly the null model is incorrect).

+

Let’s do a quick example. Suppose that you get a t statistic of 2.5 for 15 degrees of freedom testing \(H_0:\mu = \mu_0\) versus \(H_a : \mu > \mu_0\). What’s the probability of getting a t statistic as large as 2.5?

+

P-value calculation in R.

+
> pt(2.5, 15, lower.tail = FALSE)
+[1] 0.01225
+
+

Therefore, the probability of seeing evidence as extreme or more extreme than that actually obtained under \(H_0\) is 0.0123. So, (assuming our model is correct) either we observed data that was pretty unlikely under the null, or the null hypothesis if false.

+
+
+

The attained significance level

+

Recall in a previous chapter that our test statistic was 2 for \(H_0 : \mu_0 = 30\) versus \(H_a:\mu > 30\) using a normal test (\(n\) was 100). Notice that we rejected the one sided test when \(\alpha = 0.05\), would we reject if \(\alpha = 0.01\), how about 0.001?

+

The smallest value for alpha that you still reject the null hypothesis is called the attained significance level. This is mathematically equivalent, but philosophically a little different from, the P-value. Whereas the P-value is interpreted in the terms of how probabilistically extreme our test statistic is under the null, the attained significance level merely conveys what the smallest level of \(\alpha\) that one could reject at.

+

This equivalence makes P-values very convenient to convey. The reader of the results can perform the test at whatever \(\alpha\) he or she choses. This is especially useful in multiple testing circumstances.

+

Here’s the two rules for performing hypothesis tests with P-values. * If the P-value for a test is less than \(\alpha\) you reject the null hypothesis * For two sided hypothesis test, double the smaller of the two one sided hypothesis test Pvalues

+
+
+

Binomial P-value example

+

Suppose a friend has 8 children, 7 of which are girls and none are twins. If each gender has an independent 50% probability for each birth, what’s the probability of getting 7 or more girls out of 8 births?

+

This calculation is a P-value where the statistic is the number of girls and the null distribution is a fair coin flip for each gender. We want to test \(H_0: p=0.5\) versus \(H_a: p > 0.5\), where \(p\) is the probability of having a girl for each birth.

+

Recall here’s the calculation:

+

Example of a Binomial P-value calculation in R.

+
> pbinom(6, size = 8, prob = 0.5, lower.tail = FALSE)
+[1] 0.03516
+
+

Since our P-value is less than 0.05 we would reject at a 5% error rate. Note, however, if we were doing a two sided test, we would have to double the P-value and thus would then fail to reject.

+
+
+

Poisson example

+

Watch this video before beginning.

+

Suppose that a hospital has an infection rate of 10 infections per 100 person/days at risk (rate of 0.1) during the last monitoring period. Assume that an infection rate of 0.05 is an important benchmark.

+

Given a Poisson model, could the observed rate being larger than 0.05 be attributed to chance? We want to test \(H_0: \lambda = 0.05\) where \(\lambda\) is the rate of infections per person day so that 5 would be the rate per 100 days. Thus we want to know if 9 events per 100 person/days is unusual with respect to a Poisson distribution with a rate of 5 events per 100. Consider \(H_a: \lambda > 0.05\).

+

Poisson P-value calculation.

+
> ppois(9, 5, lower.tail = FALSE)
+[1] 0.03183
+
+

Again, since this P-value is less than 0.05 we reject the null hypothesis. The P-value would be 0.06 for two sided hypothesis (double) and so we would fail to reject in that case.

+
+
+

Exercises

+
    +
  1. P-values are probabilities that are calculated assuming which hypothesis is true?
    2. +
+
    +
  • the alternative
    • +
  • the null
    • +
+
    +
  1. You get a P-value of 0.06. Would you reject for a type I error rate of 0.05?
    2. +
+
    +
  • Yes you would reject the null
    • +
  • No you would not reject the null
    • +
  • It depends on information not given
    • +
+
    +
  1. The proposed procedure for getting a two sided P-value for the exact binomial test considered here is what?
    2. +
+
    +
  • Multiplying the one sided P-value by one half
    • +
  • Doubling the larger of the two one sided P-values
    • +
  • Doubling the smaller of the two one sided P-values
    • +
  • No procedure exists
    • +
+
    +
  1. Consider again the mtcars dataset. Use a two group t-test to test the hypothesis that the 4 and 6 cyl cars have the same mpg. Use a two sided test with unequal variances. Give a P-value. Watch the video here and see the text here
    2. +
  2. You believe the coin that you’re flipping is biased towards heads. You get 55 heads out of 100 flips. Give an exact P-value for the hypothesis that the coin is fair. Watch a video solution and see the text.
    3. +
  3. A web site was monitored for a year and it received 520 hits per day. In the first 30 days in the next year, the site received 15,800 hits. Assuming that web hits are Poisson. Give an exact one sided P-value to the hypothesis that web hits are up this year over last. Do you reject? Watch the video solutions and see the problem text.
    4. +
  4. Suppose that in an AB test, one advertising scheme led to an average of 10 purchases per day for a sample of 100 days, while the other led to 11 purchases per day, also for a sample of 100 days. Assuming a common standard deviation of 4 purchases per day. Assuming that the groups are independent and that they days are iid, perform a Z test of equivalence. Give a P-value for the test? Watch a video solution and see the text.
    5. +
  5. Consider the mtcars data set.
    6. +
+
    +
  • Give the p-value for a t-test comparing MPG for 6 and 8 cylinder cars assuming equal variance, as a proportion to 3 decimal places.
    • +
  • Give the associated P-value for a z test.
    • +
  • Give the common standard deviation estimate for MPG across cylinders to 3 decimal places.
    • +
  • Would the t test reject at the two sided 0.05 level (0 for no 1 for yes)? Watch a video solution and see the text.
    • +
+
+
+
+

Power

+
+

Power

+

Watch this video before beginning. and then watch this video as well.

+

Power is the probability of rejecting the null hypothesis when it is false. Ergo, power (as its name would suggest) is a good thing; you want more power. A type II error (a bad thing, as its name would suggest) is failing to reject the null hypothesis when it’s false; the probability of a type II error is usually called \(\beta\). Note Power \(= 1 - \beta\).

+

Let’s go through an example of calculating power. Consider our previous example involving RDI. \(H_0: \mu = 30\) versus \(H_a: \mu > 30\). Then power is:

+

\[P\left(\frac{\bar X - 30}{s /\sqrt{n}} > t_{1-\alpha,n-1} ~;~ \mu = \mu_a \right).\]

+

Note that this is a function that depends on the specific value of \(\mu_a\)! Further notice that as \(\mu_a\) approaches 30 the power approaches \(\alpha\).

+

Pushing this example further, we reject if

+

\[Z = \frac{\bar X - 30}{\sigma /\sqrt{n}} > z_{1-\alpha}\]

+

Or, equivalently, if

+

\[\bar X > 30 + Z_{1-\alpha} \frac{\sigma}{\sqrt{n}}\]

+

But, note that, under \(H_0 : \bar X \sim N(\mu_0, \sigma^2 / n)\). However, under \(H_a : \bar X \sim N(\mu_a, \sigma^2 / n)\).

+

So for this test we could calculate power with this R code:

+

Power calculation for the sleep example in R

+
alpha = 0.05
+z = qnorm(1 - alpha)
+pnorm(mu0 + z * sigma/sqrt(n), mean = mua, sd = sigma/sqrt(n), lower.tail = FALSE)
+
+

Let’s plug in the specific numbers for our example where: \(\mu_a = 32\), \(\mu_0 = 30\), \(n =16\), \(\sigma = 4\).

+
+
> mu0 = 30
+> mua = 32
+> sigma = 4
+> n = 16
+> z = qnorm(1 - alpha)
+> pnorm(mu0 + z * sigma/sqrt(n), mean = mu0, sd = sigma/sqrt(n), lower.tail = FALSE)
+[1] 0.05
+> pnorm(mu0 + z * sigma/sqrt(n), mean = mua, sd = sigma/sqrt(n), lower.tail = FALSE)
+[1] 0.6388
+
+

When we plug in \(\mu_0\), the value under the null hypothesis, we get that the probability of rejection is 5%, as the test was designed. However, when we plug in a value of 32, we get 64%. Therefore, the probability of rejection is 64% when the true value of \(\mu\) is 32. We could create a curve of the power as a function of \(\mu_a\), as seen below. We also varied the sample size to see how the curve depends on that.

+
+Plot of power as \mu_a varies. +

Plot of power as \(\mu_a\) varies.

+
+

The code below shows how to use manipulate to investigate power as the various inputs change.

+

Code for investigating power.

+
library(manipulate)
+mu0 = 30
+myplot <- function(sigma, mua, n, alpha) {
+    g = ggplot(data.frame(mu = c(27, 36)), aes(x = mu))
+    g = g + stat_function(fun = dnorm, geom = "line", args = list(mean = mu0,
+        sd = sigma/sqrt(n)), size = 2, col = "red")
+    g = g + stat_function(fun = dnorm, geom = "line", args = list(mean = mua,
+        sd = sigma/sqrt(n)), size = 2, col = "blue")
+    xitc = mu0 + qnorm(1 - alpha) * sigma/sqrt(n)
+    g = g + geom_vline(xintercept = xitc, size = 3)
+    g
+}
+manipulate(myplot(sigma, mua, n, alpha), sigma = slider(1, 10, step = 1, initial = 4),
+    mua = slider(30, 35, step = 1, initial = 32), n = slider(1, 50, step = 1,
+        initial = 16), alpha = slider(0.01, 0.1, step = 0.01, initial = 0.05))
+
+
+
+

Question

+

Watch this video before beginning.

+

When testing \(H_a : \mu > \mu_0\), notice if power is \(1 - \beta\), then

+

\[1 - \beta = P\left(\bar X > \mu_0 + z_{1-\alpha} \frac{\sigma}{\sqrt{n}} ; \mu = \mu_a \right)\]

+

where \(\bar X \sim N(\mu_a, \sigma^2 / n)\). The unknowns in the equation are: \(\mu_a\), \(\sigma\), \(n\), \(\beta\) and the knowns are: \(\mu_0\), \(\alpha\). Specify any 3 of the unknowns and you can solve for the remainder.

+
+
+

Notes

+
    +
  • The calculation for \(H_a:\mu < \mu_0\) is similar
    • +
  • For \(H_a: \mu \neq \mu_0\) calculate the one sided power using \(\alpha / 2\) (this is only approximately right, it excludes the probability of getting a large TS in the opposite direction of the truth)
    • +
  • Power goes up as \(\alpha\) gets larger
    • +
  • Power of a one sided test is greater than the power of the associated two sided test
    • +
  • Power goes up as \(\mu_1\) gets further away from \(\mu_0\)
    • +
  • Power goes up as \(n\) goes up
    • +
  • Power doesn’t need \(\mu_a\), \(\sigma\) and \(n\), instead only \(\frac{\sqrt{n}(\mu_a - \mu_0)}{\sigma}\)
    • +
  • The quantity \(\frac{\mu_a - \mu_0}{\sigma}\) is called the effect size, the difference in the means in standard deviation units.
    • +
  • Being unit free, it has some hope of interpretability across settings.
    • +
+
+
+

T-test power

+

Watch this before beginning.

+

Consider calculating power for a Gosset’s t test for our example where we now assume that \(n=16\). The power is

+

\[ +P\left(\frac{\bar X - \mu_0}{S /\sqrt{n}} > t_{1-\alpha, n-1} ~;~ \mu = \mu_a \right). +\]

+

Calculating this requires the so-called non-central t distribution. However, fortunately for us, the R function power.t.test does this very well. Omit (exactly) any one of the arguments and it solves for it. Our t-test power again only relies on the effect size.

+

Let’s do our example trying different options.

+

Example of using ‘power.t.test’ in R.

+
# omitting the power and getting a power estimate
+> power.t.test(n = 16, delta = 2/4, sd = 1, type = "one.sample", alt = "one.sided")$power
+[1] 0.604
+# illustrating that it depends only on the effect size, delta/sd
+> power.t.test(n = 16, delta = 2, sd = 4, type = "one.sample", alt = "one.sided")$power
+[1] 0.604
+# same thing again
+> power.t.test(n = 16, delta = 100, sd = 200, type = "one.sample", alt = "one.sided")$power
+[1] 0.604
+# specifying the power and getting n
+> power.t.test(power = 0.8, delta = 2/4, sd = 1, type = "one.sample", alt = "one.sided")$n
+[1] 26.14
+# again illustrating that the effect size is all that matters
+power.t.test(power = 0.8, delta = 2, sd = 4, type = "one.sample", alt = "one.sided")$n
+[1] 26.14
+# again illustrating that the effect size is all that matters
+> power.t.test(power = 0.8, delta = 100, sd = 200, type = "one.sample", alt = "one.sided")$n
+[1] 26.14
+
+
+
+

Exercises

+
    +
  1. Power is a probability calculation assuming which is true:
    2. +
+
    +
  • The null hypothesis
    • +
  • The alternative hypothesis
    • +
  • Both the null and alternative
    • +
+
    +
  1. As your sample size gets bigger, all else equal, what do you think would happen to power?
    2. +
+
    +
  • It would get larger
    • +
  • It would get smaller
    • +
  • It would stay the same
    • +
  • It cannot be determined from the information given
    • +
+
    +
  1. What happens to power as \(\mu_a\) gets further from \(\mu_0\)?
    2. +
+
    +
  • Power decreases
    • +
  • Power increases
    • +
  • Power stays the same
    • +
  • Power oscillates
    • +
+
    +
  1. In the context of calculating power, the effect size is?
    2. +
+
    +
  • The null mean divided by the standard deviation
    • +
  • The alternative mean divided by the standard error
    • +
  • The difference between the null and alternative means divided by the standard deviation
    • +
  • The standard error divided by the null mean
    • +
+
    +
  1. Recall this problem “Suppose that in an AB test, one advertising scheme led to an average of 10 purchases per day for a sample of 100 days, while the other led to 11 purchases per day, also for a sample of 100 days. Assuming a common standard deviation of 4 purchases per day.” Assuming that 10 purchases per day is a benchmark null value, that days are iid and that the standard deviation is 4 purchases for day. Suppose that you plan on sampling 100 days. What would be the power for a one sided 5% Z mean test that purchases per day have increased under the alternative of \(\mu = 11\) purchase per day? Watch a video solution and see the text.
    2. +
  2. Researchers would like to conduct a study of healthy adults to detect a four year mean brain volume loss of .01 mm3. Assume that the standard deviation of four year volume loss in this population is .04 mm3. What is necessary sample size for the study for a 5% one sided test versus a null hypothesis of no volume loss to achieve 80% power? Watch the video solution and see the text.
    3. +
+
+
+
+

The bootstrap and resampling

+
+

The bootstrap

+

Watch this video before beginning.

+

The bootstrap is a tremendously useful tool for constructing confidence intervals and calculating standard errors for difficult statistics. For a classic example, how would one derive a confidence interval for the median? The bootstrap procedure follows from the so called bootstrap principle

+

To illustrate the bootstrap principle, imagine a die roll. The image below shows the mass function of a die roll on the left. On the right we show the empirical distribution obtained by repeatedly averaging 50 independent die rolls. By this simulation, without any mathematics, we have a good idea of what the distribution of averages of 50 die rolls looks like.

+
+Image of true die roll distribution (left) and simulation of averages of 50 die rolls +

Image of true die roll distribution (left) and simulation of averages of 50 die rolls

+
+

Now imagine a case where we didn’t know whether or not the die was fair. We have a sample of size 50 and we’d like to investigate the distribution of the average of 50 die rolls where we’re not allowed to roll the die anymore. This is more like a real data analysis, we only get one sample from the population.

+
+Image of empirical die roll distribution (left) and simulates of averages of 50 die rolls from this distribution +

Image of empirical die roll distribution (left) and simulates of averages of 50 die rolls from this distribution

+
+

The bootstrap principle is to use the empirical mass function of the data to perform the simulation, rather than the true distribution. That is, we simulate averages of 50 samples from the histogram that we observe. With enough data, the empirical distribution should be a good estimate of the true distribution and this should result in a good approximation of the sampling distribution.

+

That’s the bootstrap principle: investigate the sampling distribution of a statistic by simulating repeated realizations from the observed distribution.

+

If we could simulate from the true distribution, then we would know the exact sampling distribution of our statistic (if we ran our computer long enough.) However, since we only get to sample from that distribution once, we have to be content with using the empirical distribution. This is the clever idea of the bootstrap.

+
+

Example Galton’s fathers and sons dataset

+

Watch this video before beginning.

+

The code below creates resamples via draws of size n with replacement with the original data of the son’s heights from Galton’s data and plots a histogram of the median of each resampled dataset.

+

Bootstrapping example

+
library(UsingR)
+data(father.son)
+x <- father.son$sheight
+n <- length(x)
+B <- 10000
+resamples <- matrix(sample(x,
+                           n * B,
+                           replace = TRUE),
+                    B, n)
+resampledMedians <- apply(resamples, 1, median)
+
+
+Bootstrapping example for the median of sons’ heights from Galton’s +

Bootstrapping example for the median of sons’ heights from Galton’s

+
+
+
+
+

The bootstrap principle

+

Watch this video before beginning.

+

Suppose that I have a statistic that estimates some population parameter, but I don’t know its sampling distribution. The bootstrap principle suggests using the distribution defined by the data to approximate its sampling distribution

+
+

The bootstrap in practice

+

In practice, the bootstrap principle is always carried out using simulation. We will cover only a few aspects of bootstrap resampling. The general procedure follows by first simulating complete data sets from the observed data with replacement. This is approximately drawing from the sampling distribution of that statistic, at least as far as the data is able to approximate the true population distribution. Calculate the statistic for each simulated data set Use the simulated statistics to either define a confidence interval or take the standard deviation to calculate a standard error.

+
+
+

Nonparametric bootstrap algorithm example

+

Bootstrap procedure for calculating confidence interval for the median from a data set of \(n\) observations:

+
    +
  1. Sample \(n\) observations with replacement from the observed data resulting in one simulated complete data set.
    2. +
  2. Take the median of the simulated data set
    3. +
  3. Repeat these two steps \(B\) times, resulting in \(B\) simulated medians
    4. +

  4. These medians are approximately drawn from the sampling distribution of the median of \(n\) observations; therefore we can:

    +
      +
    • Draw a histogram of them
      • +
    • Calculate their standard deviation to estimate the standard error of the median
      • +
    • Take the \(2.5^{th}\) and \(97.5^{th}\) percentiles as a confidence interval for the median
      • +
    5. +
+

For the general bootstrap, just replace the median with whatever statistic that you’re investigating.

+
+
+

Example code

+

Consider our father/son data from before. Here is the relevant code for doing the resampling.

+
+
B <- 10000
+resamples <- matrix(sample(x,
+                           n * B,
+                           replace = TRUE),
+                    B, n)
+medians <- apply(resamples, 1, median)
+
+

And here is some results.

+
+
> sd(medians)
+[1] 0.08424
+
+

Thus, 0.084 estimates the standard error of the median for this data set. It did this by repeatedly sampling medians from the observed distribution and taking the standard deviation of the resulting collection of medians. Taking the 2.5 and 97.5 percentiles gives us a bootstrap 95% confidence interval for the median.

+
+
> quantile(medians, c(.025, .975))
+ 2.5% 97.5%
+68.43 68.81
+
+

We also always want to plot a histogram or density estimate of our simulated statistic.

+
+
g = ggplot(data.frame(medians = medians), aes(x = medians))
+g = g + geom_histogram(color = "black", fill = "lightblue", binwidth = 0.05)
+g
+
+
+Bootstrapping example for the median of sons’ heights from Galton’s +

Bootstrapping example for the median of sons’ heights from Galton’s

+
+
+
+

Summary notes on the bootstrap

+
    +
  • The bootstrap is non-parametric.
    • +
  • Better percentile bootstrap confidence intervals correct for bias.
    • +
  • There are lots of variations on bootstrap procedures; the book An Introduction to the Bootstrap by Efron and Tibshirani is a great place to start for both bootstrap and jackknife information.
    • +
+
+
+
+

Group comparisons via permutation tests

+

Watch this video before beginning.

+

Consider comparing two independent groups. Example, comparing sprays B and C.

+
+
data(InsectSprays)
+g = ggplot(InsectSprays, aes(spray, count, fill = spray))
+g = g + geom_boxplot()
+g
+
+
+Comparison of insect spray. +

Comparison of insect spray.

+
+
+
+

Permutation tests

+

Consider comparing means between the group. However, let’s use the calculate the distribution of our statistic under a null hypothesis that the labels are irrelevant (exchangeable). This is a handy way to create a null distribution for our test statistic by simply permuting the labels over and over and seeing how extreme our data are with respect to this permuted distribution.

+

The procedure would be as follows:

+
    +
  1. consider a data from with count and spray,
    2. +
  2. permute the spray (group) labels,
    3. +
  3. recalculate the statistic (such as the difference in means),
    4. +
  4. calculate the percentage of simulations where the simulated statistic was more extreme (toward the alternative) than the observed.
    5. +
+
+
+

Variations on permutation testing

+

This idea of exchangeability of the group labels is so powerful, that it’s been reinvented several times in statistic. The table below gives three famous tests that are obtained by permuting group labels.

+

|———-|———————|———————| |Data type | Statistic | Test name | |———-|———————|———————| |Ranks | rank sum | rank sum test | |Binary | hypergeometric prob | Fisher’s exact test | |Raw data | | permutation test | |———-|———————|———————|

+

Also, so-called randomization tests are exactly permutation tests, with a different motivation. In that case, think of the permutation test as replicating the random assignment over and over.

+

For matched or paired data, it wouldn’t make sense to randomize the group labels, since that would break the association between the pairs. Instead, one can randomize the signs of the pairs. For data that has been replaced by ranks, you might of heard of this test before as the the signed rank test.

+

Again we won’t cover more complex examples, but it should be said that permutation strategies work for regression as well by permuting a regressor of interest (though this needs to be done with care). These tests work very well in massively multivariate settings.

+
+
+

Permutation test B v C

+

Let’s create some code for our example. Our statistic will be the difference in the means in each group.

+

Permutation distribution for the insect sprays dataset.

+
subdata <- InsectSprays[InsectSprays$spray %in% c("B", "C"),]
+y <- subdata$count
+group <- as.character(subdata$spray)
+testStat <- function(w, g) mean(w[g == "B"]) - mean(w[g == "C"])
+observedStat <- testStat(y, group)
+permutations <- sapply(1 : 10000, function(i) testStat(y, sample(group)))
+
+

Let’s look at some of the results. First let’s look at the observed statistic.

+
+
> observedStat
+[1] 13.25
+
+

Now let’s see what proportion of times we got a simulated statistic larger than our observed statistic.

+
+
mean(permutations > observedStat)
+[1] 0
+
+

Since this is 0, our estimate of the P-value is 0 (i.e. we strongly reject the NULL). It’s useful to look at a histogram of permuted statistics with a vertical line drawn at the observed test statistic for reference.

+
+Permutation distribution from the insectsprays dataset +

Permutation distribution from the insectsprays dataset

+
+
+
+

Exercises

+
    +
  1. The bootstrap uses what to estimate the sampling distribution of a statistic?
    2. +
+
    +
  • The true population distribution
    • +
  • The empirical distribution that puts probability 1/n for each observed data point
    • +
+
    +
  1. When performing the bootstrap via Monte Carlo resampling for a data set of size n which is true? Assume that you’re going to do 10,000 bootstrap resamples?
    2. +
+
    +
  • You sample n complete data sets of size 10,000 with replacement
    • +
  • You sample 10,000 complete data sets of size n without replacement
    • +
  • You sample 10,000 complete data sets of size n with replacement
    • +
  • You sample n complete data sets of size 10,000 without replacement
    • +
+
    +
  1. Permutation tests do what?
    2. +
+
    +
  • Creates a null distribution for a hypothesis test by permuting a predictor variable.
    • +
  • Creates a null distribution by resampling from the response with replacement.
    • +
  • Creates an alternative distribution by permuting group labels.
    • +
  • Creates confidence intervals by resampling with replacement.
    • +
+
+
+ + + + +
+ + + + + + + + diff --git a/manuscript/LittleInferenceBook.pdf b/manuscript/LittleInferenceBook.pdf new file mode 100644 index 0000000000000000000000000000000000000000..47170384f0902ee1749a90f8bb3d55e1d9beec54 GIT binary patch literal 620923 zcma%ibBr%tn{C_1Y1{s_yHDG8pSEq=I&IswZQHhO+kM}e%-no8Gs*q-AC;A-lB%7Z z%35o${S=wJuqYiPJu3{^?BdWW3=<)M(AK~LhMOCPLD}8Tn2WEhMPbNdbTUtC5Ae?o$vkY=%N>o zxXSuVHR8k%%?J%5H{dVdYus76J-h}`>3nU%tXYL0B*!ibQLQYz7Kl-n-7nQwXW7!+ zDm!G=lfRHbFlycb;R4%GsgC`gTENqw<>8`WVMvKMg#1-=L1-}I(6VNhECxeMn#lSy z12!;QBa{H;5lRs-`>+s9yC|r7!AQk4;8kSNKJHWCA*Q>-TM|-mq?R~6APvk6%0Wv* zgM7370DpCY`=WrnCqAIS zc>J?uFdn9tyGu0oZwgx9?VpWkDYiosQ$~$h^cXu|x%7hDCt2)Jk(x+7`uSYnR;<2j|D%`5R3>i5Fwn2k8D-3N;?L?Q>0Rou~Sa!d(QX zTkS~E=D`oqUe?gk*7njs%-iOos<#FiI@?H{I81YqAL-ef@h#x2Z_mxUo2*&Bq@P0g zW5GMor4v>df1W)PB#L!mvBw0l9{d`j-q%`@{3?GFmo}v;F_Yr{T>sWV>d|y)p=wKz zam`t8xJs5bjc1GM?0TcVmW29a8>GqG@Ikowi?)ht`A#5~E@_{R(N4)ISGn~Z;l=Z; zXT#U66u0WNvHLi%cyJ+6wB9Qnd0Bd}ILk5lV`R8%@48{5WVAN9{ngRpdyibK(*ZYI zlP>A@>&_vOyD_n&$4#^bwjpThQbH(U?|Q0JzSVJ_-rX%{>|OigGX1umo54t2>0`;f z?NNr__n?W?(QX69stcmW?9i(E52wNKHl%^YGDDRh>ZzR9PVB>jEz@?h7GA0rp7HY` za#=F#4QN?gZhk!}a<`kuuGC&)3BTOUJ?TOtqk8Zlpx`ZdRNd*i8bNTGB82;xv!2R{ zM*^&wB<$kPLy9QYxJ0i?WM`lcArRu#M=Nisl%MAK0D+OisO9ow=w-x-x|fr4tz=}| zu;)1|7PocR(mf}T^wi7M#bz^EL9{q#RkpEL7Dd8Y-CbI=<900W3G49?ebCePy}ZM| z)W?v}<0cpt{3J_MZ}_lM#Or z{iUsY(yrqigBXrqAKz#1!4LJj7b}rB)&vn9K%B~WCw^2RtS3dr%ijWaH9}z(vpu`? zuV7g%1ssr^dF;b4Mt0+TxLNccF#HE;&lht}pRmqC=p-wJbUPeBIj|UkKR}vN8qYAs zHb(zN+5e#bkOL+F^S>z!94!AL46L00zA$LUXtE;t%+%cC(4j0ozdF$V5t+>o0E05~ zPw6&FXpCAVognWh`pLE>u972vwx6E2X+gYue`R`@@;bOPu4) zYQ!$E8u85{7J9u%df42dN|`l{hY7N^{1a+oKv%Zv?ghDoWfL=Q76lW8bp;5FWC{iZ z;Zpl|bTTcADF0Sg!yz*($*pWC4PvQ_PP7lx3}D(OA3~J|qrD*Fymvwz(-uw%IW*FX z3^_ytz{bi*ZYE*Uw1fDZ=Bxr$Q1TC7YQV zCyxj~r@-G-7g-++8v`==U8pfBupfR1%pQejgbhC zA5o4hNPG-pp(+7Q7f5H`gj>(l$0D4rTY;Wz`+|ENMJ8Ry=7OA(&W?$sr^4XJ=p&&| zw-MZ(aUMJ+Yzwr7s_$iS%3C5s^w*oU)C!kNfZtCq{%)X`Ha+DnpxF z#>vj9XS(L{S}gwA+up3efGq@W*{YHH$IKtpIx1woDM?*?i#171Rjd=`_>M8rt;x-S zS9Y;MK@ayocfuPJHb9OwZ_dfQDf6b_$Wmo(kgjxVW!_!TMh1IEH^7~ z=4l0nnk}Yq!O0$mke!}MT^;cgFJF1@O_Tg3DRp^@J1?BUp0CX2Ur>EXqg&|ed*5zs zKD~1^REM3Tuu5rvCPXEOW~76EjJA>r(HU`J4<6 z8?x6=Vww}b9m$4M_8OBSToi8Iacx7q7YvHvf=6_2JI}HuVctnL>T`bNlGU@5AY$yr zteQc|z1F{mhq=~mt&Y_0g@!$e?4K3tl01YjX^j%`gVa&Q-=+OxvBc|cj3cvKl^!#- z#!+9Iv#D(~wzngul0ILgEz*uowjQx*%7`sF=mB>nb zQqXCcSKUdNlRZ}(M`Jt2|J2KYJRmbVZf%>Ti?;idObQl{_-h^(bp}P+S>UQs-@u5r zpjRB1d5ub97$&*q%#L%s_UznhK?`*Jz;6W6^sA6^Wb%l6(l=U%>*M9volQ}CD(Dd*Vt|K z;=4usVd^hA%FuZ_-w}S-uaO|gUAy4YSle!t_E?d!g{dPD5+`&2a%q&b{h;5Yalc5} z&~fM_P_xy|TAyaHxeByXWr>-=GR1ln{Vv!)E$ZR6ua59?pvqE${fJWq-tqPV)p!r+ z)(p`e&G^=+|II`L#%EV+gXe9+l&7P-jn^Cny5BU}-=EFPuX>MV|J!1rdy0`O$UpIc zic$I3QjX9J2Zh`=SuWRYnVR(vFqkNb+5glC7Pf!Yi2vRkFtaiKyBct%DWiJ8hPL%U zQ;+kJNv_sRDmB5>%h{+c9g^uX=R#yaHb>N~xONQq>5eBmf$a-I(F3w3_8lbNeA)JX z;TfYiC^&dv!2;qKCzOReI3tvU7D%T#QJvDM=5yNJt1ofbh9B&<% zeOq=pWnw*HnqMx=x?mzXtJ!}KiMxEP4USUZs>@c1>`X##rQe6Klt`l^lSv2j?Hb<~ zjzO+bkBEjG1B^k2{1~v2Os2F$8=qw5X6+-6z53;@Pw?Wzmtdu^fks%eag7%!?i-d% z8#`Jc4#orJcCa|pFF_pfE(&6UM)RmI`#|WH9sN-@dzgvk%NA}zCth(uZ;O4HCQR zZ0$ojwLn;R*5ClLt;+7B?wP$Dwc{0zd{c$_lj%6r!s9ZWmBQqyQ%!2joFduW&{ja= zjU`-36=iwiz57_*xfzORupek*JJ74mtw*5niS1gcHJ%onN8``i*2$4ga}9X6uxZpb z65>&?yXEQh!8o0%FMsQ)BI?Hr&^j7L-?u}(d!%ca*1v_sETDYJdr$iFmzK_poly{3 zDL`9&6A8xP>P$WHRqv->DhRKneqCXtafwvJxv8fOW&Hj|5#q@lonaij<~#}(;PfV{4`eh_izhIXHM8jN0h%&^(9Zi!6 zu(>#5o$_m7@eLrKmRlkkbu}m<2E47XKD&rr^Le5w853Y1J>@GYSq0lRJG)oI-QN5FadVLv%iyQ|{`-ZRwA z$U~6&>g2ra*I2PZQ3L@zW}&}PHjSLyhha&;6t#J>cO~c#c{NuD?Vy%Tr2$4YV{V&Z zdZVlWU{*V*E6OjTd{?R?THQRG&dL%kj+S&K5F4lf%FPo31Md~1WclyX7hVu8kpSh~ z&%rCH_^c`gL3<4tPlvWk2kUDgW5kl^qc-N*w#Y(y(ggTUza%Kqdx{(drAJJ4OI5gy zneoMX+BhxrlMa+?7l6-ba<;nqFU$VyQ#YZ* zIQDW!4GChX`0i=o2OCWNBP(ef!3&|3?j!`Jh*5i+SchHne;4p#{% zAR-Ff=pp>%uOww-TR-cTVJhh8&XU;{%o6^UK1}#vEPP|MR{WkeXD2q^cf{*;q_h9Y z(5y`V8t?pHhUWaYUHX;ggu})^4E;gVNYg|pgoFrEA$#!K{I@wp)zVr16$mjhj#8MD zzb{Tc`&O{uaV5Hx0xqZ7+T%*g1gy5sc{2FFj zK+dgRD_gV-@6Un`5!n5t15g%^mn8@8Y>HGOP)4bBGJaoS9_r?yu)ziamp{=g&Ao>8 zS-J#AiqBaE4bKgRj76N0?9Lrd3=Acyq*@ZZx~nWk#z7>h{Av_PFyQWg0MR0blA<-Q z{X@&aJKn{J7D=;=VvnqM)7~J`b^+t^eldKqWzo5?MieRJqXPjQ*jC%9D;Vk8X2gkF z9&nTSGxn9>T%zf(;w@7-oWf0J)IG&TuqE>86n6O-Y~oI4RKal_1j+&GX8t)4%a7tK zNrA#6-hW}%`9&ID1cd?|u&IQ_Fd_dQCnqCTTGAa$kcM_8N#fEbl?*tHEky<9ntK*- zAkpLx@8HxNFuhj_ebB~p`3Bv0c&QKoj96%(aH^J71d5Ye_4Yy|2&KN3^N7BhP@;5k zeESd(uAIg&U>T}j?+x}Ye9)Cain?p7g-OiQFNat(0@05*++Vt{e$V+XygdOe&-$?# z9>JXO`l+S~_>F*CtW~jQLq#iI_+D++zc&ixC@(h_WJNbbr{emp+6I*=!D6 zkDeQ$EROM+H=p>*S(DRtYF#JIKG$+OBGeLaR%k*cu(UiHfA42;olmlSNN=wemOsVN zhNqme^(%Z&gAjII`oYnpzucJerDpdb-pI4E2Q+0igAa&c!U`L;M&OxzSNVBeD zc|b>;hCRE9KV-J0S3w+Gyki`2474NWZb&?RaJ~qWdwcrT45h7d_*ymx z4@XZG1jq%fJ7p%`cAwN^Z3$4!V4F9}52jJ1YkW@#8eLNVUHs0iPy`4)Q(4*ina_MV2-W%tJLPryEGF3R7 zpWxP7L&wg5%>jt>t`0(&W5*91-N?< zk<(qAUuoFczW23i4Mrsi4=O={HbD{Yt0wMK=YvUT+X&cWH*o6OFOHCkdlOT04MK5w z>Bxva#!QB$1KHQ_l)U}nJK=@hs@%~9KlB@F9^2L!LP(MTZk#1?ix*7n!{s2n6ceLp zDDVXa)Nf1tPabChu>5;z3@0P&|4n0<*f`k!T_@h8DHDU+2H*8kGi%Qd*(4sG2!;t; z*=3JH2prjjn6b*su`Q}+#Ed!_33{?~QkQPyOY6b{r5r;4$5~Z@HoOr3X|YM1x-2~OliBMp)2P@2Ch(`;>d_)K? z51w7+SXg@w54Qd6G{n-tphq3G;hjBME!cMKCjtA!l(wkMth`f|BOe0n=Ta(%P7brx7k`P67=N86I= z-ZkV*=GNhqQ~k+xC3zBtp#0fr3Vj50xy$u*3kISV8zM|gS1HqpuWm8*lVUim-6b*o z^&mI_K}s3U5(hpWr$m;%MB@C?7Wz#4_jZN(`GzKo)YfKjRnfK=n5LX}?GMW?c5u#Z z?`}ZJ?YGb%0(v=I3QR+hzD=nPDs9L%7jC(0OR1-_zem$R5ezq^>1|oR4bmf+@53GD zZFxQXsGnw%iry7)5B1iLS8v_8*|+6iLzK_vefNME0VarK@mbVhc+>M?f>Y3H#H`&H zJLAuM`E02212}KMZ%#7-f=DH=5MDJsg1_uYAC}Yx0jqlEbfffDp01DEO$ZlXXIp01!$1>3_2(^&zUP~@=jXG-+g-~@j_&8lLj=Xn zcFECeQ*zJCKnTMhZ%>7b=LD11Qbl}}t50_&9Ux0nja%euf5L-edCzKt`4e$ORg*ElS z*Uzt=o8}9?o{smEz1P?GS1xj1;N z>ENSzI9Kl+p9Oz+$DJY#ma+W@49|(#aub7gU3m2p(E_d5!q~b7Jq`(4&wIuL+M?|m zQEgjzthKIy$a>E_^D*(9mx{cLCB0IAJprZ}ylKmMW?RlZd{(P2>_T;9$HH{REkU{o z;wvB4(#n+rag`k14nhMxgmU%wL0sX9r&(Ygr1EMp&|RSQvC%kGp5>8BHaRSQw!in3 z(Q}dzZ0V;&7f*mw{b)W78$-}2b#s%Lc5e4xF~=SGDgJq&V1)^0x@E|!a6&cUmo+iS z)AD62=lFgz??n*)69n9`+i8TL-2S!q)YG};3k3)b!K%Vg4L8UXI6c0(0)`dPQch1~ zp{!8`6@8ciP~0BG6RFGC&BZmylvsYTyb5W1@5~R;h8EFz|WT?AVscv04td zJnV74b%WTB#qa4rlV>%8W}t{r!L9fZRZW%q*)+~&F<1aV(XiuCQBVD_Swi-tjyokk z*{(DW%2DzT2IE~mr&n=}NMrZcK2j-hMp(@WW%0bkuY{*PVo?4@&tF21?Y=+}@)<_X z>U<|sBe*e*0;9$7j@592E~#sVJc7S2)-a4x0UHu1{K(D!ye%X`FZ|^QKWI{gA(Yi^ zZxYwA3xDapnLlvvtU!QaD!-xK81f|+rj$PAkqG^1F}1L+y*nC}8VIKr^1N+%DTNz& zi^9CJ+EodEs}JH)fQTJMi-hJvIO-_@)^Hkf8R)sR7r$CX2%J&DkcadTg4K!8tc4V& ziKU=5BeChCD^X#|)pFx-KjP0O82F%Y6rL!!(l=rEVpF6=Ptrfrqd=_AW;rJ_!*cY= zDAhN1pJaOCPG{hbxr9B3a~_5(-FPosR^{??&ywd*XqwXW%n**KdQlWiyHOCVGOp1D z9K@`Jr~XqpKNLSep$)?Foy;Hl=b-@^u?8Hg^~|=sOX%rqhTe^_Z2a)j_UKSXhM-(X z=Y#mxA@*gd`&~by__Z>{Om$)#`dXdU_$i0T3d{-xbTjy1gp^TPq$109-ax6Q{9RQH z4~$?ZoI@B#zvH=Lp^j6+@GrI+Ny5@+=rZBgKskg>qzt^h-+~sK>O94br4xaK2FWQa zed8qfOj~^3FLvS`YoU6hRXaR>CKzgbA9zdbCqS?Ja2L8eug;#9wu-lvpzI{yq^mf~ zEP3cN7n_-B0!yN2Y^ry7`SzL??$!|H7Vd`52DVmMrc|XfnS7@PsZxGY!W#5a*GJe= zi+@Qg!4{W+FDwR_TlBFs>7{D`WO!05@b5iFFNkVFI{yWTj^Ys626ZCe|0^(cpXSQZ z3{NgdGP7*Nt`gbEznTl?hH*!`r#$@IVFt*;mM(!BiK7vO^|L?`r}X85T{W?SYj<Utt=SQYRXGVhn^4I3clJH(Rm*FW3mho9~ zQ;hwN)9@Q%b*Z{t)^ZV>@5cTa`b7iZngI|T!*Fh<@;B!vF;PlJ~bT5SRj9(wmBM~w3zCdvukcLp7b&PGr z>QgVbYoMu{JRAepG>j)jK)S{dR$f3!6P=*$VN?*Yp>cez=?4d#DN-G^DMMpgOaoA5 zst}G$DbIQgj3-SB`8cZr-!XXE<3OVLgqUSyG!WgaF2~4;3{>li?SZV9hDsMD4x$xc18K9vdsJ_TobNt3LXVpc|YG4H5l=OA8^c25~yB z*|JgsTog|IkV0}wZ_6x^8w754*IuKyLC1n}L4W3q8yCX%tZ}QupN&#^xRuxKVo$(} zUVD>~E*dZ?D3jri46mV!Ibt!+gRkpvAeY-8;8!PRk_t1#avatk$Bxm$HqMrd599Gg z!i$8p;Sj5jjaXr}cj@9o`lW8rigN_d?;QV-+M4-B#y>`o-w6Chk8)P*QE?rU@oCj3 z|D2l8ANwAoNHb?^ci3K~Q#l9-|A0P(pL+jKyUWVL_Ahq#zhVm}j(?6M|8xBRyH2ju z)ng8uk$qljF!|KsCr44r9_(GOy6OSHIubAiG9jY}&xkZoxPF_s++QdSQJ6e0-6$ zdjYfyiH%>yEvOYu^;Zf{Cbe7qs4t>>QYgKU#E4Ei*S%;9rJ{yh8dy&N1he4B9ikOg zE{5m?O3Hdin9tDPoHgm%xExoZov$~_6RJ+TV7RRuB|m&uFj#&Yzo!}w;-L#4HXMiJ zrfp?fAjcJOe`c`#Hrqc{H`5^)I2C3OmafZ@&U}M}zP=x`EbsZR zAX%wTb_l(=r5NbvL2`AnV-Is^1|Y+kUMJVunR*p)Q+qlJ17(4NJuF=bCj+GCH3x5{%Kf_tUe?ZI4tunAD-^a5UrmtAg4O6`FI{lK4&o!XvZR?in(2Hww@)YX1b*kp_9 zYZ+}%`3L*?>DYOnk@j*X_VNh00RCQEXfI5ft#qUY<%=0AC?T?x(yEZjfuXG?#GjYt zpEnh=d6-11dLpVAcO85(#7R31aOa}jaoSD#jnZ440uXT6-e`|S6P3RFo{W)Uuo7$* zl|RH|Lrqp$1~`^3;jMB*xF(99cUqPH5iA7SxQKoW)rP;lk=_b1%m2 zvEKX^E*2I)Yx-guu0=acmFcaYF|IDRmH5&NZVdH0+flv5up-&1$p!L+tp|~YBvLE1?fIu}Zl!ojRgJ)z-lB&bnA=P4dI;$?nb-VsLog}C8!=)p5 zlS&kQhk_jHD=Kj?)MMKY*hAECCSZKbig1TKdn%P4(3LeT7$JIe=aOvR%S=?KjPO!w z%00Gze*~R+YZ_K=#}WUq2MEwUK9--}m9OZndgf%<4>!|+zG2=|gmXV1n+RfbzmK`J zMk7hEs^;7f!st|bvmnF+`Z0iyCa;#fPdas38{|wP7~_edQEj6T2P>2eAc*v-3++M% za33FA#vTtbu?A^ceP(W8eOc@;RQc7A=wv`q;@l@oBiFunsS#(d`Rl`n926uMa7nN_ z_M{m$<|i;bCgoLgerv}^0LIoFDoAg@l()9IxJ}V1iAvX_9IL~7b2WHMB0!{}`C=~g zG`uZ68`!f95FavwPxMp={b)W|dyCGF%*yT0-FFkab*e5AT(%AnU)PlY&yA8~NO8mx z%YUd-Ca5^sAJD1Q?K=fgcP6?-$ViWjHILiWRevki0F9ZBZ1yt7>hNu?YHf??!e3@sF-v47D221~i@V?y3KTRnqhyuvQYCCm|L9~_ILVqr`>1w4P zwkVuG_g=kCnFV*S?*Y}+LBaDwS@7Xi@a&v3D}PEDwre;N_})4f=MwfdoQ=yrL9Z2Qle>U&q!X!ZdsfRrvWJ83z83Kr^9V;WAa z1uZ4dtbru~Uo8woS^}^N(n?f3d8net|J201{&jS?mdQ1dI)q~5NILGzE4c2Qd(x>n z@GN?fd*o2OY%j!kLgthP<)ial&E9DdkMoPoL70@)XI?Pnd&-B_bZ~@Tb^n z(n3kl`s-OgSea?~?6A>AvCVq4&RS=l>NVi6MaS_2hBRc1ZNjuRD8qumeTsR^aTGy- z#pCEn{DuI_cf1kLZsH}CB|ezhy8LEXu+(c5d=@Na2qqt%PwqgO`EBoDG^72chZdXI z>wsvi<{hzUNb6^#xWt-orQ23O7HGZ6MsJjQRMeAom!_UO22BrJV}e{suRNG<0xa)S z`)KkuA>zaskv-fu<669TRaLa%>`#a~w%n=rsytArc4Uy0dE0#xf6laecL{Ftifra+8q8j0hic zXOZtrJiGXWY*Ut3mU8eYp;)KA0f(n$iKF-DcQPB-=i;)IArDD+n!`=^3$*)IA@Je= z>xv@WNvvC@WWN?dnBmAGH!*&XFmBe2xdK}3-OXu$D4P0gwO(lhUg1m9RjQ8`n!aN3 zhEOo%0{aZRIPvrV=~&B$JTmMid2*={vCweKR`?+U&q~f|on&R{*ezCdLMagCog8!4 zDEH(zY818!8mP;Z$wKN^B%_^vrl^@Z^k#d(tJxFmwdpo-@cIp@w^rM+_3&4j=O+Qc=%qS`B(|8a6s`C+*5&U{W7}$3;0mXV=EAZyU9|zx{JVe4J*B z?;Quh_P~&yb^SrgXx&dNuHd9VxwqmwSUUpJ2Sqt|VuDDUr5VTOWtEqI zw$AOjPm(t=o+(xU-^I&m0rOF@_Nk(?XvgMigPUlQTJOgr7q;KP8+tAOn}=ro*U0$4 zJv7@t)qm4NS7A*=AF$o?>h4YC8|P}qmAcXgsuKoI(u06yKpfXLKsFCHrzMLZk~Nh) zJZTP5E7or?gOd6q?iYwJe;oO4_q^?OiL;qwvx^Y1y~JclT9vbeB~-3&x@P={_Q*KQ z9~5e?+1}Kbhd$nxR$^<2EXaCG)_cisJ^Q}=yzhBy*rc~oiG;ask=e&BW^l!gsE@sP zZ!W8omblQIn<1(isk2-?i5Hk|Q3Vcgd#rAUQ7gs_Q?C#F{jJEw1Lv_i^twqNg<2=E zn*uU#za1*0u413Gl<@)DMMmtrKys{(E?>0fCW=X02_0C5d&YI;zGoISQ1*w!HcjmP zdNG%FlkL)MJ3a5}L?Oc;m`uU&IPAo$fMg8!1Rue6VHB=aM=WL;cNz$?U<5L?<*99_iy*GfKdIQd7vG&r695ReIIMeI)aF9>Ji(2~C0suCS% zE_7Yv6w3W}s?&R45;yQ&H@wVVsd;e(JcRgDO1fKMVGz~sT;eU1-jYIb6AmUO6RJ9? zs9LtUfIki_fnKNR+RlVSXUcpiKy&fJ=M!})(FFi=c%6s8x;qtpxQm2)ZHcA&BX(jnh+J$Z=MD7QO_7cFA2X)_y{Q*ez+FPR0Y*X^Ex zCP>}XZsF!lj;my&s+LA1xJ0Jc`1Jy5|O zs(x#*_OsTqYj^ts4unJI(2YaK8<@c~oG8T=Ia+of!Rp*o3sas1z2^LKe$k%l=`nH| z$omZ;Q)G0|MqN;o5PU}TxlVB3^TKM?zwdXB#B6jB%M>hZZxY_Ku4X_35b$2JGspJ> zNZ|aG<*d9t$zVwB8{*aCjCboW&k#Z1*hREz#a$q*;$`W3a3(gqKCaR>E<^{e6V(^I9)JHk;?EzUM;QfXMag9TBS=W*9 zoLKR&n^!Z|JnCF0b+!0X0|Ja>v$OP2yX$ITlxobnOG$BA9v=i~RFp7>Ku=)8a$~6A ziet-*F9$A91|jE6jjowYK$MB%$4MH+0q47&8X3}tQIiDmlTV_m0h`b1KU=4oacJKm zA0ivp6xGFFEEa+vTi5}gTfT{I`+<&6ICle6M4-sc>j6;VS9Lki^x99ks}V3;VBsPN7gPLwR`c2N;6_@d$t*ETlBACH6)A zW#jPV_joW69tE=K(EG@7O?yW#L|Wa!k9P_O(Zd1OTpP$!`QimRV-R6OS^8Ag3VB_t zo}HQG@w~jHAnrVYQ4a&9nBkM`58v1`3W;cQC+hJiXCoLn_V@72$0$|x(_Qp1r_l;p zI>1UGY9U~=C!`ay>@77Y>7A8C87A=hi!NZ(A#TZ^zX^k~&Hk0iMb|&CXhnb_T(8+&YQX<*KEA2g*A?5!Ua9SA~jH zd%GQpKRS!z>HoX`Qg;0px}_Wq!RZ2H;qzZZN}Y4NiF|MXC-Mv9690*4#Nmj_jXS-( zD2Nsi!DJI6IdM2uf(}_iQ|LT7eMVCdf3~u5_$bEKNSgL3O~}jZgfDMx$zMAN)@mtB z2iXlIBw9gPUghZU0ye%ObFkkmx`VEy4O;LdUsM#LceEvLp*ie71*a%&i{A+TQA*?i*G}BB;GPcZM0$a4U z<_I|8NN12pWf(%XVb`3_Z|#VK?3%{tnpUsz4(fx5S+Se}!HpydQG=)6eyf~vMqf6? z5-^chAtVEEz7NkDD_I(&1wXQQW*i z=Zxs5*%xOHm3-iM69q3e6cT5EF%jPjqb5T1vx7@aiq)1&|EK4>c2@@G*mn$_nfA5N zbj`y50rf`Fv%F+X#I`(lTluO;)lIB^Kb3O7pTcqFliTCDF+4*$*7OIxEKvntP(`kv z2Xkj4?|hxiL%wcZAI@f9423_f1}rw5qe{{zuwaBDj$G9UJoise!K)1F^ey2 zdRk*irC}LG>v@<>y&)#AlgTRCI5zBXc??7rvJ6P12c&iW>SMF;nt=d5C`{X zRQAnopW5RPh!7ioDm^ax$KDqQv~ z1O!8BBG?lr2PRumYSfWDRODv3E8#!$FaH>Rsyd%cnr*u)aaL^|RaE;PZ|L#n91iN^ z|2F=>m`v>K!$o_tJ>32F)k}cUzIJSY(stQ_KlM^FuS=L%*47>|@6Y@BoME);$@-}= zSUOSHJn7YAkvaH zRzdM_gZbI3?EezUtQqGRavJ-xu5~i&{?$#75=Q%J+2o*ME({&ur%@@6O$qeCv4<}b z&1LJ;?)%_NffHx^zeK)l|4XMDGc(J--AkOLAr*5ljNJX9dTIibnz!b)fH5%b(U1{{ zY6J-g;DD60i25xQ$3(!D?CX8LG(|4fz|{vr6aQ6tUQ^y(o}Gi={dVn*Yn=#FK^wu2 zI3j~=3xobwLD32yA;T9Ek)o?hE}Yiw^4w&1cW?-k=v*um~UBx}~ zfcr%u6>aI5S9X9jrR}?kE+IEV&n_Nppe!RDl<&o3ho|{iZmQ-$0{=2yV%)NyqQ#`O zOKySX707n;zB6iq-|=P`j>PwiPI7*xnhg4p;OTAtbfdwbXKpd7#Ur%dS{c_wy}M0S zM2(D5s-Tq$rosWWkFTlhzBW8uCL2RPE-@p7`SINybE*yB-WeU-IRb?S{gZd8Rs*~O zuV!;lP-p?H|H;t06|aSq$E;x%SaiNaNBT_PM?6=qXp_cSQZSxs!yMpZg>eHvZ;{|@AVKX&{ zpLi{q!VOO;FBRG+v8bbpH(W@Lh{COGOgmr3GVyMq0D|xGVKDwK$wi(K4_YB4SYPex z41(w{V19Vd>}ls8qoUFKTpMz0;AD{=5ql#rMTAbnnu|NyT$Ed2Y}76Sc#?Za%n}m? z5@aSrq6F2uq|#b2E;3eM;lun$&TB6Eh~3H;x8ag@a1wy$_o`)_uK!*dtS^{%6ZfK0 zsK7?<*=(&_zelQ`UJWACbR@;~f{RJ4jtXK=_IVu6R4-j2@+W{LfDTI#sy7%B2z}Mz zB?xC_2f@-VZV@Ikf?DmoczBDK3;cH}+L&V4nRQ{S5YvfNSIUCEX>4`uKfBZ?$Y)lS z!5Fah5b=q^_#j=2g-D=2BSQJE$o!;yegKE9_upq=! zt&3v`K<5FE^ns*a?TvO9^VB6UR70%~&;7(~R$rv_2ctpZ~6Hfh_uGI~_BS1qn}kvp+(8e{?fZ!g7Zw z1ID&%D_)29^&p?vmlZ~%XOMbSXoKy!e=R7et5fCeog}`q#wzSN^>n}DpUMgb$n}QZ zA~G>?#zqq(piHgXf`y{pl-ItSj+e4_Z#{!Xb>$%Avu!2YVohh5Nm-Vm4>q^ z2Ht&G#>N0bM}an-@+!)K(#p%;n>B5BNrR9;N#f-OA&)&;To4=UJNXzNS}@Z>1|AFR zSui9P_#7GaZ_;ZrvK7_Jfhy-p26QJ~5wJxZf7_}wfZ`d7mU&vxAEMq@qa@IB$pFWe zDPp^vDLWzcZ{_mPemJdV1Nuc>MY=~wa?2+SaL>p~2>s_#0*h;V6J2*dh%GzV1l&Tc zOWX1{v@`fk#t+q~?Ic4X7FS-^K|{mM$+yY($U(AR<*h3Iv|sj6cL5^66w7KrNK%ZD zaO9fc_kS4*Os?~fRcqh&JOrFvY8U*Q6ZZN7R=nV|!Aa3Fla|W4*{zah1RIsEr1mBt+4m0B$@#{W|lk~QdOc85;v!!H_yO-L*^`8*%ykFXgWoM{?9>HN@ zsePvVzsYt2b&Ss+RUQt)PN?oo^Na=d5Qug7$t66KB9St&eVu$qSbHTOm;5{?ops`F3r7n z`Jgy0NR{_Xb|=HQo>}r6vT`HyuLHk|v4wK0OfM%Ua99hftw8?^!yYQGS@G~AoV2L;=Re(KTA=dw<{CimFSIWDahe9 zAVxNX_UR&*xtS>hsr>mmcdpogBN3^Q>H_vl{>0AYU<6_zja9-R5{W7^_SgNuqBeC^{y+Hc;b?FBu@Qe&2Um^ee zsUS8X`T<@LBsU&F4PkYyK#;iA7~_#gWR(&sVxp%6^4iG;N@jv|U|*-I63sf)@6P`# zQ~fZ`{$L6!>(CskJ8-CWInc(g$i@h)1(n{8n`4f6&%?B|-Ord-YmHpJIB8eiC=orM zC85!yR)WLVKPIXgF)*dQ?`Cg#5fpJ>V;!d=U_0$scnnKGuROG-e8MkD0$FpWe z()u{C8f;cIsh3fRFpBtxR%M)vop}6?pbjl$ii3sL>LO-Wj{fXPysABN$qKA0JXdcf z>P%a5=hgy@qT$)ltHvM91>-Fg?nlt@ai;2qmezpUcqojmCv7@<2SkXs)u&+K5rE+Fa7aui`DTP3asL=SfAt zT)otIZp-l=`_!godKUr&-S*`vg0@y&E3sMF#{}U-a|#vFgB=5ALG2dk8B@~gKt_z) zSG=u1*~l>hSES3|a>xBvadDOR@zAjElfu!`R_UDu10uY8Mz8lNp%XYXtA!y|475d* z-7Z;aL*5I})yt&23EMKijJ3w}ld=t&Tz*yO$g7A3epK%vDrcNoXNxQ~#1Ka9O&cP8N4q#fIK(y@&@wr$(!*tTukwr!go z+ji1Pra!#znwjJB}!(GknQA?~5)j~e9hkzRx! zRjY}kl^U8({ox%sp44RK>0LeOWm-6 zN`Bva0{h^USciIqZ#`u?Yvcg8HJBl?uS`~*_9%{_0*(e0Wr7Z}O?Zn&!#!|0$2vi|U9%Jg=KGnvr0L53&r(vBYCz?_%Q zlzuHaLnE-k*pO{>q`(YpW;zJX(V4AY)!3m?M`D>`9CJiz-w~BV3VYjI?at<(=?Xj%=drVnu zm&nPL;i8V{?h`rXCAz8dK9bv|9g2J!p8`#LMJoATSU4POq6ui!2!%w>0HAhvF12o7 zoJ^v@?!4_uF<6fQGp6AE?ik+}baS}t=Mv^Orr?~A@R}{r7rI=tj8sYw?i7A;O{;Ds zDD-XpJ`CA6-k#&w)gymM@^#=yKv^?d12MrDhIdcnPNU{pHL6p+r+1EDMtWN@7;kub zL_fv+z5}J>i|3X=}>4PyYz+YIJ!UFgYoliclASIDpH2|WC22B z#Cnq)q)bzBQTv+Rc|gS=uX=7UtF?Nin#kN|>CyhdM*O*#mtC=){ueXstThR)sOa5Oi0W6BrUXJ+l4S#;^61^e&o-6s0{cI?uzq2G&Pyo*ct!Z z&wv(9?YPZR^nV96^I22|Y5QLfXv4N!Hc^FwP5#Kk9_cA!i_2)@`|(}d-ygY~(GoRA z6y;yUvp%!W+^5kyTYVoVx~bKRthQ6k+S|8kF0GYIq;e0QE`7RRC_X=gyf)EGwqKWD zyS*N1YE|sERn>znmKbF&J)JX=fI1T&oI1NH(uf;GPoJz)kLI?@!_@8lw(zZTu|u(@ zd(_h1bHP{K4MRLX@lQH^E%YX7yMT65-R)ABtrKkhk6+SBmrpAmol@=#_&b9Xx65Y9 z(#u01?AFBGF^lFn@{C&IFRU&CRo5Q&9MwzX1RX{0J<4OsWrxI2(~}$6~6Esab~Q4m>i0ZJ1TZBfZ+;u z9e+GztD)Dl*(w=m7C&7;A`ZW4{7fpN)c!^15vg!2}PQzBhX zyP^LLv5Snsk*9}-M{yR+}fj<0|UkR{4#L9k6gEQ zAcz~)Gv=ZxYQ?x0zjKH{gv;i5S}q~z8wPc0@v+w*@Ca|X4L$3adkcvYO5m`D8lgT^ zxU*Yio_bXQgK)!uYa_Q*yA^&-miB<_;4yiY(!Wv)rR0wvFfI0*Tx(ifRP%mLrY z)tb@T2`+WNr|2!{?yq^_xSRRbDGf$8Yg#rCv_%exB+ueu_%y;t+XS3+6a=qLi`L)+ z1t{r>X#6@F`mt#GQGUVBiR_kH;Y&v6BU-rk1bKOz49>dln9Jr}?W;wk)fSO4qijTCy!ZdD|Fg zWq8{*{<4QBE$uC-MrE9U;5TWFOaqJ7AoBpiT!fEP;d zvWT_-Kj-Y}ef>tcPOVq+B7>T9vOhaLud#I=uG&%w4tcA zOpJg3Ya;QrT)QErINd=P_;2z1G(hNO+4{6Mq=!nwSg~-V7z_^uaj;^FYZ!DKVS9Mf zN}_w!NIG9>Tws)jYA50%l!y<=tn?Rn=sZIt^@n+#PkRPe0r~iYOn$y`SlSJ1e$xv` z`lbZ%n{wnTzDdJewdg!N9|5gIENc91ob0aI7Lp2(N!TQO4uY3v}5=owIY+z#X6YcS!A}y zbsg2;5~-;Sl;rJJkH80v43I=8&33G{WXQl|c<`%}mR5Y_F|Rm-0+Azj4-`VPAOxnPQNAj*_Os}T2%8_vt0iNq) zrtIP=v2~m2L7t-n9k#G|(@uxx&8+n@wR_JGoZ$w_lKXToqW8oMGv~wo&b7-Nt50jm zX@Y`j=>NiuH%+k}us}(iR&r`8Ndd<5qhEoj#XM7m+)U2Ctmh`%`2)1Scz!d?CcBY* ze^G{3{$=Jfc^mJ6y#1N)ErJ2Y;irTdAk2{G`Sb7#KfSt|)6q8IWEvX0Pnii{u5X@f z*-KNqp3<8yzHb_m5r!K|IdJMUKS;H{s~QK7Z{L1nPfU);Y=4tKUlIVDGOtFm_)8C$ z5MFT#_{Bd)$5B>%(TIAsQN+NK8p`G@McYpF9N{RV(9o!-8m}u4Fw55k&%Oy|XMxg)_K= z9IuSZmXH%Qioy#YDeoNjb|=V*wQWnRGKN$#LFs=7s+>=k>eYo3ayWR+53UzzKJa4F zpn|mcViqe6Cyq!)I5n4Oye>>QfW)0PT6#D{NfliGH8|T8PkBBN@6)Oo`H~^}4K%{7 zF$UNJZOE?ozCUbw8re?&aE5vOb@e4Y0H$(ugdfP@;?nwE%E#8bvYIzymS+0+_V5+ud-4^?gufs-!|~KxK-utK(!D zTx7NMsUpCS?18~<>@K*`l&HbjUGz1MP5{KLB&x5}{)3bhy~zo2lH$eaqrYdFUM!@@ zkDO7!PXhBB$aTyPZ4Q5nOet%V9k)vZnYXIX9a7?vbhc!eOqK6$!UWWh6!3*#6m}i! zis%5j8CPrt#rPT~D6R0ZhGP{|@vniT@4k`w_5M1MMi;>vRaPzE_uG^+xEAflAlMq$ z3lH`GU6-(Mu(18Ng0cq3+U}_J-pe=S^A`ZS1^xb) z5{v>`8Y@*T9YWZ!1s~~G{$@#!P&1ACa_-#MbIv~8v(|N|91N&lA~7c-!s&$FP4C-< zmtpN>GwY(x_O`yMrKXPQ9Yt+*cO$Vhi)r1>i#k7rUk$(R6tP3trM6~dabaF9m(3<` ztLXtBzYu7t^D{x|1oQd%5QAp!I}Bv5WxjP*DIbrFYvM$!>l#QI%ms`8SnZdFPjyzO z_}Lt5yITssYb;kw9s?AhIc}`<;_^jc_uHo$s$hfttJVna%b~BX^MgM;u`$Q2%nBmapOCo?1?#Hf$^2?rp99Rxae7 zVP@jARxU1F7Wr(u9$3>JLqo=VjwpZpb$L_)@y42hyayZjJKuBZ27fY=J`quXt}zk1 zHDRy!gU#T={Ig!eOW)6C%ZB@6%ME#w49}bbKKX<^rIJ(%c%jZRP$r~Bz3pF~r=;$t z{YD9N&NC$dH+gkA*eaL99t?WHKT`F zsH~E8SX>O1N+=iA2HTDrE+?|%D2w2P8wWsewzyaE$e}_Zq5_by;NR~{h3LNga`wdr zA7)^}zKN5?DiOx*0Ll3fSTeXKP)MJY$tc5U5hhoL?+egdzS*CT`)aVlIqCT9wv5gF z4Pp75kMj3w8qG?NFY43XPj|)<)O>~*4zla~68JEqwme6^Xp&xhV-_~&L{kvp8;6r* z7cPjjqqwy8ITpGZRLclGF9<(4Ddr7tamou+)0srZx>QZYlw@g<(PQlDpl{90PT~uZ6hm#CdB%=#F~4tsg;b}X%O zG6bvls@wg+nA|_XK&leZXy5S|48g(1e9aMz~_masuVH@ zbKx1cTEq!o$s)wj|3x}tsz61$!8k`ZW*5J*1GRRw$q{WC{iQQd^1b{`6pPu-F$y83 z!Kob*Y&v52HHj-QMU3#kJrF|Y3a!n7JBu7do8}j)0(tLP{|U^)87>7Xecx{NP$rC8q9Q}1OW^vRk+?fTq90Nz&7V1R8jki^vFH0j8Ki+1O zBJ-29W|3xf*P{S>7cC1{ZuRPM3u5s_@l$CTe5^BH(>EvQ8Zm}*7wszhJJcn3?OnLi z4|`iDm3LH3n=O9t_7Fcn2+=u|BrSS7#EKdkE$00q^N8b3;{=sfRqqJc%mVsv^{oi^ z%X17_)_60jumVh04Jj@)5d49lb zMDyE8L5@dtV1#^iqX?O7q~50yjCXGmsbg!rb@+y^0|V_Vts=cmaMNfdprQ;ULI(u;wD+R}-&-`SCuR`;ngqE}4S4)r3EU_Zvo~+!l;H$&Nil4^TmI`XOY!!3Cx2N;r1+wrDV_H zwhQ$mTdB5G2(lCK z2QUQ%vWrIQ02)(0Ni|@xVt0Ji5(%=4eYQ%y8i~^JBW(B7(du~LO39N;106bzRac6@ zW-)jpC~@t{zdS=WRt>b+7JwnV@T%&64Ak12fN3laXh$vo>`=l|h14W;B+&|;IO#^V z_5$?NzX)58WH*Vf;nfooE%x~jlrw&oYZk{GZyU!vpK$@K|AxY0Nxge zcAgh}%}<-RoiA_oQTR%#^NyRHypCf&uG^(A-HMd3V?^4tG)Y+Hq4$;NIxRk>v#mQy zMP_$+>OvM*2|5S@yZGLSQD;*KQNF|xiuR&KvezE4jFeH$K_GQx%Gn$Q!bc4v1CneO zc7X6PlTq^3E2G~1=Jk)!u})H2i!BA^81Twpee*l7OdXC-H{a{U467$Mt~WyXE^2M~ zm%M3}1{)<-TQvD!t(95Wm9{^i-Q)M*~ z3_WRaQy&`(g8h9d`%rxR|46MaIdUyJ7B0m`$l;>qIK*nILHusF*v}O3+I4Nf3sX)| z&FviqD>=48XQfOG8nuGR2C9eU5c&i8?I^%bL%_V!A!|C|F}uQCpp^>!Y(;(BEF-}- zXw)KxrO+|59FdtP&!>V2+tRNhtnh29Ab32ss8EVlTCzcg93vu%!IZJ-%KVl!GN}yj zZwgop%ZAmoNmcrUnxPl%6mc8V51|@LS)_yC37UERO8UU4Iq4E9CIuI)R0O&=*}>nI zAZ2RRRK(a_*Y;VGwN2qpC6IwIQl%r~iu8+2Byn=td^uQL>2ko?XJuKd&xAD+*?LEH zy4{b^JUI(+ZE8-P$G}=5_DD(-o1J<#ctU_17XX&O&J*Sj9BQuzu&O}|(5`U=wGJZP zvFNGR36dGc`z65GpM~5H-yF_fY~5uxJIHwd1N|6B^DJ1`vBne{j$Y4W@6BQgt&Uha z02U5+IUYxJI#*lo>Pf-p~u`mi%3wgzXsD*`JA-({4)Y8l;Kgx*)gADf%0 zsbFc9FFNuC*miufdvkm@tEb7zFjXuw# zU%=4gLB(Iealijjrgdbt%H*F{9VV_VIe13B#G6`m**xIK*#`VV=MVdP?^AtUwy3u!sufXu3YgDH{UYR`ACxT za&wj9mTto0ETh7{`ahQQji0c1LR*Cyz}yZo>mWekb;N%^lYJ3?y={tD@7(CBL)sx> zKAH_p{aH%R_iu+rLt1F9dfU&SuoBAozTc}Px*RAtEwfQY-N6|+32qRLPE&mhXKWRH zIlidsAof6{aa+bqC~I@l)_Tyr%RPvyddnDytqXNPq6)9w50(h974g12{nyHhS8%r4 z0%iN_`(;?Srbl}7oLA=6|?!t+5IMja0*dz@s*e6nGYsPn8bej<~ce{#ar4S ze1LX2wPN@_DW(jZ5maNv-K%nUPp&`4fo|>MDRj;bTGyN%Hf)EOoA>D}>v`_u)dck> zM$uzHp&JL#^ORMA`{VG?Nep%OfcHr%#YWMqxk z0}lFwNOrk}2ybXDoss>Y;7z?zg;DtNVqk~u8sScnTmuNPl)EF=o zvZYEQU!|rX!)g+xm@k6S7Jwwl8N!FC?^v3(7%;in;9BJ3VkSYw48%qkOi!3jUsjw3 z6g4!Oll#lWjDm+Mq7mV=LorLBF@fzh1OQEIAyK2(GC1P+VD%8l@ zRS2V!G>la*JpsW24mJdRcRW0&ZS0Z2KNRP!udv`<>V4s?rnttk6)efQVz~l1p&3oi zO$NoB&#Jo()WO?z(x_s0~&%#LheV&XG-@ZJW(Jxu|bg_3c|*LQW&N} zN@2qw2Sts54yc$_QZP!3;v}cR`m+LXgkf>Jio^MccX~`rY6`C1E&XrJ?==@!S0A}!>eBDfZBcK zRoj(My?1F`t~6y?a;UXEHU2*)^g2mv(@>@`jJind6 zz8PdAR(0pyltY3rqMCi7NxWa4kHHp5ARd%*t`KSduvsZ#8$8&|j&x>@4iU5iCF7Gr zwz){ioY>mO9dX5!0n$n6;hT8)dX53wppJ0#VM-W|5wu*js?!Pk9P|iv9-ZugN$+r> zG<8lY!E+Vv(>vOt(tSoFw|2Ws>&hB5a;2WE_;7sF4)Pa@Sr|SbVZyD{Gwwq4P+}hJ z{JMDr&8PUbQbz=Di?^s>>XkXIvW~Kuasaj$ z=f(Q$6Y?eEb5nW(`iuYSO2_oZS#5Z)eU9c^qjviHTm$_sL_F4Q8ufHO-x{WMUces z6=BQS=KuVw?1A&nZNRsAbnVYeWz%_}Ax^ZHd%$rH?Qfkx2LLN7z?9I1 zMU_be!seumpU8$e5s8qYD~97E9op|Vfk0=K4fPk2KxJn+4;$)KRtp@cL%%Q1hEHa6 zZ)on{1?7jMX+>y=57`eUDMW{|Koi=gOLn8hxwrgXG4r^U>k!o@xIuKin6$*G{?@M^5hk6NC|J~L#KxLIl z%}nB6i62(@RaSEdlF`(jz)!KnTV%vkH@>LM zZCxrtnTC&=(%a z%M)(VUaT6TKFxw{qFoo$Mj6&wPD%dS8WU!gmeqdt3i^7kK^;`A9S5Ze3!fH#X9{fz z#0Axt&iXlN%!C`aXCI!pFtMa-Z@mwt4RRoT)(EKPUX&#zNW*NJ9g~~%lPnVIWCy?qv!~Om0?r|4t(WhCHI;iPL z8`f^ILa}?|NIV9|OZCThpgJy;>Qo9@^2m7{wafX+Lff5huT4okmGqCcB+plPNu3QV zNxfkw$`g%LdWln6$Ho+fJ1=K>C^#QBXt`P!U9gpmJ`=E9I zpW`|81y9)VkU|y`%>Ii0gr6e^l&6v7-y;w(Gm5(`h>EorMMCNT1l@grGx*^;dZ4g6 zW2~+qz&Ye_9zA*p50hj6lG0+(G^dLT)YxDIr?=Mkt4x1I?hymtDnX7J zn!j)_jTOX}I$gTl;^VEE2-VIUdR^X6{lJ!wo#3f9JnnNkJGYA;p|0{LTk~Dq&i$p_ zr_Jxp*(-)P7fEM<0g}t?MAY9jW8)H~j*~-QYB;u%3w+P|^kq?z8bdRBYk)%1uHEUx zi)2*_k}Qp5c_-Xe(~#Kt|Psca!m`U7+N{P%(2P1SJU&Shi zhnhcQI#DYhlLZ2mxhqS-B#X%Y!R--_gO7S%S_odhSL|HPpGv8PnzYxh_3Qw8VQ=(p zkvhZ%>h|2fv;fi)yLC6V2=aBC03S@(*H?9!nnEpFt-g>?v1VV$(?a>a>G6YPS&)hS zy`A_$4jbop^pi91PbKeI2EageXDb`+3lFVoDCXp$0qH#z9Sp2E*}v~ zT>D7~Pf@rokHDK&5B~d2i0f09&YTBnD;{j*zKI+8$m5*3$?AlDHIPFe#ktFvm)>rH z5w0eUWQg^|ozkagnA|&~DKf4slvDR*o?KjV=x-wbCP+m4=(LYKw?`EpsFfm^qLw7Y zbfyoHc5y5l0ifczeevBpBbE`y>RfTCPlgmG4w1c$e!e~5u)>hEnEz8LvNQjO6B`z` z{~p0OqV@kXYtqvUmdC4!FU)}WNqZCf5kS0vNbXJoicVx`s#3YQ%y_?_E27X=injYh zLHlGLDHBJG@byzp=3evu?oTy$FEr5apMKwy&p1~n_f{m5(pO!be7ipiKa~yj-TXiV zzsq7@QdcF^M?O;bcoZGzV9JyPOM!2yI%KAB056mY6@wC|42M+OqLjLuz@R&(jCHET z?h%GH$6f*8zxHKQO&2C)*2jw~njq0PU|*+8+RR>?!P2R>Y~_Ve2wufZC9_XqaO@p( zfAFY^`&!U?zdpKhwx#h;O+8)CXKw5k8zWWqgh-+uMVOAiCSJHv9zdk_Z983QuXTGW zw=3+Dh&MK9qAn--$9|9;dsSOz;13&TZ9xhNU1uu~JKr+QuX!yHRQ46^SzLfLm%mLp z20=P7oc`F`?jOJ zRqS>vX@C4LQh&s$3fE{3@w^8|FMTFGdnkZZ`|0(h{I)o0W2zRrSW~V&dRWy|pFdib z!}P832&RQMTUb2%F<*R#NVxlw1_n-B{SNLS_j+SP-$giNjEp!s6gh}0go@4vNb(HV z>i(EJmh5cN)+&-$d%=fSjK!XM^JEr<eLxA^r1O832Yyqt2nM%5a5nkx%0#I^|Crw`3v%4w^kOL2DsjK`GG` zv0Xwdf1oyeAmDYbDrTh!Y9s9X@PRzm%s)<>z~al;X|s-(4)Fynwc4E>J- z3VATiT`12T?j62(J6tgCf;LF>Zc) z*fyU~VkARKlTS^`IEE()pmAy`6>6e!zf=^CI?sT}PxYwzra_zDw>xi{P8gPCBvW5s!jS{A zV`KT+zJ|;CUh~f6q&;xa@)I@Nk!3;ab*zkniDeG~~pkR_caoQ7Pt~2ze z=yBznxZvDbk{ee*Vnim~{nNyI=(ryQr(5It6DSs5S**kZp>mk2g4Ee`zVqHOFVN7Z z6KoOhfsI)PT_eDBDPgvqy%Ar{2n6}lgtcl~=9u3$=*&Nz^=ts$M0WT>0p=aqh9Jz# zNv}GFlf>hwb0E58k?$hgv)Pb zaog_TflsFc6$*hhshExwX4mfJbBdUkCk8LI8}SQ~l@$0Y8?)di5ENuZx!SQ1b{_mk zqd6@KQRm?H=&61=Zn^2<-|V8O#{+cl>F{M(b{M5C-QnGz|7b z9I8_+@WRBr@7CIBg4MwH{@7dAF+&qxp4OHa-2cRPVM5`D_7C^$=r%iezMBn&{Vc*? z6g*P*YL*5D%QoFHI-Qz~LGljeXI90TlSmqy31+dba+ zqETnt36jx+guV1LF&v4aStBHnC{PY?r+v)oH}D3(W@VOE#A5h$H&-~^D5@zTfo7Q@ zDF2!=r3nb?g1FS*>SeN^f=FcS3b+*m{9tM_0&%lYesd7caNm11U1>eqfeaGE${g(j z69$7I3c&5dY5f+0o|&iMx^-IHXnKM)Fyw%##xZbsXaZND&2ivP=JZ01z+kh?)FziG z1uA>wcMQBt(v|`CTa9_qfYAU5Jg!*h!`Ce@ylsG^7kYf7&=|D~z~|EX6+}F>O5;5F zyia&6SLNU{aWSA+jfYJp!CZ>^bjx3?GIoP}E(H|LB6W4;1r0H5FV+9a2{O#p-?yJ; z_0n0@uFY`^%PhnuQIWWkPd~(-c#dVXVm$ZW4U*ndoWB;lM|9@y86dc*p> z3Jg4AwDF4vlbZ_ihooaDF@5C;*hU27T(>nG#V^D!T?DpS2re__5)tVT7x4^&X4_6Q zo4Jbfbr7F`k%7P}z3{h-e?|S}PYjEqsR_j8gig7rrTD%yGUOQ(#;>W)-;l)0(c#fu zq)dfnuP5ztVJ#gpL@U|_Bo|x!K96c!K&HxDK+HLGyOC*$&6}F6c`7+E*-64d)}|>h z$Qu$=JzeKpc$b?5m)$Le?6VPz3aphLMvGQ!CU(2jv}Y6&8Kq)Blw|011Le=h)Ot** z^iHV+eVAEQS_peH!Ze;gv2?Lqp&{f8)(N)jkzXb??RpMy-VRG8(B_0<+8FPcaj@Ky zVnDd5-B#uB{1I%r?j$KH6Z;Vhjq)k42b4Q3CG_1-r>$Nh&tUid@0;YrE4Kb_36) zE9x=zKm}2q)IzpdCl5$KKvSc(AGap$5x>KjDqZ`UL_41mwo{(b~;sMjx3y(Zv3$#S0w1{F=(* zmu!}%c4C$sfEyBU@(dn1$q8$}0@l46A^X&XKrbLT@*k|5Dz%I8T4qM762rIaHR>z_ zuxQnx2j>TLs5Wi#!uu|+pUlMQ;^MPP>kX|zVREJ#zNs~!qa{Ccr|)kEthsOPiI<&} zSQ_0BmjRuG*HN{}t>Z~qCJvrqJqGzT(>mqfDGvovV~L6;~`ybk6`N5RDBpptWWSCc>tP*j{^wXr#I>bPV4intj5mgKDxi+mQ=(9#lB!RnE_XEN&9{7%nej*UKHm z;KW{X#Yfkcs|c@&a^?@5cm!NG2eH>159dtnD=$-mBE{UU(H95wwyN+Un!)7#uakB_dQt3M}EJ@xs zg7sJ9XL;M9t`9%gq0h$+AsqLj-K;^?qzfE&FtR_m0(t$`*&jzypYq?ZBjo|j|H)Ev z{!a`l3p3+?-`dd9jyr5c|F^MdpH~Sy8Y!VzoNPjf!U}7nh$iXgk74nV-*Vhg-*_zX za3kmIGq=xKEDAJMML*|vR^;3Secv!_Z<-EQmXlvjJof_9 z(|6P>>*e0!*6p-6%cAvo<)4LTztivP`CJ>y;~<`4=6$TGVxAE|O1;fF#&I*qiVj(1I zMTC1cBabXRtSwHAy>{XTh~2p&_5hARNaf3FhO4~}-&;iLfW~39g#C?A``IjJEb1Ok z|J1Tp{x})+IoNy61k*pS;=QNLPlOkUzt`mG%10*hp@J38&|Kq@vh{C1YOS|;-v;G~ zqGY#Q`hW#3+RYyXz;k7N4bFqG$?tRZ3X&3FXbdclP(j7+D8ggaZWkD5i%QZE(=X!G z0L>3Z*PVo87$TQNF%UVrS@Bw_AWeAZx)my^)vqDgmkO4vXS9YEx%P)eEK|mZ`bskE zEwgfu8Yh&*s_w`GtX@o(!FW}`V~z-2?ExD-I$by;=$?wX@sKh|V562RgH9PIi7B7G z;;F?g0L*pK0S;(*c2>Ek3x&M#3WIJt1#I^ic|_(c4#U*5Fw{(n?_eY*UK2`W%#vw4hcaAF2x(d-IttTNmqCk{qd8fFZ6jgg}Q_LJ4 zalJ)-NSv$7iA1x%nfi@7yn2v*^GT+UXOX}rrj;U+AmJGkIXDK+uGG_}q~tFFb7P|; zC$1kwQG<9D23B*Iv3i`0{CW`1$C){UyMH=QC0;!sVV&}ZxdqpQHhJ9rx^n5|4p7K? z$E`$5luMir(%&3UaCNH6Q_CUk{T5fg5#1Np?8vf77Aq->=`z{t0d#CC3SNL)irWph z_|$um!)tUgCz&R3UYb6<6yg;no5=3qahmYOTdHNlGl7Bye3=$mu&H**J{-)Cen&&X zF@-8M@6ZvfVDZ?oFQCu~2%990*^FU`&7!C#mO44l6qby48$v!zPKNL;)xivk<;37> zW0(V<)_;+PztSaJyvEHW&$%T%Oq@$|uS>wIGgN6x3c5rHZALS0rARXK4E1z~T1Ogs z4Z0pW2*h_Yi9 zuCc!cMxO_fV$a^9b4d7Qt~?y8H`pjv-FwV?^ce0%{7OA zW&^(OrF}Y`XCiG2n4pwGlr*W))Q9{mOWC~kS5hR7jfVJb^B9;@gbL7a1;21i^g6ub zK<#I&ITo;73dgfkYP^0mtRoi(6!y%g2R=?M3ms=5NG$zRPp78f@LBJ=aOEHNOY_|$fg?6u>})T{?xob zNJT5%6=^6$(LKXwppl~s4QuNq^vrXGT=k(Hv9ED;m>w5S7iek4zh2Y0rI4>gRp~kS zX{6z_7(geM&PE=6Jv%lqb+x5eW)BRtB#TL7 zUOSjbuc9<$xV~ZFsf4Ser~vQJnasiIOm22#Do~+~3{Lr^Iu6rObCMzP+b>*u%mCCe zbeasejw+8~#a0^b;ayd_2Q+D~2UfZ!J*kvg6C@gZ?JX2XA+xa=|0S7PP*Q55)Y2X0 ze~DQlsrxYhRb~m4X}7E$!fRFR!vz+l2|0*)|1H=acdt^>JeZf$M&Z@26t=rqLg4H` zR1L0UV94a#)hC|*A}xj$e}~cBj}cKDKY4r;YlI2^&^OBLKwpxg6fpYM4>vLZjkmEa zNAxtmp70RvoLZGvN+~1$tcwQWMXh@Ziz%*0D< z;C;SbS6x{`@Hsu3d`s=QRqnED4>1r2J!exCe^}3*yMA|o$gXT}2$@B^qPDO>W-ye? zhldIbirph~R}}fGX-$=m8UHpQw9tM1<*NjEBHz-e26p-by68K{m8b(o7ac1g2x+N0 zfjCo}3!D{%DeB*Ige4}3y>-YQe1ya?bMvRgzI+HMW-ONWKvn{~TUYoSUR(+kKPjX` zJNS?L5F~8LuL3kQ_~qL9yIV;x1PNwIFEPsVEjgLh-i;PGVC=51sS=18#>lKL329@H zIy$l`=XhZ*BrU)&JpDugZ^kdSGXSnF7JfLV&O0tL``;Y|ZrZwzFYTw0-leVk0Z`;x zqDQ{m3oK$?Ny+-z$uXQ<2g^yX1&$*zsbO1_Wb2ELroZ2{NE7})Y;Sq0Qoz5zGV)gn zWqd!DHWS3SbrulowaCMJ_878C3}zd_=c0J4ehd&F<9m3MdTsW!r~w8hnA|KOyLs6W zwLLbF@z{Yu5ZN z1rtgKzOO7L0y`3#zVJI@Ry?TP}n{Hu^w?;{3R92%F8(@a?0&K|MZLx&qvAXzJ039SjB&J8c^c*1kVC zryC2>+TbaQ=wMHu>Q|RcX2zhWgVv_Nx?U6ZTgJFR4U|tC;wQHRNG0o>8(YF)nFdolu2CY*cY)$d;PrO`QI%lQv#i#92~kiv8L-Lq$*~*W+*Oj~ zjFA801krM|%Cg&6wdSXuB8pk(j?W=m1WH7rrbBxr954T*jng%1c`LTm_i$EGN5BtB z2!U=3NriJ8q=}^!faJueef1HMXRzTr;EZHr?0mqDjbW^OAJ%-mtHLjuK+0QXDvsrA zc@nbNj7+36FwFBX$U|6zcUw9+w)L zn%jgd!U*^mQeZ{HPv{H`1c4I&G3)2{gK%HkB?nA<(ZkKR71~J-C5eg~Q+B%p!{mKT z1f0=#S8#UE5Z2lvefzbTV*rg_ACN$<+Sg3*tbK(-ZXuq5t$~8 zhsstyd7n<+LIz8ji>QoorM$)F$QlQ3lV|1N$;mlN-QntD+0k{*!dh6g)%)X8si2%< z7wG@Z6=NKiE?e$rMh{ux^7#GfSm>+zsq||@>cQB;tiBw;{M6v-I|hNNU9OU?q41co%^C?H4{5s5?2Ip>_?kTU}!X~-Zs-yWXl z`R@DO^R9K@yUx1jtTR8Vr>ghfwNq7f@7>+S!TKLREfu1wVYSGO54?tsPfSZs!qzHc zGK7J2@PR54qB`DCa-YA^Dq~WXQ2oB>1~^L!Ev~2i7`4rKMb1}x?UATfZIcC?E3?y?EqM3q3=;Z~`mHaDFGkp=uBz|p z;N%p2q?s+Fc%pcsUf!y}t!046Cul;F)Xe9=qSK{zP%+_~YaJ-J6^lKmx>VfD8c z4axOODmsNPFr~g6)r-CQ6P+s1V4$cQGot2tYSpnDX!Vtm`% z@pxqCct_oXM^N`!L-t&-U||n-CD_IzCVe?ndZyY#4@a~Tq)jcyuS&Xp_Y0Gg)=?R> z_jO*Sxun}YVKYyK-;Hp+e3EHNJPY$Y&N^*J0L$Xoq*s z+d-gG&=jxi2wxR%{8?H6xt7IoDy660#o`;_m7*Uaza=iWJ|To(t=Iaav8&?;uxxnfwvNQ64a00 z$fnSltBaqwLb!r)8>yUW{C#T_G3`n z_ILey{P76Nd(ID99H#m6o<8i}FJ8gdltSG}ULtw9pIe^QnTGhiSI+f5-^;GHc4CiQ z7M38b9wnbTSdjeEO&A>{BP@*fKGgQpi&Z*nr_jqU&m&)_HeIu&*g)BRRkU?6MtjXv z&0Y?xD|TP)?mKubF@(?p->RVtljx#yYbpMN6UL|qdCARGmle2jXbrila};pk{K3(! zauQ{w&fiobcc5}ISys)|6_<;c$v6J<1_1&%M`htHC@ z(0uuEc0QfrMUv+DMZ%EO!@4||?Un~2gq-XtkIV1R>eRo}2J4nvQ0}eSc8Ok&So#znV%v*XgG-d_BE>k^lL1a?nW`ovy2f)mO95 z&jmT|La?6&*ui5j+ag}O6(T5YeW5G~pX^<7A|-!a^_YhD2LEZo_vW3|469FaHLy98 zm`vXKI-z|b!h390Jg;0Z6cN5tsl$js-7|`<6V;~!_U5KeO$H)#d!~2_JD^|pwC_~hdq;5QfE~WN!d-wl<<-k+Re`F~VX|Gy4cLrQ+ab#bS{( zGBYuS{qA2Hg2JeHuvkQ_EUavmto026Y*8ZzGee{I62bsQS$#Wm zD(;)Dd*1)ov#@MDz$WoGzyDy@9^?=-9rEOj;?;GgytMPDF9by)mN}c{bH?Io-Yw-I z_Wap_`?vmQe$fjj`nYfOnGrIWSzCUvTPlgKYi^@3fAQUW5RGein`$S~R6HEzd|LCo z%e=@eZzv@6nsD~2`6E@$?V92xjA5ML@lyWuy&S_hFrOA9&yym4198#TWURT6fl^8X z*S6rVej-GtQaTsln$J+{+MSx_+M$z^a}enGJ`4l-L5>q90#4#5t4Rh89+f16YHpVg z4dNhgA>V+XQs5v5ajHW6peRg!w<)02%a0M|2lav`fgatoqA+WkQKmrC8xb6U%p33^ zj-R*)7>0qS47;KH{0=~w2L$xZ%8BlT=PCNH{-wPL7(XQDnh~f)r$MR6fMXH}g;^+p zMh7?mD7*oGB}Q`qP#(mgVXOt%NJ$1@YTia6kH5m>Lrva*-~0aN8A1b7Npr%e)xZga zP1FF=SXUKZ5x6#b^jO4bf_z~jJpTf8ae)x3@ zz99h?ZDnFTdrkDLINc5#;E5Zd!wP(G)L>&poSWYaXn^{r?J!Mp0laL+0-O`~ZzX{$ z`X~U?EQAi?aBH&xXY!)sgu$Cp=fYsdC@nx&t@+8I%?b?wdy>)7AF9p({t8_}(gD&I zDWT=2tqg$P6q7WLfl!N*cYsPiL;?;}B(Y6`{HqX!k|HNR`T)vn!mKv$oT07|gpwnR zZVJ9_3S<$wX!>2{j%*r_zEz`dE^`Z6#O~(>RVNjB1KtecwIpAIVnjJo$Y=`-LQ`j3OLkwy1t{GgL}2t>en zUdhQIzIe}l!24TI!GH=Ez>vR-ZmM*_+Ecd9ye;h|drH8-7CM$?*wQ~El{5vq~dOzA7e|ombm6+KTZ`Rm= z)L%=0`6-ar!8qjebcmuD4k3zqU7dLSKDX7dD1R%u&L{C953E{n`)H6ns`_yI*3UM; zS4>PgVo1+IZ(GA*5w{^%z666Ix}H1OgE+0X-7;^9jIePi6wy1A?!8_5y{cVTTq)clc;W9mAe7%AYru+$>Zgi!rZV4?rM zi*^lLGDwdy0xE+N?@LW61L@1Rj6Wb&T`D6r6^UbtW8+``_WwH819?+x$!u&Hv^);J@vr z_y6a0fP}5S^M9@U9PAwb@qOJE(ltKB{#n!BP7E&xCI_aWc5hXC*Q@trnw7%!?`l3p zsJu+id!jWNf=ZdDjr?9>8>C_Jy-ZlS7b|_XeDsG+uMD)hFF59O+eVu#Zvw*6uo}M_ zKX4vq=fScMSvhu#=UsK~aP|m$@IUJ9Ut92>ZnJZ;k_hk%=;7sW$UW3DGyw>#c?32@*~2)H1co7?{I?B)B(K(i6|!OMJYUtlvS$jhpPxAtU- zWJlVY_n zo0vbIM`4k``vo#i)v)0Cku#~OUl1!a>j}0nZbzerXY=yurLQJzFKN6HE6-HxHBlD% zFu#NgJ!}ZbMdDa3t+cw=_E0~$wE9&ILe|7iKzZf?KxI0c0H7ki(M%6eSyObEL0Le< zJxBlm9hE`?h!#UkLAJ1o9rd9#7bLFsrPh5cc>kr6=U0FZpNiEJh*pr)hz0=A%9TqD z?^h8@5%EQ=7;lU}0mzkpl1MHsNA=ez07SD^M$=2HgYeeu0BA~7C=WnRwLD^QD%RD< z<2j5QENKUVq}Zj*bi)}a1|L7!gui|z4X@USNllHPIo+wLg$+9Dj)~F)GisfuJ`dfR z6)q3rN#2(mU{*{OliMP%)@+3FCfOyvE6oXjQLK$3E;$PW?s~3X?K_&tM=@_tCuGqt!_G#wDfoGcWS;J zL#?_#@mOwsb8zv;T=3DQ_VFvkPj z1$S3n<(t2N4{OXlT^yYVNyuSpXM-Eg2iY61p6Ec{#*a)D@Az1u3M1Dn3a%8l7+~DT zG0@I}3FE(#+hSA{_ty!I%If^~g&RlpJz&>+AP z4mp+B@O&}L;da0tp5v*903&}2&%O^%%|1e^1R_>6v*(y#6)U>A(mF8i3B7%{5OI@v zmC4LPzyoB5`bvOCM!i}zjlAK=$3RVDKMvsQ)kTUGAi$;UG0pJQERplHG|Hm?Fx>+r zWnzbybtMPr?*Qtq7C%3wL(|3An0=p~x8Ckx_rOh{tiToD5gMHv_H~3W431eQC-!pX(CB)36_pn5#9(ylcK%MFCMEAFkd}8XGuF?*OUE zfZ}Rg+}{^DJWIdNV*74YTHN9~Z-!T}gJ2}%Jn3Ak+WXMG2pTHL{5yIgq-&!iQ(lty zcmg|5#D2tFmon@@AOibPXb5*be$qs0x{F0=;Pmj$hJs4g&RC%*>&FJ^Ll+yFRbo3l z8B3DqkNI+WH=T;wRN-`pwb@j)y5p-eT7P>F#V}<@1(VDxk5ttPLAvB2-mHn_vy(h* zByWWsYoUmJ+MtGtb+i74moUivt@97%t!Kf2A;@J}0If~r z&e6ash%ezDz!*Iy5pbb{_v^qE^avqwaoK!}fMG*EnIxBT9lKmc0Z5CnuB5<7oGnZR zM&cab@t19s+k$rfv@FVk>5rDqHsE6toI6;+$h38Zp?h&cDup6eVpldWfJvr+sD>Ur zfs9_P05v0aSRDYNRnvF}-p|!DCMayO4{lDc83*PEq_P%Z{K)BaC7>VOF}+;C4eL|S zUISj+UPA{s(6*{23#>xbHV%Ogi964ICI|F9J#|I`063qyvjO8@r`%_7N^_6$P~@Qx zB4E48!9%?Clb;jtVAp#SsOkzT>y|@-ji|rW*MAxLw>K}6uasXrf>54|I;M(!-k5!R zsQT~};ZEP5Lo;muu?a#llZX9siSIbEk4&^B{@U~gfFH)&(JLb+a|CDo21Y`4 zFPG}&g?Fyc z{ef}sxH%_WDWWWpQ{^&h62nt<`|deRzKC)7qUU%wU<^bd(V4I#?C7nJ_TfmoMoaDf zYRz*%_wd@R3L=17D6(J#nBU=g!y+UjKqt1LxKAK)j=D*4K|=sBiX1So2=GOnNSR>6 z`}u@2C4kAK&Mu#k0&umRi|`Dh%u?soIAtp|XQ8LMtH8+S!yDS12F}LA)DwXT$k9Wu zGBOQ-I?8dpk;iIS49ytEfvEs+w%|{0128k~#OLry13lWAqnEM`#CCa2n-~D)&Q%%i z>uPeoAIh*fG4x~(erY?mC)!U-umZodnX}nz!?lg1a6u#YTqT-q>VbpCvi7DTyqg`>tAi`+}G0_pDZD3?*KRP^ZuC zz#i3w9sh%C@_(uyKF`bN>1h)RG1CpVR+6j@8yd9u>@rrYr zqlVVxa^PI^T=0|t?m2@wuUjDp)2*PB_1AK|>e$~O?>>*|hWh9%M;-1i7ndJ7K{JG$ zrn~O@@E4LY3pg|&@!AyihDu+1? zVbjiZxJ9@XFd$JjuPsB?nMi*6(%3Q3wNEm@{DsQ#tD1z1BI`m>LQLYMpRoN=pL=)x z`<Mii5*H*NfS+B|=FZ-m_7kQ9J^zME4E>$Fo}>c&=I+e^Fg=tV#Jni`Ok`8t z!exr41vt`3qkyPQjF3>7KcJabF`$_sRpDKk@)NeXjtHm6Vfb`=Z}z-yy=5k`HO*Eu zq%F)+OwFRD6*c4}!~2tn&!DPEh+VtWchFa7b-aI(L98OMbM7uQg0#rjs`#Jt-^0NJ z&+*$2`otVnF=;D7!?%ANI;DWTptll~ydI3`Yuu7N2KB#xS*Di&{kp_G`DrN?M0!Ps zIt=ReMq#4FnZc6auO>7&D)g>tKz#PIe|=AXN&DyvGC%r1P>byCR~1Yd#6zsgY&*($ z*Wy4Z(nVmW=B;6Bar=N3VU-X%?t2)3jxzh+M1SdPZ(_~1spx1l5WrK|xCt>AqT|wD z^j6$JKqSfgpFs5JO;j08qC6B$LRkSF6h+5fWB-XnCyn1=V!c2|SOL=iobo{jlkIt~ zfWYBbaeZakxQWp!__r1Sl+cmb&(Lq(Qt0bi5NZ~igCUa^?rQ^0eyfBAJjj~xZHyoQ@S}=X2E6p| zX!Kn{AR3*ZWV%^Bj)QFKNB4E434Zj3#){s;X2{t1;JH5#Bt~H_U~m}peO?`?(E{`9 zImtc0;F%vj%nULf7{Q510^JSm!IKj*KNPyk1y#*cSqGgTZ3ta29|D}>dwlae>z$gv zk?78`;|c0g6?!5JNZKO5*OwmV20}-*AMmzFB-oyf-^7^<`DVuFu9|{&&_U;uQ-Sju ztx6#9p9DIn%ZF1R>lRBX z3`aqYjsHzdoML-rP(Yl~^O)lXzjT6mb6VpD|NJwB>p|PIn?V1}mZ7xy#%yC<2B6c0 z%)>HilLK-zlamev_Z_zz4&d@YmsA&7x|dBQx`4Z&Ftza3K1+~OJV8U9IGo37X>Pl% z8{Kr0mZ`nF--MP+f2U!ncms-8e^`uWEOD@RmHE*P$0^10Rjq24(|G`?&$X)bkx}jvG;mkNYBc|;U0U29qMC!|PZZGV!&_;$+b+DxVGweYrQ-3HnJ-S&;}L1zX*_&~Q{EK(ic398!BOFo|{ zQqDYh_t#^%0{x&*Sv~RMdM%Nb1K|vF=Q3zz9v_fktBx}RA;v!T3ts#UQYp?3S^nY- zfQyKq-P@DDog7Nj%LnSsuL?kfz?Zt$2d2%U`m^lO-qSTg{D_p>ZK_7z9_<%NI{Uv$ zAZULNnH613+L6zazk%xf4X!MkJv^AdQSw*WhTXs#F*Fo;o?GW2!hh~6>c%e$dv`RH z2hbi9?!tK_WZ?$CF~zXC?k8x%ep}FuiA!i?8>_C+CZ z-nc{Od|r};xOaqVm7A3Ubni0TXDi2EqDn&m1tfv!<}AoPkDPIW(UtCeyX&_ULAvwk zd2}Ij&3k6s%mU4`BmDbly9K^sBFy?USCAbqZ3ZM zr5syl8pSt?iH?GuWVx&TJzd}-LZq*s>ZAGk%1yTTS~ipeEoEDEwENjn{q>hh`&NE{ zeTkJmb7S$FS>=UxH{!8tzWQ}F^eH`HnL6!SQ-I$-w zw44Jr^YVkXQ3hr}VX2B>kP=HH_Z2V!Ym9|B9T`k4xH4-UTOkKSd}|bCZ(G>VbS}l7 z_C^3M_jF>5TM|rGj_prwFGCM>Kv27d=`3e9%3M}ffpA|@0gy&1)nTa#YYA>ckq-gs zx+?gUi=aMy{!H$x1i)!>k|eX$JYA#A#Xwh*Od|Av-ghC+5zsns`r%@upP_gcQ&9PKh$?eELVD}_!b?16)qB=n7px3Gk+MVW7}Wm9t>#;)3}*Nu*HB27r>`jT8ORSO9Yv4yU+{b4sbpflK#M^I%!A9z&^omyw7d(%+qwDBnH**Vg?5qAu*(=1!V zRybE6I)%#n$}N&6;gq0y5Xa#s?v7ZkXW5kxgt)#ziW%n9gQd`{^G=ER$E3LIfW085 zI5=lboS-@bRHJxJLdQel3@1H)dQ0Fnvga0Q!@^w{b#CJM9a!sc^z~qDMry+6JDwu4 zCm7*Dg#+zc4kT8jI(5)0C#3aWjlMA}MveZIeJlIApV0(PZx-k$vRz9Y$eS_ig0<_Q z(WL)esd~LUdCcc>)h=OkWt^5!NdISk<;Ow0mFt&f#QVPXed(s0n1!vM4Abp02PGl} zOv$Kp`-7e{kI|6bizAm?BQ={mJ6W#v6{9;pQry2i-0@@lI1?iI6IRPWw?Rp*y}PXz zheCQRA!#>w|4N}z7l8yCffO1AuE{`eI>;W?fmiR3CNKpw$_HGRL^S--B?xb8-`R8+wR^{97~k{qhyB!E;VG^a$NL@eCsxM8 zH78}y&J3=$vrZZ{cQG5@SnvHK=6^rpr<`BoH1CXHwv0R{dvgz21QhZNx3rd_fv!;zTx0k#Z?F8Do1vTg5-WpGSp580EdR^$XzaXf{~vCDToqJ}fIQmj zJyGRv_=+0iR9r01Q%^sUm~)nwRRkf&jkOG0#d>eJIK zjX!A_o8p8oDAO0QQo%73!y@r7%a-P+h1?}Md zB*Onuga6us|Mam9=ODAKL-8x3Q$yP132!cp6oF!lN+D6K{J@%%x#9U&UsTO^4(kQ# zg9$a$JsJ$#z2oJvze^>njSjMnbm*==zk9Je2{YMxTmOFGQK#jfsmoQWF8=pviK(in zfT)4Xt}Lpm;h>B6FUJ(k3FjbB`Q%ADVcRuJC$0iN&Q5!%HsfKlyzuJ-?VU{AgGUHN zZh9XZywP+K858u~yLcZz+$MDGqJwIuHw}q*UvCxI`X$I-X7%y?OeZ~)_(u&0rSz4a zBSiI_>BRgo$4?>)I3H!C#n_zLg#D_W>nb^2P->~e#Mh6skPYGrX$L1%9#XWpKovHg zzxX^R6Q~*Rj)o(qbu6Wj_VjT3KNe9s`mk}RcUO5JqDJRE&EfTgH-fC5QKEKGqn1*o z_pX@9)}NO`-iT?VfJZ6-8TSB2fEqM`Tt21LK`s7!KpO~!VR)=%=j5$HqLH;2!E zOo#>&!Rv~}0$?7PJrc|H{&BUbAhgJb%CYNWrS?qAFSv8=sLj%T5RVNI9(Cd40CrN}9 z34wOg2dBIq1tBn;!sJJI-7K2MtM9dh2<$?t&|Y`EKQl|*;pBzdIx zTLBbRm(N!n1Yf)ztj^k&m(OV2m_)EhvBGp+3wO5GjO4cB)wdE5VFOj028r)3-8uvB z@g@?#QXMDyrW^}d2T zl`6iy!>04@bcS+9={znL9Oon-(1e2uj9cFa(C$tSjcP`zXCPoE8NeiiHQy05c?5u& z4y{~_Am_=FfKHu~OK;~lq?bM{3` zHwQ2Y0UxO9OeqbB_@xLSvS7n;By%!KAJ+OJFoVlb7A?<#)_tJzQh8*`cgtW%qw#Tn zo{WiW572h;7BKy6_8=-A|M4V(GLUE!9(ABvKd$z3z?`TFF~QYXVh-7nQ3@*IJJ1FT zc)_J#zJx@z(b^k9e?bPmEBx1R!5HVwpeKaL&nak@3->dE@QV@6b-kohf;=8ZgQt_vuoj=@RH#TsX(h$2lu_d%%)jB-4&+wO5Bl)f9%i=~kV6 zrE1(TOGEcvVU2w7)Yg;f`3Uc~{pAYCV+b6BS8sd19V-TLJD%k@kT|RF^4IYUHH>?Z zw>PEqGh(3=P%xyC9?%cOG57J22?iV+-E!T6KAT_&_%NIVFb-^6Q^bS-eoY5po-zj^ zXdslpNJ3qB4^F*VMU26?hn)SWPrk6CdwkkE&(v;ytBHrTE_l8 z3c#AEiyvsMMghz>UNIqTpk#gsM!KwU_u(1h~bBhawAGHh}DxY zG*TyE4M?TnTY#H+e83+P`L=WO*#{`VWop6=r2(sMd2g{QtMDdHCLL0gQm0$S4(Ufa`8iFiWC0CrPG;$St zd{#I;FpopxpFDjmF<`tAQ^{apkL!fOxhKaytad^VIR2jp6It5Hr7+Lc)AAFK#-DR* zt2M{mCwc};iyll3oZW@(;A$u2-#H+3ItV^iZRy%bEdSuxd{305kPp4^auP4>neVC6 zOxUllwnwas^Brx?94F-)EvkB4YE1J|gZwP$oa?s-a(pel5jD*68b2P+07Chjw}m zBdxF)TdG=FrMs$KR;5tf~}U$eAp?nuh}1G3H(^Gji{{?vPyMp1l~_g z%N!YJY-5fx%W2ccDWAbus<`!^vF<-TC|LvIo;&u%(NSQxs$x% zqZk_os`MaCROeB6rI+zcYa?LlyT18C`+#4_Xdy6>qp#Cva9dQ&IXo&oqEkaD&Ucp_}G z`gl^Un(dSImt%76s@IM3`vB_20XJuEsUxzmd&Zqd={n@yIY+JjOhpBWTvbz7$K`Cy zyyr2@BH7aPC$9di*1udNTdY;o*7mp|MIInm3fA6V8M2#_DAn2F5qti4wJkBso2&7bb|8#{2$`jbLS-C`MMRv4PqC)2v0m zrWUXn5_)${;M0FxUin{16yFD?p)xc+231HBkVu6uy-f%g5dLVUO5%#tn-Akpd)X7a z2d%x0g+c5e)!N!>cZvwt)zfpkq6r0SMAH>;PgKVFT0J5D5-!O-ao#`BJg_;jrP8-Z zg=N^jxK%jed3u4PR7#{6ogVkV^AUr{t0{3K{*Q`v)eZMQZ?yi*Y#!8s7W#I1UnwXo zPI7}!Wi=u%f0{HC*mGAp_0UTG9&9D9^Atprflo!L<|Ez_P)o8MRZ%I9Vw+ zLF7utZ0@qcncVKAyA$VnmYum8D^z=(KT|05&-9<|4ueCDs`~H6k342G zY!vEp=$WXPz6bNh?1sEc{nj14RBUyi$+i9M>G+@`m%!yVo+XD@ajghPunwojy;}Mm z3xd_8AqjWyjLVQ3%HFFt`R5r-rX7Psl_FyG=PBDg4UCZf8X2EThcMh371vn1()v5y z-X5c}dYcrK&O4=r-rJH9j>|0J-}qj8!5lmy{TL&TE{t6rR~et{vm3Q}B@$HErF_`E%R9nArh%Ri`fIrcONKK}``bg{@-9}Kc!USvy0g~$vR2F{v zVc~b&Mvx;(DBe%ugBLi9l9mp~Btq9@mPbzqb6!xPBsjJ_TEDcFPG4B>g#}HRBcyB? zC13Jp7Itw_ST$X;D3XimKhb+nSt;-5Ar7go{{5ow>A4j-Xuj=XG)ZvU$#DjYqvqYm zr6q4tU31DI2dAztR*D6C-WvPe4~yq*a>i9pX_0*B30JJ4b8$gD!S`S(LCHlyn~JTr{@EX9lvzwF z_>~0PO{L;|qPXwtGOr+wnYjz^7U_9DK``Ub1I1pEic(+rZ2Fz&|7uov?#7S$IY*6w zPhZ=>WQ9phHex5Pz!(g?gGHEOJ;kn+?8UFM3;4P&i*E{ z;HK!Ba8b*9!@~rzRtN9#xR)Lpr!0eFdK6md$Pil%F>}(N?y@4UbGn|2v`MDRFI_r^ zv*Cun%5KEOB9Sl%bY_op8gfLcEW>Y~x*v+xJL`bO9d512;oO=|mO;6(20~-gEJeSy z#hLL*-jRmC34d|yCuNRujW`C$8nUN?j7fXoxT=0&Dt?u9*UYarW1b6R;^koeHJsb-N9Nke_o$% zwCK@;^EfB)_gujowfA&37_yt*Y$d4Rq2%D-a>vqDDrtJGka3~b@4YIA#gS9nMa%ac z#>m$Z6F5IfOg;+czRafEx!ad;Ab;lpt*bO~ZJbQ2v2(qwSnr(gIpfLd(43mN9rL4U z-?#mQ(H`=XZ=G}2Ofq|DjaKMhh0Hj2($5q5XH2&e67z)yey@4iE^8AHLaI~I(kW*} zt9UCKb>T?ytWd_Lhd!*Li936NTZvgPePLYIK9f;0&NWBQ_Bl8k zslp4#M%N_N27ox)kOOL~KTzyWyfJm%xxtqJ8eGJ^B&{lV~?12Jc%-|Yf@o79Bxi? zELv5wX;;C73=c59t}F%9sq0KjX5t(><-J=KnEuUdbxiyM-@2k8#K=Qd)|BM_t&e47 zh?LjYDLz?>ik#v;oNrYPec4p4wTy=v3VbVw>$fZ!AA{n~hxz@CSy`1#qn(KjU!+F8 zS$gJ(3zNQs?0SYU?B2`zF0!=d9UBlBpYxHm(fKx>kie`xWU`%{Fnr$Sho{SD_|xip z!%)x1ZB@JA7k>5PEh=C7+)0AG*mI*yU$YVmW@7a|JzE)?KT|tpJiylS{?QDlQKx!V zj@Vy1^v3VbA{)Q^((BvZUSi{40c{*U*2tNc$)dmZX8Z9Ni0^gA_uv&lPi>5y=jPK` zPT~#>pPoH@(XQeq=2}pEe}{r?&wFN9N%nEOQoWn+6yvLW9-rL>yWgh89mfBBhF_^SMc*S>v&%*M>`Q8E(Sq_vU+&Q|Wc(y`?)fiB`E z<{s59Y;|oPHUv>z2yvl9YV1&Fb9(05Q|WcAsUjV6s)n;{j) zaSNp~hEj=mTp{P{#tE&TJ6!VI!kOREhDb%C~_=21%o_aw_Y=TeL(uLghS)KL@6cStLFPr zd#MX*fd`EF))gO2z-y*CuGfT9C5A@p82limZ(Wi#LYx@3aboIYO`qJ%-|jZ9;o%m< z&7=gmYEm6CH#)MFh|}9x;c_y(ba_KF{igAQwYPRBcxV4SRpg1_pwGxx&J;gtg(D;? z%y1AfL-eb$PMQtePsjG70-D{a6}1uOrN3qWD0DuuD~i0Qs5^X|Q1T)y_k-GVVFtTN znLpRO&rce$2U%7zE*!h=XV^U#_I>g54Nt(hhJcTN=fe(yda2;<2ea(F$1h}BM%b=Z zqVYJ$vqk;lwGNjVR^K-jJ-pgD`xc{-5Z;3A?gjkp#Q;Xk{_qbydl9mv#QSBuNpA#Y}V(vC(_+x~9 z`llLEs3RQsL-dka74_B32=OFbuI*#jB&lpfva;kI>BI$ zSKCMFVYa#gRqtc3?q=Sxo)!w3UwhhsnhADxGWqV$PQ0%Ms?jdFMWWR8NbQjOVf7En zpN@g@imRGjwEW+H1v>}u{Lqf?mGhX$GJggQ?DbDh4*ikLU>RXHWcZ@BB@S%z>BT~@JErdf8MhnaE#$VW#b zw5%5~Bp&^7wUOu-|HpSO_s_g`cwv^hAFLbt>ZWA98;LI&9Dg2l3#D=Ijs!$sIpIBz z%yJKO*%hr8qw5Jx5du}s&gM*GVv#aNF!aVOTQiv)S>&|6+`E)UxU?8)<`-%8QD zbPf_RSxTE&aFW~I(io->_-yVcTww^dJjeSt_blB1`efJN_blvO|M@FB>7+BzS?(jG zUa~u_DkguP6G8SphagPJA|~QmQ-dVh@BD)tB@$nm4W$V9JRAL&lKqBFHwnMo4inwm zoh@{$OOe9#!SB{jVpK&`_Y#K;%qw;>uYnn(f-fEdSx!SQcb8K zf}Kt>yp+GXUY=0;kN4x)?j?sPH<@eu`f!;H|5{tlMD`fVt;XQ5pvHv9yaLO+0lz`5 zdY>i=gy!!&?)I~JudqIFDJgWHC)WQpO)J3w(_q++>9ZkmDREwQ^R=be-A4_a)^nb< z#nV5%BKH}$wilv^)++dpMAXTTmle+JN1{9|vtmUq`&n8X8z*%+;LqaBGlpw7a4uNR zZ&wOEyA6{l01JOyi?i~uSDTA76kwL)Jy^~R4l)_&l<9b@@2%MK=N`4c+EbD|s#s#X zq?|*^opbY8hx}LW@j)$jHD{8^n0zGNyxewv-2KY_SSotiS9;xsUs;byd+lTMom57o zN~`IRs$*yIx@M!3r$uEZ*qno<>$BoFANmzke0zSc!HdU?@CS0MUSJmeC-IUQB+=6a zR=TY}NM25O-U~2o;6v;RSAtVU6ZMMMctHHn-T_Ijyxd#&q{VHNUczROjRPN| zOk`wX((_)sR7|YPdmK}SVO$S^pRN5pQ@=zGq3oeA*m3Ep3X*Qeo^>05iT}*tE6j(~ zPlaA>LD?bZ$#ZlX(}y0fo<@1bWq~hSi|N9PqtcijFtx>})wt5Tvd88ey{r>J{EIpHJcEPovuPtWaO02O zXzQ{*qD@Kr_Fbd<&C*IdivKwLypme7$yRF&9~*HmCOjp&>JBe^tmN|ty=AMLVdgj= z$*(=$J0)yOkMP|gdJ>)K^LbGFtXzUYqRc92>{R%3LsDM-r?LrtYw!Eiar8_}RmakA zWoB$4_-vIPKF#xgrZ~Y2PaDc;*^kX;FT)cl9*rCK6sWHVW_ERZc{6Q{oJGICy^YT@HUx5?s&*ANIedzf+41_zf|Au zg(;W0k_4BkZr81Y6`lers~mq|^Pi@2@X>NEQn7=-jBl`Qo5p(JMd>Y(s7LQTJ{#}( z+Aclk3o45d_WRoRSpqix3L?cDtYX-a<9_n6%g^aZdShWbf7wDJQ^KtO_JGJswxt%f zx}zE&d+y*!wog1HOo}UOjlI^h5mC)8xcqg#LuDD;&oEbGfX_cu zYrF*INH$j=OAOMxSZ;oy(oDACw{sPId+8NN^*UVG#uB`Ix2#D@u6Po3>_NctyGb&D zmZ!~YMDX=J5GtoeJK)y0kN2>%M4H%iw;9H~B@^j);%7ZW8}^i5*&wKy*mtwMP)JRM zOy6NM39h`{sz8FiM?On!s$&nX>s|;;*-G+*7^tg%90n8O=qtV+YDmT&jpiB!^|L#m`Wb_k7%mBu4;KLXC^FwziDTZhuJ zyQt>KH_|*&bdD=a)6m^~H{B4M{v@Baj|cmjZJn1zVJ5Z>*B{JU*4Jz?zZSn|{`&j$ z;%TAE7MXRN)R36W=9Jjd6thU4Z&Y?6!MAPVyn1}gJC-uj9dG*=dQ4DA|Kv9B(MOLh zdr0l|b1i5qdUz}jXn4}FUTNJOb420Phlr!rE0ZDR`ZJWX8a5wp;~T#w8O1PeQV}m} z!^)v^j5K2j#`9*+VyYh{FCvq(oN#UnOMd>!$l$ZXJ02Us7%I`Xv9~Vo^sF!KgV5tj zB63KJey_ABK~#DXf4oXP70*MXH+~Kkj>tcwm~@fPWw(N0+u_ITyjUQ~n;F`Ck@<6X#$x-tK>X)la;tJB6g6)f8e^uHa#xH$G zF?6(I5qFuJtnaS#%F%h{!D^;Mn?WK#)c;3;^^$#k@*s1cRzsvML_*8o=lEAZv(BXf zZpMYrejlTgv8hNg&cO&4>QS7?nA%{#BE%)Z)uZHdZ@ZM}Q-9U+ym&9(3hEUP`_&<&>SLpP_DM{L&#pQ)2%k`}BvI04Gb74I1<(?AXI|htO&_!Ch<3s#~U1p3Oh#O6_D;~kT^P`cFZ=POF1`v?RIM0Q{J_0+qN-nr`FWAZBK2xcWv7{*<_PV_DS}=KL4-pzf;HY`&S{o zK16mnS1J8|lk@3YG<;o=ZdiWb{#l(^l$Aaz|9;Q4{3GYc=CHh`1#X|*M4Kt_62bJa zWbLbMnfO**6C@(!`$vGV+nxx`!l`rZ4(nEpEk`DrBVbvcv2*-Qg3$zU5Sqp$#|;56|E zL3FVU=9KfK<;~r_Jr+k<`P{6O)pL1}Z^j8O0xQ?LCk_t!evk5ZDJWHwF6|xddRpxb z!E7}bXC?ZdiGJTFZ*FXD|FUCd@4rXsYGcymzYXXp(3fj!ip|Goz58owT&5eZE1u4@ zX;K9A#>mc}zQirm^!88MoL(GvJgzIBr&@6F+sF~zT3_IQ0N~%9LWGV;v%)0XWNjh*5IAg+YnZ_cR-OXpZo1_U2f?xg<%@#6%NIljX5T# zK*|QOYZMz%w$hnkHiD^N8ZIF_5-~Nde~zahmBQz!alh;fZFft&j*IoUW-+HaX%!uB zF;SaJ1lRHGOO*J8BVFTj1(}fmTh3>6>5EOZGauq5*GW>DQlo?{z7PTRinu}wjf<2x z$$(aMJHJ>F6gis13F@ZsauEh)YDPs(Yx4(GmwsA8X?K-ZZ00F63HhYzWW$e_toM6U zIy2$I(^LbV7rYNtz$ZR$zHY}njCqXU_RMRTj@(!iHyW1ENX8PvP_q5Y+W_UVAlEW{ zg<@E^(I=Y896?%-*jWblr1ZP)g)yW1%UlBRuu^U8pdN<_AZ`QlN2Cit3#_hUmV5@G zu&Hh!hI)8)TG0Jv3qU#9SF>-h-dIE@Azc)I`?*SVhKQ5E;aMTF3ND%|S;wXGz()3) zMQfjlJ(UC(>L@;j898LXY5T|k2qj#`?08P}AtG6OZrw?EgStzUs(u+bz1oQ?=qYX5GJ4eR`ige(q34Ki2uWPPOg{j&uxGj>mZ3&CWk zYlm3kWELeAFG|x!;i{yAST7E&H$2Z;4bPDM?_ws-l=Q@%lV}2iF4E5o5u--8d*%x8 z^XLcmL1XR-g)_CID(YXy*B$PbXL&Iq!Re$GhdwL*U=w%P!5VEop5c_8<^sTqjo4J^ z-_UmM+>Qp1jAHMoJ6uQ*U>lwN0#}s`cD3$?4E7*aIE;M3boxFbZ$=G&@wAoqlMnPG-`~I22W&bVT zBD~PGVaHH6Mj+y4@{d{;gj#gQQU@J3uTXBWb{cEz_^818mTIIJ{8KbUk~>;uWjsw& zssx*+@=VJBtSQSml`9*cPdfVl0Np-g0Dkn)qx{6zKL4&NoOOevrr5M^>P{6Id2Iyb z-m`@s1bArWJr^E%f=8)ls;@S6X6zC^ebxGm$b-T4f`o|s^Lc#z&8ffy&b~po-`0%P7Zv=v4 zJUS&Qk^%{ixW6EF2SfW-Kp<~Aou;;^%CL4s?Nrm_1wX`eMlp<9aC<3Ot)W+WK~r<_ zreVO9QK^;^938NgqyI2F+vCEpZ0(QndLWl+f^bv$5HW3%lMpJMtqJHW#yfW13cVW3 zgGa#lme|`E6BEn6#KO0>S$;_6O&T@Pnw`#~WW8amViqS=A|JXmaBg^5VX()G^ zkBibh_i-HQVVJzaH|eazxF1%~e8~)Hd5+dEe|ySCZm#BtuhmY{thRV$g^fsdRmzf_ zgr-lwE;1|zbh#)M;Jtp`abgt!~IL!XY9(U8|@+NWvKqvSTxP3oUX! z&s9cHZKdN`<>`5HzJ12IPf|@pFV#0cj%57|F`&W)^lv+L;+hJJ|DbY1T-HF3dj#(c z44J-st>zG2WvprxiTaA5Oz`$PUa2M!9X8@P(#VT}1iWK4 zC|uv=!qAI9b~~ZPaClVK0tb8CemLp$hj14~_sJ?vUh1vvwaZ$YWF1QZb$wVJ%tY+wg7eEGBN;`Sg-pFZLso$?5>SurL%O4 zY8ms`)T7fNjXq<$dH696If5mdWMZ0uydE542s_(X<^6FpkKoq4=aa zbd*9|rWYm}goL9tLjvzPt1D9WHS4L_Hkh)MEG?(x?oWXwQe8`L1=fiR^*mdFY!@*C z%s`Ul+P#QtbZflEs(B(rPayW@oI81#JR$ivpSICxu6KuCoBCM@k|hp8zKi^;B?>`# z^wXUpzZ5T67Fcu1kFw8an&6S-`GISn_q%n@yY)SlY6Vl%`+AE9o3!%t?&Sy)h0dI2 zQg)%Q>gJK$il@kdvoLTQxxcA7-yLxhxW$cQI6x+A8SNv7^{zOzD$5(FHIp#oI?n%o zUc{lPU+7GbO?>|o-}B~=*8K4;hG80tF;vg>1}~ZzO+hw-$+@+zxjD3k3~1Uh(62rW z_Unwfm)}~wdqg6v7DYR&!qIqRv!i}ZX~TM7L!eC)ggULD=DYPnQ^1cOE)zHb(ss%md%eCUxa|%ob^p4rP9*N=@=cy=A{s_6`h==rB%e&(>)utm-t7y)Q8&a=_K1ELlG!>706l7QKJM_ zM_(nX#y0YXU_csUb(F5Q=)pV6Lm7=hH}2t? z3FL%}Ik%kQx5a!*y!@vOj>pz=WptsMO28bNL3UAuz?q_1D20P!OMVeQJEv^Oqn{5$ zsH0+6>yRn2{lkDn*ec!#Y1rhYdq1VjhO*h7?DY2ou0s>iqm(9~6oyP-f!qA~XM`D`aHIhU*}t1`{P+7@#%l8yRK?{kcH4 zMK?+v<(g+jawW85Ev>}vx@9paa3y?R%-~M`yC>|sT^w>Y$pN)@crO0()+aooQr4#qQ?ZxNq7ZW4`I%LV_=Cw}ghdae?3KM{Ja2g$7`o%=TBp|Be)nAz8 z=opNm(?e0>%h^)YMw3513pB|JVaaH$v(hBW`=GED!GbQ?zgMRFg3fEdV=Ry2k?(1L zfxPUsg|N~y;Rz%=!1h9tmz?s-JnXp5APo(l6A*nsqq-Ii^W9b#~SE=H%DP6mhOY z^Vo6aaPrR)iS!Ofyne2uRK=C4;xVV<5b;pT&vqxP-5g;|zh#wO!;8+8?h zZ+_7HMW@t?A<7-ZB;w|pdf8HG1Bqzpk?0N4V1$~ zJPkcB=^K?!les^L(5GF z&6c(#8u8S`5oOW2A?d?@NQe2bae|dSZ~ZaMO$G;T@!m4UL9c&IJ1DLuS9`IPvrmkvLjW!cpFXBumgjP!0l>L1C$ zQN!egL+axd+eGo*1!BAT%W%scyBcz4oC?qJhpyO!T#N(Giqgh}2KiCtoD1~yhlwph zNG^8x=q@vZ{<^Z`KX&c_$In?SNYUV5-)9YrG7}95^1a+JY1T&Z!I*{HX$%OaH9Z}j zZ-=x5jqwTTnBK+JW?*-#N1l)S3TxKzOuZq#KkVwg?+Qjz(7xEG)gzXkHNHE&_xna% zl^ngDGHYFp1h8HVp}g(w%zHSAU9>>IXo@nfPJdxT9=!KXL@%%;qtm953seZT80Un#ogCpi+|bVwJAbz5nU6%H*zn^pZ?cCE?>kUEhwwLet138K~+#BwjM8+C!}z0-CU zPQNi&3V;B|?Y@$aPQOa+DDAt7I0rLakEi9kVA&db^a=E3p#7Cx05&9}bx#C(Tw|Zn zu?AVi_)ld?Zy}AU8X(HO8e*c^uENfhNBwbQi0Z#+isnhbmFxSdNVK66H>00^=xZ0-&Rx4`dNqVi9eqjGW=nV5L|5=u$(n9152Z+9$i6m~3c>IwGMx#t%n zVkuSmHRfxzj~!Q<&ruW>bg_^Z18-GkJN3{GHsl(u9C7!#QuDC5~a_}Cx( zgSx-DO~`dF8ialMqpKfR?#N~e8ODpk3*=`zE14a()bd+!>nKc|Tb%QCx>CA45aUM> zE-wOUy`wqzFuH}hNFP<39MwBSJv;SajOSceSdfmTgX*`x?F^6~NB<6>?fJC2>5zG} z{}_G|D+w#L)b@c<5*H#Wet?N!CEIVz`?giGqyTgs4(1PvIOnfb176me*lP@65*(7^ z9dOAc!__>mYF2h;{XK$k!-&jx|13;|_|?HF?8{t|g8C zv#w~@VBz6clyERr3c4G1bKy|pP&9<9GE5XUZT?~#bRjb4>n!3I5A~!&9S78sFX}d` zzG?dsp`Ou?OdMtw&0I1EBwL>Fp#dk%J(1r^uOGa+i4VIcD^63zdX2?rd7&3dtGIjA zFxar7zc5-~L0cszNl&pN3c-Wfj0;>n6wu3rZbGWFRuKu=vx&EMfAId1{s5=K2}XM6 zD8`B5z&TyXzn6r^V%v49VxBL5ZWd*(xdf5y#OSr;l(ccQ_82Y3z zG&TFOEkATiPp-N>FP;eQbDMgJQ*#EuVka0lx8q2D3=&79 z3SnS%aLuPg=gEh~M%&>WvS}fTZ>#bRVJL+^??z3NB1Tj3W591#h88-c zs~%RZqh)$ZD5ge6F&-sxEZ7?HyF5IMA^i!2#142^v@$ZY=`aHSi4e;ryptrq&ycD2 z2Gv*0E^vOwInl6El32r8Q1V&!a5QNo@kF9REuA(ZrPddt21+5b8cEGBxsj>WPZD>t z-K9>`$IJn*OKeVl)(XqS%#6@Bg8h*5{UPkqA39z-@qLGxX$5Lgr8d)UVMT<^1p-K! zO`i?6fiq#G=>f7Z?8wQ^&MhMH)_a=BZ8DpI(<$OvzngPiPYU1QiDKl_7x>GL3jqWRWwCxE+ovMyEv?7FXQFJ9Sz%;evD8U$I+!`t#YUtU>?Yis3heP66Ss$dVR9j ziHaGiRr#cw-R!#v(}`d>*dUwE9WH}BFIp)Yq0KiaW-_d%h&facMBW1^S1vsCRtjM% zAAS4ws7mWQD?k^&eoK zJzs~aoU!BxUvID*XM$b#Z8;S6AJC*K9DE(QD(jd)c5}^tLXNqWfUK;UK}3WZ3pIzF zH)KeikIrX$YF$=Y!wHp8FZ^Azv2H>11_9Rju~qx9|8exd_OIUg-{1eAeRYX% z6}K_``2Gdw2Rjg^fpDAyYX~8iokyd)=+;6LML}%&t5Zg@07H3v`18@v0m&JVzmvs$t+Ob1+#Xq0H0Sok^1@Ef4%5!&YM6IDUn|buPCsJTaiE1U_0U-3 z%lz4S$G;PcY)s?5K&m#N=kq+K2shj8CJw@~IG<*0=)(OnKg`O&cS1H4`ycvr#X-?|luzo-3vKgWc1FK}P#rq6vSLh+H5kt-E% zIN${R|Fu?HA}nR&jmwL0ZxZm4;e7&gCTQ`DrS_A(>#3ch!@e~y+)3C^Bl9L{6xf)v zT_ot|vDO{@XI3_rAg+Dzu*%3N6kttafdK>H;`0|z6CPJi^;!=1c=8^#`aNRGAimBC zOdsq&0?1w`JcM3;A}y+=|q%XtfgIB?FJ=Z45a>=YsLc00*xz-o>jG%Tw^2VvOhpNQFd7F zw(`%s^oJbZL?v2OYxjWxB3=Xo4YA9xBa?lX-vYM}!HZ3rQ!x#oc|^x|+Dbb#F}C4e zEBD0T`}A%#N}BoI<}Nh&e3r!SYL`RYp#Cj3j&SbX;5riXRuB-XSuwn81bs?~k18`C z=-+-aM365F7lkccZl1ZX?y;;-3;IO7M^cgFT~VH-dw#RnVTFkHh0JLMOofKq#Ica2 zC_kpkzFJHY+|L*bQ>ROUrcrD^MouhO7xF?Lv+FOr06a$qcNrRna)y(F)P#?((41KV zB^;fFr!7n#tPsgob*6o6wqyIkw&{7yjl2V4vWXf}@~m0M?oBl5Ks$ats{4DLF&8T@ zNP<8qFvl8!EE~A8nXcx)@rF1M#K@FPiTmywrZ=&xtQf}F5x^AV5YmIr%mfP#3m1^Y z?j65oH(Yzj_cO49_GEY=3+}_^Gwrj1q&4&?;xY__0Rd~MH`VFLzTJLj5wF=%OcFrI zZ1W~>u!wJwV+5G85m*>qjB0Isa$LyfIgudFphSVD4|5Pz#qBPhZpaoZ2NnOt5u^{zu74*JQ8iwx<;0^Ih|oy zL+C4ODwR2s5d!?j+nl`K3{2?U`c-R&U%+ErWi5-(3V7KDg8Fz|KlD9@FEC(8yp-LG zlWHVz$*eqFYVTKc*vl+!V6&?EI;|m~gp~c<+|Sl2fGK{8dh`ah9GR5r#ZdYE;y%4$ zF3%if^PA4BaPs}2mT`&G&;VxDBDTKlV5A9g3XoxTmfixJ>16HqrC|Nd?`A@^G_`f& z$C42aYsyXzTd(Dz%?Z`u=Z)Mng>fqX-nA!tSu!(k{}x!8l(!Y9R4*BboKN;H>Kk78 z5^D>VfpHLCzm}DrY{OP94XdkfuZ8YSuym+29D7W3Uo0E-Edwd-;s#&_*#5uWH*ClSmz$WR51uf zgt_pw31&VrAHgBV+Q~)wQUxMnw04h5t;M`td^hV8@?NSItV7ugC*E2~GZJSDOlNZK~2Co#s8XE{fD)v+cQQu67BD=AFv9SQKw{`Iib-&A9=U97HsHqIifafaN?CE9@72p0*7|+s z%eg8X%!k`y2!q`mw%wf)8lWonkdegfgMVnJYFUM1Z%l%Es$MNcH^3Coi>zruDiPVo zzo!&e<}c=>1k1r^cIP%#JwIPr?O?cRwz(odn)lLG4r~hr{XI8x7%712F9%0!LEBaO z!Sddp2s19)07sJ+28IlGL@0bsE3DNpb*`tjXpkSjaB!_jSh3ahQ+{IF9W6E~>mi)h zLVpYU%1`UK1u0Osr5NPwE+2xfg?ULC)Sb3c9Fc2#Cy^coL`uK=jnpmo>v!~)+(e?A z*Ef(>2h?c6!NV}!D=lYtPA-d$HtBBqbAZOp=sbQA(+ilnZ>VVGIW&yO{{F~YJMoQi zsm-PTq0T5E<;cU_uST;=`RuHIKo_Calx+s_H+W_G=)uU+p!ri2W-v2lXCXl_xxAOR zZ5w3_u=o02FnL`m7eaCUmJAH98i)4q`oO#kBV$LDYb$%Q9ilR=8boy9nC^kq>PEP? zk$|zbOPV^dIwg!DLXf6l0);XxjsVTs{FdQ*Tkm(2M}|K%l;Sw`*0lZQye6u}oh%pK z@Jk5`b-Z9iS?3GkkQ>#7gLYhkfvg8~>?JD3@kkM3;#{o)_MHbVswIw!T1R6?ypRm0 z;GnW^m6y|y%?yHH>onfLcqv^Y?AT=6cP-Tvnj*xf50S6&Qqp)I^@=fqQ9EDc$wqWxT1~lA>Uc931a?ssP(f3NT`{d61jB3>=Qve7 zYO(t`Ot9Mq>{#n?e3zaAHDRG5H;_M9QP7;LVj-h~ky81E!dRfo5V}!J{%C>U_~Ya* zadAL0wjH5l$R(IbN4AxekG+(k{LRLxVuH`Q5lA^K z4um9e%2OCVr~AX6R)mqkz-^s}4Jb)+4nm1J5Od9_pTLV6&NTl86`Q@o4o`4h9f?UC zx+jT29P9meq};}^U!mQ{?TzR{dc~ZsbWuMPf_#8>pGcM(A}Vea!(_ao)ln8%EW!zehl66gJeU&=?FP51eoY`n&g`F$ zJ2$-Bq~hxHKM{cL{i?AKFz^u6$s8a0p#Fh~o&3mzzWbCntr) zmEK@lnD`hOi)Srn#!+BOj<^l+7-+D3D&rRft0IIHcfO3 z5yBk#^hj1B<0*!bUMSM~XwV@W*z`HyBFTKVK&R)Tt|SW50AKCWFP=IUomPvFPi;M0 zTYUXQ*KPB=KMQ$*l-e6c{%#zawD$#oyhkEbdJAGWp*d?)xQ^#X1&6je_}6BiGN;DN z@ovP&&C}-}_|!0DTcTX_#I2rjIr1}eR&Fhyn1t~<)~$fq9^Jirs+AmN>Nat2vK_vR z;92A#B_ee42r-A?lS6z(HDX>a(OQ=AXJp6F{?n`HZr@EQ*W(6DE7;I|vek784bmH? zA4#jNFWS{iCfv7Yk!#TBPo~>`)?C+n2G>l#t#dec(#0&pwKyeD#g-w%pRX1Gxw-?J zTmoVE3y-Sj$~`K1&^J|s%~xJBTKdLn6@?{4m$z%_p>lg17jv!{2%VzSQv3Ai^95ki zZZ2%!Uy5;GL$;V5_zUu{v6WKu^k>`fd~6Oz-?MP90{irQTnZEv!xX#e?4)bjHP+4i zZ!mkR*nt1H7~^98uWJr&?*FlRZAyE~1-}(FV8y_RDO=eDzW+JG;=^!djJ>>6&O~k@ zc0Yp63vDJJa2T6d`1Ppgy9Fv4pKSa03jG?X&$zzx<wgeB&Mi=x3>(V z8lxYEO)ZOhYmjSat_eFIjYgt^Z5U;@=Qu8(SgM$`nC@aDpdSodjHM_g z(oa7OsUmTdy2K9pHv}f8WGM5F6%p?rTe1>ax73F>MMX)nDrmo)K#_xo7+L`z{aG33 zBAVI@-lbBJw8(g@`^(GT;uWWiHrw~!G%4#@h&EIv(L-1vhMBv1|Z4kb)*%k#)HT}d-T z5YJqyc80l;HLj6^NX8f!&Z*PX zdRd@n@YZ_SaGH7JDZO;DPqWnc`+N4oUBzKzs^ht~X6*5>(ag@kek;x5ZyzbpWt1TP zCpT3b@vN3FYEgUE@Jd2V9lT%+38^I|ED&lXqAvm-B1aHJKna60Qt^orr47YAZ0OHJ zNADjHn}k`dn-o?WBS6pyY0}-dLREI#hT_m~ZwwDfC1khf?WJhRJk(rpey1h=+(3QF&u^=~H{om6pQt8#g4@M-s}usj?VrmUTT{=EdZ{SPth|~KIK8uVsyLqh zkmNlE;ECvh8^Ta0AB7`d_H!GuKVzm4dv@_~g8$gPRR?Tw-om0mB52XS-FO=XKo+Hc zB~NCon_+o6h`v+T*~G$wqS_IM^#K>yUO=^~b}|;hgt7Ur)iisK&TojlRvCCsqf+ zKV0g4g^=6UHmGjtY&ScIeo-xTn z;A~wE?1;NFwCCfy6F_lSMDcO@MKbGK%Y`qQ=cowHIaIkK^!TN+#bJfduN%m5_UgV6 zMs&^4-f_P#$hgt&r}tLE*LrFHF_Gd23g5I+*;&ub4fAVlLn79#bnX||nLr&H0WX)ub(NeOYue2=s5d4e5$ZN)qO zHr8&AK9!)u`-Y&uVOY}hDezz`(8FW&=&n(2MG>OLjl_~61})=v=0q`dh7$$JT!bV( zV9@)GjWFN$8cL_bvZa2-p|*w>%*FN!o1-veF4DC#;&30J%-GMls;$8~CCz?Ma8{6JV|jp0RJ3=L zMg8UT?z=Yh5<-vmGd@hLfA7ll5XDdm*z8AOw)I;v!24qvdcxK>P_$ zKM4>2PB0v2E!SF$N1`?PH77tg?mVjvhyd7QYi0EFPXY3Sa1oNTLMgC+IF0&abH*`z z?`w~^TB{MWS=;W+Wlhy^>YXgUf5J#y(rdOYlO7kh=qzsVmYCUWWEm5v&1kWie*##^ z1*uJOHX@KOJ9ao9$tuvZwIx6Q?bjXvQFTg&!VZTZDKxK?gOQMM{7`t^0Xge~w^YFX zDG2_}tF=}SnjmR?!K)(<>zQAe$@&=xpE*Kw8nT-d%yr*cL>27Pk1W+tx&!HiB~YM( zd6PdC<0T$&kjIm6ZmmA)57Vb-f=7x+S=$k8RZ9*u!pEWg86c>kL2tkwk8|f)SjB{V zZ(va3;qj&mk0YR9H&p~1rYe5ND z3bbrkSUkpoNH}%5&q^?kvAKbz9Ee zI(x*zq=S^plI2#-9a`${{5Pi0Qz)C^0@r9R_PkLqtQh=v12S>$I2Rs`BHx!1iUuRZ zRo+Vh8B430i9k9|oqb{ET0eoVx<9GqYX%Zma02g{uLKn#s872azWdT9)(NSworE<$ zs_U8Prf6bB{%I%Z|N-1yTWe0q_qIrIlftpm|UAh&`RI0&TW!`j# z-ylkUaaw;;b+5@+yl*$SceWGm#|x4Tbls%GgQQ9xMnAGBf@V+4X;YhBbIuj6yK z;#8J)#Q^N&lN zM}rwK1E^esnDrMW;aOOg58ANcIocgIlMegepHMDPnS>26%x>no4bG^Z{8MR&xv2t0 zNZ@JF8wOjwC=GQf4VgeE64(ejnbXq`3;gLVc8shCNNEy$Cv(h@7aU=TNAK3sm6O*5 zc(yGsI`0f4=>L>gaHAz&omlV7CpmBa&4+?v#kfRf`wiL!eEZpwpH5n!xyJcURbWUx zCT?UDF&Hj#I8$2LfWvfiah+K$dPnLspt6Hhm0!ibm;#klaOaa^h=bFbL%r)V0swUt zmdphBP8=zcuoWjzpo{AhsfZ1F2(Wy5wpN?>QaySzPXy^e0eFHd8^{-;36Xg_1CrM& zf25(Y3=%IHVhO|ap;^qSUuD%E@a3JD_LTs@pXDaT7ijY~hh(4a{I|?dXSKx-R`i^o z?|4t$R{M;5gMKDYQ~2Gcn4M7s#K3jCpRze|2rfcUWgUEU*YjY!@#?NdqrL-?% z-Bi~FU7xn|_16dvAQ7~|o-ewqNS5f1E7aaOn9Iieip{^eRv>#kr1p{(?YnS!850%EB`MML~0BYw)Q#2WVk)PreTBPW9IPpG{AccY9 z>B=g#B8HD4y3(TiVsPyFuE>|_h|Q!*IR2aJJd>=PC@*$EnpB}wx{KRce4Lx?fpmB` zm`v1VUo#oa+kq&S+^BRl0HTE44$I%ik`q7z5=a?v{V@oG3=xWjJjKl!0k9}2?}Ib? zgMX)IwY(dozh&v`h*bvqYKvTrS9B_`PmU0@!vd|UrGYw+suIG^s z+#xJS1SSa5QhtZ z>42mgZzFcUcb)d5_e((4?(9<(b{s)&Nx|v0Mq(7|Jj>I(@i<eU4@L%xw*Y|cam5zojhuY!B>3=1XXKTyg%||Db@gqqOn30ws$0;NF~pnf_J%* zO#rbBQ<(BRlCsVqgu<2RGh_51Ee{Hjm=FVQOhnElnY7475vlcNE|we~%zYP3*(Q#3 zv?gOXJichGo8$2YJCLqL#4KUK3KkFafBk4)L`)B$qX zpNdT2Y&!QzQ~N!}lgTjwbl|b&$bc-+qSquL)N7qS9;K;Q@<|9Zs{Khx8{uE{M5wo3rCq#Xq8kzRW z%+#a{6GgKdej=!Oy~3Inr4^gA0M{ zO}0@z5AL?br(e1ZT^oN)xLD|$s60b8uwi}R)-onFa|U6ma0B^%TdkSlxgSCIxgRV- z@LQXSv=!LF3{$R-3+s%8SMdG{chPw!bkgP~ z^nO2jZ`FrC+W*)elJr!sD5}2A%Q$QQnlH*ViQqG&5oiD>n{aBUUNQ}<90QgP-1hJ+cJJfyJD!6slgfC; zJ|0>>5dxIWh%DA1^+q?+DsZw*olyD4K$4L(i!kUkPt=@rX!msMB__^EEyIhOCvw#5X3bSW>mvyhu^na-ErJrZYPyKU z*{d5gO?3i1AocIyn0~%en#P9ya1`?Buja{fI@U7P z-3*g5z&?mx#w(N5Hmn?ah0}#dc=5uIkL5B(yqD+{rY*xCj~3ushg24&f``1bo@v!U z2uC~Tv(2%F%iptx%7AItqW3Oopf*wRRn6%_kl!W|WHQMfkBh`+s!XJJuKPCbo-#`y zDQGd#YHn#5H1x)D!&}>-!e;(?razH;o92~uHj`N3j+ixrMwU)pl@cHpIYSE0lZ0`` zQ=>m3Q4AcNpCQ+tXj?ozR;p%GeExV=^b<{Ug-fZW71T&`4YzB0WVCo1(4d~akg_8h z6;d_FtGZrFi*^-F;Z+az+YntK>=-F2pkNrBuGxsymSXs(Z<=B|UXN<;u27cthM}4y z#Os+XvwE9r!Ai8#mTotdh^=P%5R<>~TM#7xYn>6CD*QLjEN*4!^%7^ImN2W)9rfH&boPz#RjWLka8vGVXse}eNK z8+^n)tY?2Q%>~X;O)74>=+bCAIO#DwdqsNHzAV;fTi9Lc{{4%&ksl4@*_oyt&PQzj zEHajNx?P7ysd&f4%bAy`Otj`T_tbM(nEojKxc6mep&~n|TNkgc>guzi`yQeEUk|h+ z)K$TlXc>%A6{K(()5((*a!uZdyKINau`r^x4W#*-abht!K2X(Ia)S||oWX{+Mrp(y zSe)A1pl?D$>%?z%9`+bUvma^NNK%%xTJY2(T=|(Nm+P>HzRiO#V%DXie6A-i&oHJ# z@B!I|-%YZl^3GBSb3zlNG(ABAWZw4zujSiKZXQLq?{-iKK|_zbh7p+>`e?T5Sh`f- z*%_dH-0D${b^OR(~X#@6`4Mjil z7OfJ)mXXO6^QFD(ctWNMy(@4CMTM3wti0162R0K{g2^9$$8dn(QKncq7b$+GWd=^nCsqa~J`fRwEZ*o^%=#Y9 zL}OzU4AIy2Q zR{C!5_$l&M3p|ATQ&R3==51(tW}J3S3gX-eq@JN8{*WrFZv`C7xm_3Bp!@XbUAGDO z>|q?TNq#{_lUC35%#0X4vJ)bByyYDY3I9}TY&Fk&k~zWw&Il0ts`tH^UZ5#|jpyAM z;5v3vnyj#tqO$$Kif|<0ZuqSo`^}fj4)K^d=SZElTNuO+!kVV9NE;cZb!GB7Eetf} z;>_5F208Sar+2vV&?IxA$P`+KNfrMEr@wkZVz{>gLrR8y9!M_^rJWca7$e^K@B~F} z4Fsf_q~xoKK0ylwfihD;-GG9BvR>)z6xXL?U6z1%QwmO(2IZ68jt_(S8yNi-@=nEJ^Hs4Yo)HA0j$>tA^aYHj?t5$x^DZef#G1es>Smm zArYIAgAtgeIvwshAhyBB_0!|*v9kT?ylKXLmBGsWU1yQ8fG+*S_dTa-GhF^mg;%3f zo)Y>r=h8X8yuzysR>}C6rjQ&V(La~^{INTN(L(hWa|6^ChG6L;-_ZWJg1GwgH(b}5 z{88)JolG2x^F!&>DsDT8hY1qq|vUKfgnGVGt)jM>zZ z^w(AXi_V}QT1$Q#JO-ZS!QG2=_DwDsa5i{?-gTcou5PgyJITrI*M_2x<5zSZ#tbYv zXjwx3NIdm)VB0oPF*mhRn{Q`_DP1cmru-@_}=1ZP7P`e;Q`wm4Ev z&r-E?;;Y!4YNBI}(a$ot6ZDMEsVt!uDo!sIn=7tazRvA5{R!^i8>ld-mA$mSPyi#Z zJ81q;Wm)6;qnIl_W`lb&9L77q9l4BWrjqE5G|tDBzhes1jFdP z_co&j(MIp>Zuy<|p7*?;dp`I6Gc$YdRi65Np0#GLwRp+N;9ZsLl@tr>-7XR60k~cx z#iO2IFZZrt)WAu~%6f=|pRReI8Y2MWM>AZr-+tOjVDfGVkIdsIC;7bgt5W)8w&MO? zlVSUS{i!m!k>8p3+$O)jib@nGuB|Ka7%fT{_oUsyxKl{;NmbR5={d*D5Ha-GUJJz` z;h^;+;2)x4gNX<~V6l z=wDl0;;5T)(4_A_aA}!L2ZHh$3++5)HV!3tN<%m+^y<*VgLDowlzT znfzS)J>kJl_Mr`{8Z#kW3vYFXevjKO;ap|@@p{hyM;I#M`zV7{`=2Xe>0mNgm9AUeM-}6s&EWv z?|JtDFD!4PLT>OKZpZlcgW3xMHuf@9;lbXtz~|?iujk{?OLzDM0^;kqckZ651u@Dk z)SVE?kIFKC1(gLYad%g^8aOwRc=PhA_Jtho>k5l7-AYQXoo?#zBb9-Qz-Iz`)JfUn zz2V~m9X{;rzNODh1C*|{Oz=B{JWtM;2!51J1es~V0%n~I#7TQFD9(mn=>k=Mu2$7$ z-*4r^jka2bZJvD{h@CM)V%LZf0V82SYE4b`jQ7E>`vpFC&cK7bRITJq=FCEkUS3En z91X9phJX0gn?oarqxLPy?d$QoX>Nbhop;kg4qD;$w@jx^x;{uYE4)$O{8mJ}BH;Oa z%umMEzhBDvRsf$S`)iqI*x1;-c|}>S6yF0!*pQYsLzR$I|&MnADr{jS~f0a!$hxs9g0iiH1x9@;zEkp3(>>9Pi~1_l0E27)2-l%BOM+|$NIZy-;mDGHsFDkb;KQ- z?bnq&MXl@#*U;OmN>(3lNC?NfzPvSiz4>(rjs&Ib{E{|#0GRW;X8((z7kLx-X+^~S zj(&!U-^Htny@+a(op5CEBEU-n+|Dayf$FWycxFxBF2gb+_9UpoJRv*S+#vxQvm{|t?e++Xu~(COn{ zI5WQIZ3vrf`c`C^x^PCW%-12*fA?iA4mKyp9k<5+cG$(8J2!G%R8;O~8(u>om~m7FyU%kmN|ukZ@ztZ#1oPXz#St zX`{R|DD-p&pKx?#WDvQ~K2wp&wyBfd$ag09?)`|{y+!%6FwLm_@tDZ|*|;h8<3+gL z*9ZRThw3nWdVQA23lU?cm_~-ttjEbqL_dN?zYADRSDLcZQZ>*$GEI2)K0q?tYxYLQ zq(v$G7XJIsvb5sNz31sk>pXlv)S^7jt>JGAaDUy&xpsIBYcu(~lkzUMp)otPX;#wG zm4%U*CEEVO#+#3eXjf)zxb-RCcDAl!FNJ$0x#;BTdQ&^&u7|0E2v1B{#hlYO$B*%n znXxaCpJ#_D9!wkOZLsReics*~U5kfhq|mVkO%KorM}M?@cQPzwoB!C!C7%7k+oz9G zD$)z}=)uyzy(!vcBX}R4VVWNDeiaZU{C#YN#;88N@#SS2rFL{3JPJ_0V>;ByMUXR!wpFmhbgnY43s9)>HB7O2H zf>L>*0>h~=|Cp$5zcybQbsY(I@ky^|3|P0=3bZYyRACr-yXFf z;t^YFTcmd~I(zeOiocN9>Euv)chfgwk#99s6Z(wrtJtT;SafUKV=eBqKySF(udfth zV$-XDn$a<4xuVpaeADA;#8b2iVha$ zG-F0&kC$I)Lic}#sAV6_qo!OZ?w|~Yo)NQ?nVPn9alD${1ZTM6l)UN~{YBW7q3BepnN* z;HY{{OsaIgtVExOr;A)RJu|V|=?RSITgp|VZ)fdS4S4I>n)REsuE|t_>2?1TQ-$vZ~aj8;gT&6bY zn>FxQws)K1cTIj3cjM{h=Ets}ZHl$5*_j*sVxV}F!nw3N;Stv$WY_Kz!dUIm6iU}< zI$l(n31|cl;g7clw6<8uKe??1Wy2|{@+o2UwV>9+^7B${omF5b? z*;5sKw+(9SUI&#Aff*RYMQgjY=XfZqIqjZmN3x}k7`n&i-OfwiH6USm(35VOsOQiS!Ur~JdR)J2vz%LlvQ7cPW*S_l%3InjaQ z_j2(EuBj`chTr7S$IdxMf218(e3Z-WCb%ZGhYtE?)t2Tt9yEk->0g~T8`~WYn9oWZ#$bW>W5)`@|4S}+9$WEsXLV-d#~KFDtt3fX@-9`+Cf^Rf5i|G z_z={LEJPZKz4%=d95VUMgI31gd7^gkS>}gpa#}|j6?tpNPqRD)nSGi&$lg6UdPVo9 zs01~fG0`Q^4lu)zl$Ikq#$v9G+H3(sJ*e)K*D6X6fH$gao!+{G*jym*WhFxd0k zK*a9+*7N#hv{Q3!X2`?u<>|9Wac_efa5!nJ%&Z0%IuWS4^Hq7c>a6kq9ubHM{_BzU ze;snDWp1PEW={;#WEB(;xLVlb;_AjKAud7;`uo1Yg22&;|K+lu5kp%?U~`glEaZ{t zyH6O3xcQ_eMQ_l~YhBuv=_S1kzK}+bm{tvQ>vQGFvlH2Li9+hOn7hj!0~F-k+aHI! zlj&WQh$cxm{K*@gsqnu&GkpDil+H}yVSWHd*eDT;%F*?EbGLrV{0zu`$s5la)f%s- zVB1aiz_5>RX((8lfXWDK+Wy4tdw<1SZ#QDQhg@E60>28{exQRZ7q;A-C?Bi*{ypHV z`FhZkEU2x+l@M4LwWv(DNH(oal#U1eA)xf_J$I;c33UxstD*ASz_2E1#fL9{R9^cy zhA#Q-iM?m`y>;jsozAn_Fw*TQQ-{a5NsQwy1RO;gc~Utg%rxF|e$!if0d64a`aGQtEhqS%7AKyz}_!b$k04DkskZvkicFUwh<*vHf*V*`hpTrb%?ltacqTe^6 z-?w_52svHfN#pnsG{sK~(@AW0R8*4HE0D6_Xy7!P8)9!Ohu+jucyZsX^PxIkg_?QZ z+oyJ-_esR#d5N;p8;4F4gBO3{-&%{EN!q+=$E0H+BSI#pMD|MY z1!(n`Ap?cdE2dVWl=jC*)SS<+WnJTVWtLvgbcTEJ8r3vv-Emhfz(6r$KKHs_l#OEH z^*0}BAMu)xmJM)k?aks*Y7vO5sCrpRO3j)CNZ(ULtMsQlB9!~$HcKvGYHn*^7-Js{ zy(VzvoUIw)=dph3JQbQ>7}LI zbY!+kd*{Q_jAsLx?n@xwx)1Y_lGA-no4~gtVd~+J$MyCwODGftufFbbJ6a}rB4GmbBv!1ho)S#^JI<6AAx5Cxi1Xgo ztxulWA#d?H@9i>oxhZ*8+)&1MfD6!>=Rp;R=b=?)7J}2AVjt2Yj&tJGYWq^HH|(fr zZ!m9q?K7K5^Q=77L44E7X}r!AW`z(QXsg``qX*w`b?o;aB_EO}TlowOJ9-cDW` zZM(af#`UG6rLJs!%}FYu(;;;u?WNY>9-sY>DZkYHwl7==5B^@?%O5_~Q*|<{j>(RG zwbS%Wz9?pBUoW&x`peK7?WO;9R~qqM8p22L=-b_Qh1C=2rx{f~x~auK6RUa!E)wz5 z5R)Yk9=t1T{>mz#&Gf^*3D2}(=oe;_kCo)o<4fHVelht?B|hg?YnHlJ+8S2c3G{-> z5d!+bL-+U&>Qs}7w?3~P-x;+ue?DhNJh*C;+zJ~f;W2YsR2Fp_RcVy`u6f^>S$4VI zZ!AXD*?oG!$)9)}gWax_q;@!7IkH?$J&kBKn2UP;KOUVSW|cJ)y9e*N0o@yI8qeH`~-#ZbJ_# zhrf^JrjEWh?)T2i-G4qRMZXeO*lf^~RN_}TYRl~7ce4cyc6%Ax{!CtZz%?kz>G4xd z$xmWO28>pRJ?m1vhKc2G#N>SBiacN*hAMasGK#rJcQ&Hc(VTRtImg^@c3qt0(8(2- z>94n=-FD6p(`#q_9)}QEh#O|Ydk7xmv81U&5q;OoAfwU5rs%;qtElTOXyLiylm4wA z*ZF6N-`Q1dMQ7=_`J^l?d^;OXOIAmg%6EIuu1m03^T5W_+&Ck%XV#L-3ELz$FT6a2 zWcm;v(udZ6sdjzoIG&KVbA97f5KSPR-d5qKI~PB|Lq}im#^Fm-x{39VeI#&(-A?vE z=iU7v4iTNx>xUKz&%Z*R5Rl0^YgffI-c(_E^@^MM9k{4Ur1f6z?d(%#?r_@P&!JiO zOI19zU6Voy!^M84`AJCFT%L=@yy@F~rQK6e?}Ql`JBeE|-n{T*zL1{wtHmorQx3rAVP4R3KZ zWwz8F>pC2t8G%wf1FG=`8oA>XhIILlpK@bezJyA>{R|Z>o=2)e5UVmM*?rhOeafqK}W%?1j0OC>q^=n6&ayWI&xwD8;r^<3-zbW6n=u@M(!Io{t(E z-xnXG?>2useZU%#mF9h!_m&oI2lGc+AR-*XusG~U+YJ*)Ky%s9!F zhOe5e(oRPB>6^V3k7{m#2HmKaQI1MJ&E<|TsMSe-zZW2^SFAJil0W2&O$dcZJKg+C zW8;oS9?Z*w56csJ)`QZ=yE`2@Q<~1DepT?QQzH$$^HCO9cb&|Dci;}8j@{!kJtiWe z%P-ndu^~dyLI;;-Jayh4sx_9rGg3_BuffOQptZ6-$U*hc*0C*OVwAodLEEfWxj3<6 zj4oYwyB1U+td!UK4&SYBECe5oS>N+(7P|K8ex36fMamP-pALRPJ=Ex@9OGWyDG)TFi5s8M1NaQ|9Qw)fW&l?F zN3;gKyuL^~q#mxtVSILQR9n_eq@;o~0DGbGTHt_@;O5slFV^(8M5;BnONaE0Y~K;m zuwP-X%}yJWAy zRz9kUGqQgz_q9zvSa3$;`P6&;&}XZsS-mU_iAi|=I@^=aLk<>utHxuWQ=d4Al%ocXob%#uB9N{CZ0)0?d- zm8Db3d}9Iao+9OaamT2+ox3=P*+tu6N7=X`-fP@3N2D5aH`I6pw&bp;W@xPrskL@f z`pWRg{5aI8(?%+|I}>ibqJ9VAIzieVu39?~re|N!6TVy^czY~q>Rt@x&*#~1)JjY_ zy>!#NGRA0Wl~QQ3UMqYZpy_py zX-p(S3KOjfHwgJu@ZDgYufYACSE^v=%l=Xu8&*}!gMQWlgM+>sOSgQ;+E*<3g=XGU zv4I>{?F*=DX*|F?i7Px?407y36*26;^U4@~WcBp@!>}E7%quifY~IZaWHIz@oa1RBCwhNy#@#8_-lcmw7*7OP-8$IbL!9*-Q#_g?%$$axb&Ja+Sqj)*a1{&XnLcCdUs_$$gGMCLHav6ih5 zV>pVSSey{www%{}V7s`yX|a)XzPqH>=bK#XdE6TZs*Myq4?3=^2fsz3&Xx+L#q0Lx z+b+*mq)8wU8NcQH&AF~8x^?kz&)S$2UJHBp?yCgYv9db;qs|k;zM25gt8)Yh5@|i%9ka+J=w-XI7Pjm#_h}7s#Ng#ir(nAPNP&0? ztfR6K=HMl6R)JRl*|Rv-uv*k~*5(>qrZ`|Fwna>p44l=}M( z=jQFJV*k8x_0z@sa+h23sy_WhV!M^;-wR(o3u%qJ(tyy0+Mo9SXT`twk{|e|yZ5N; zfA0udQ|JX5P$Ud^8d4yo!6I)99Mk_E&b^vVZYpqnEZ;!_i&DX zrRCqM>*0O4YPjoy8O#F$^PbHzfO);|Rq0lVrQ zfvUOcE@3yu?k|qik94l6v^%|ZRmJ$kr2LAo!8tW*?vx0NKSciin=msO{xksy=ug?E zzpkiIxBmwRqT_$*dG6m>yH^!&cl(MaO^C~1&oX=Bt`s5K|5t^z{K26=4O_vv|87_^ zJAcIfHzKbn$KU)z&6_=rMGlKA6I?lnxuSCQA5{KBoZx;G7F&1fb+Nw!IXs;4c?%TY z<~Zuy{as5Ku~fiZn`VG~2ZE99W?VW(F3Q!p-PD#Q zSDAE(ECTdY4Wn+O?t>^UUpb-BirOE|0DLe01YIfLvhgn|p#OYjPW|};`1)Hhv6Dag zA^l4~wJ?=Wppm6NZLBWl=OV5OcD)7t`AP=q`4L`8m>-gNK=4155{UQ>UrDCKy|R(T zpX~O;sJ+ua?RKB$8GmR6$jF@ew~qgq$yVl{uY*5dR}}4Z#@Qsh{?Q{1B=1;I{qJ=C zQ|i_bVxSwK$$v6&^;6@wE%R03lb;@P9M@5tyBtTvBVe<}`!J1JwK-l3NGAQnpU!{> zKy49+4Ky`sYb0?fpYRhcSDYhlUCY>qLh4R^GCI_j9@&*Gm!Qan6|=z}y&j4c;aX4l zFOOFnc9PU)IxZ4UrEX6&0Ok3X6CgQxxJ}>!l#o$bx>u%Dm7CLDpDzCxKUePe6<^>< z=WAt{+G*cjkDu&U#s(fbo|_x>lW)({)lXl_umVcLzvN1zdr`98mxGUq$5k%;sA3Sn zA%Z;hThxV!q5Qgbw#DW5ypS$O3rB;xig3cT>vG2rEkL*se-KPG!tu^hIXAB;q^|Vm zUipdD#ARm~>0J+R4g|#7aVES4?BeX_NKc2NlOIkg!|dOGyyIT-kr|~_Q8AmJj_|_R z|1rsOS*dh{(HU=xbieAM&e>64MvpqSxl0frPKBQf;6 zdl^FIuqS_Xbk26rIRA(T!Lda?@sv7g_mMpaz)^RiUpam~8l zrEQb&C5Jur71i{!)cWU!Yie7Iu^{K6xLMl(#D}z&ICn^=F(iMgZvR+nQcBLI)`y)A zQaXL*z4#_4u$1#H)>GFL)tl#tQR@raS4@B~F=H)FYHE;d1gLF1u4&y*hE6~(O|O_0 zCMvfN(x{1=^rS>>#n7aYJvgbZ3>}#nTUVyhsyV`{B`%Lda+>dT+%?amFI^uZIDUH%Mi_-?BH`xxG@p}<~Ex@Sr|Wt z_=2E?S?=@ZTb^%jVRq|$^)!jZcWeN}44DACzqA{;tUg=zM-*JTuZ*xQ1WljX(22h6 z%XcXxg&gHw;J#TiWx^g|^PPF7%K^AWl-G!uKn^Z0rbkD#GtT9H5&&$W(#zkCAO=C# z4_Wj*_YH-$2D5NKAU*?&ZZ6_3mU;2EW9}U_jyXZf>r%uXYRN?wUjBfHyPkjiBzC1B zV;R}>u(95i@&zi$@=W9hN!t^^NCH9ZH9OnK-Elv6pNjSvya%xDAQO36o&$NBI*M45 z6eT#h1vA*mOy1|nZ~5kaP&cY^wzfw}3&g%`1U0DT$gCU|wi5$HKYSBL*!hYCF=VhM z?#Hwt+ zvZyUj45B7p^qr~Ds!0q%yW(&V1#DO18?N#LZv2?L-iSa&($WH5lZ&)W(#c;!V+Vd9s zxaO1JY*XtEP1gw;RZCs2qPg@0BsAdh%n#b7==;4 zo9jvjHN@Eh)Z@%^;R%V~jv=M{-WK=m zs?~*jOZ`_&sL3vN6z6u#_+E^Xl)Oi#+P$u#!5b=nLLhYsD4`x``V-LIQAj=5y<h zg9K3L#1+r%FO@M#|$Bju@>QNxe%Br20e23t# z$KBIAdr}4@PvBwHJskit;`L z$gN-y_vx+osMT>EMIcQi9t5uxG=zWx5xIzV^(EV(kpxp%Mh4yI`Z92!q0DPY54diNIvkWR35IvC>jnCBYFIJn6Y0t=kY zU&DJebP^R0qyHR$;D)*TO*`-a>fNFqJ%Y_gBXEJp(B%WkO#+193NJTn8vDv_A~50- zS51W)EQWVPqn1 z?G`2N44eq{1I1gx$sW-NF4>mZ)w@m;0!o+$TPpaE*uuiLVmvqUmv=t(p6c))8J4Xw zqoi222OTq@s6lMs&}=>uRBkpx2y>U6pRYo2Ug?A##l!WS`NXg;j~Kv-T12xo#1;(S z>W!;h1%Tx8BMFLgZ^SEMF&K~one=@5oEQjRXr$L-53zp?)jXL$8G8r`-bCfqpY4nY z&|wn+lVf}OQek@pwfg=P)wvwtk1&NvmH@|@>#;P;Lt-=m?AndFK9_H=?>z?Bdn)>% z=+|2SlL@R8Iimr(%f)+pm};F*lQF2=hY%z$C7x;7hGCmv3Y zd7FlxHE$R$JkCw6KV%KMB(dtv3g{2i>i2P(Wza(Hz4G94I{>Rfwoz8dzJTX~!$z0Il z7eIPFPtkobI<4Id#dEn`!LZ;l6u(@~_Q)^Z;0CZB9m?&PC*{UmzEd*?&y}4ptnqU5 zWxYdGNUMH%0gGX__2`IGraMP`{&G&Z)U_6)qa;hcl!0g#ha_3qe?rH3*6MWTWH;W~ z<#+^ms?Fk6067{e%xGI#Rt7 z_lbu2QQ~x5z+yga@mH8A+W%^Zt%d}kvKPo+@fgBE!PUNj<+ncfq3%70+{#6KK< zD6>Z~q?+Am1k+C_>NU}#){kvqAK2(1<@@QFeRt?3rmqXsdUQ7yHqgpp+I!-?H~oHH zYa1UWv!zJoGrxARKe3>NrG~6q=U&`y(>9X2L&eZx{wnFf{nf!<888EoToS5v$o6VA zu3^-p)b~x5Un-u%C~cVRucq#@i?JnhMCEsIs@A;*?xYewlg$F zMso>NbopEh#TQRk|YaDe((A@>i};Y*FQK-ZZJp(~UEc zjzLsmMTNtk);0by8OSFIUla>_cms7~+kd!xVd6^~OR@>&L>GLqdFL`^qtmjO5;(!K zuzTFk*R!~}e8;^~Sb8$c>jr&XdditrY}ro|RIiZJ1fvKA8@txc%cPj%iEkFm@9`nX zyDlqlm)IM+rz*lRN`fU@u)3O;5AIudTne6iE1?$R*Ak7TngvNcQ z=YfU27sJ4CUE>U&nE_Q17 znIqb_MDOk}U#39v#r}h=ME=9!mqKE|23-H;3sAL&5Jz++rQc!p_|MnRlDMM2EAQXD z5S$ciVMt_=4PB&*D2-4=L@?v0gMyT)WN-Y~m0zm7JMW1aCE z1ndyihPa&_>$|lHkx>`LzxEl6%J)Ajxm_b~GuHIQ1KpW*TeFEltoK%kDF>mBoNUK( z77Dpw^@7fm&lee^E{h7B=DxW)i)zbLt2;^gK!tj_Biwb5nxciWcnv=gx>Uc5IoAl3 zbo|!#1TWC^)<$zISfw%Dk7k2$$1nPpyA#C}j{vb50SH{qRnamA^BbrBVHD|;)`Z$( zNo~g_fuUdXLK2i$GTxbQSgbD4w*)1yN8S?@I9o~SCTI{cd8yJHs!21(IP1L;Pe)(a zV5sxN{@qw>WM9*noFjw0ORP{sd5;7?eO>zDSWAp@yn6Y}yf~tYYX^9*2^LDlB*qk? z8Onbya5yYpRj8saBbz;PAOB^GdW%)fCX1Ry{{dD^V4@atTkl3cKOOpr4%e2~y4T#c ztF%m%Kw;Fl^&%j8UY$hh+hH3Xli~uo@zG7*ntp-#=??;U!HNdoJ)Exvvkgh~G~cfp zu#)L|@tsrJH(s#i;ll<=HAOLBu8+msqtZ3wGT+Z%T`tUXp6#&ugvC|8)*gS5Cs5GT zJc_6oQ68AOJL9Q{M{D@n73oT8yk9`9EKGC;EDbeEyDAZzBD};PBZvdYZ`K(kK4QmfduZ znYG+tmDF7UM4@uM0=K%IgefhX&F*zt?-^3nrtHQ?Ke|I2;FmgguigoxA9ADgb8(G$ z7uQ2@*?j%|{Fak5zKcV_Wj1LR`eKeRpMO2u?h4M4 zjAp6Px0PfZ`yx@VPB%v*^vOmb;lr!mJY6Qgl z6CoQr_8T+#j9dvK8uvh>?E$$pDJg-4F0H4OxEg+Ldrg1%Titk7h}F|v$yvr-1!)GN z{l*%mg>OuAT+X+)66o$!a@-BXvB@%Tb2X1J-W3N9)pug#+d zMK;EEHfXsqe@z`=O^;UC z?_8I!7M|H64Y>YOmGJkYUR6;AIxb+b{de#B-_c#4vR|51)r@a8e4FrW3_dINhw}OtMqUD^vmwQ@``_mmi@X2mH|wBiOVJUf4|@L(LQOZzyl0 z?ujmJ;fqo&MDfTG2j~PT+*T@Lx<=sY?lpFQupYT_W3-S~<>_1XP5a;Ui%)EV2M>Qf zobTsjI81i?(V0{~4PA_EzjUi4dS?Bas=h`%rFD1j=>k$U;^fXs8*QxZp@UY?mDM&fFNhwxAUNEHRE`5aKULRDR=Rd zJ3J87_I+h+_htYQ>y*cRqu*M;d+UjZo9u-aE!Q1_Q2_@xnPT0QqRo=D zH>~$|$*!|eqknF$vA$&M4-?+#Hplx!SAQ=fmhfpc)D#~w#WxK(Tb>7>p@o&C-%;bR z7rR9i(%@WwMt*i<-ikXnr0S*Is;W$8f@HPq$rXT4jcy(2ezCTn5BpL30E1I(I zhi^j!D>(h?wug3~d?j1R`ZMd6kHt8vEo#|EQ~6il<=3HCOOGhxc0&VgE}49Tw{nC? z_};rOhJ~h)mwg>7-sI94JCvT73#BZ#8Kws;+qWCpp?bbD$iSbc9 zt1$&%g%1xbap}m8G0C4FwZ0wE5$3|H%JCviFBJ?G`W0&WVsC3p2W;b3Q2#6NGAX3U zIr*91TZjhZ3)|}V!|&e%3;Ev)Jg9er1YakyoL0C&q8GaV%uBpTTETw7rtHD@v9UHW zm@kt+$|0Go6nUQvEM_w=siy);&do3=54S=%*o4N2Ds=|&y`ct*w5Ur$e$rUV|B!f! z3jXi4x`Kb-PKx#a{lE>4Fg21!ftvxF$J9P{FXoMZfA0{CKK&L1b9> zH)L;B%TzE-fcp-JiJvtS#WdlwbImSRxamLeE&Q+hX#H1jp@<6n^DUG=zyH6yg(4=v zDlYnOFQEtvi~rZ#cWq!ue1_VoTelyRbTtl|V@0*c39361W1WTWX?#RoF6UWFqb%J@ z_}*Cc9@l9x-<`OwwA=OB!YGlO_}^bTQC4++cbx^zOZ)Var0u5}l2XU&NNI0~vT`9@ z8vEaWOe2xQ*P$&~l+gcK2d@%x{T9q=)(!Ka%I7G7;qZyHzjyYua~l&B{r~4V z`CS=UndAbY&E{|2izLA&VmM}j z`g2$(pa(plkTEnHb8fdc?^fcr>#I{#G#XZ|7KY4mu7kPDohE|9Of*BLK-r^fa98{M zdc7*Gwog(Zhs(-yDQ*Y38pc|F7M30zx{jA3l^%WY1$_AQl+5|_=xNK&&A76}G+B$v z>CC)oD6xrJEiLb5SBV!idm0X$-!*(%scpD9c=HwZ)9!In#B!H+T1sDpUh3EgnOSj_ z_0M^S#eG13U}RUXtrWydAp$9Nn8(rxh+4EttqeI^)N71|-(BM+VFrW`p2a$eFY#AS zxkD@ST@eh4HeR~VrJ)V7!iRZ}I${rH#F0MeT~`Ein8OfeuISWt+y}Win?qY^jrG06 zgQCeQ5hZwAG3%AQ&NCjUWvpQp3AU;d{AhCx>)h8-d-4%gzT^a6WcJ+9hgw>uBWF`T z`)C)Z%5E+3V<#%dYNd_(!jTHl2;{ZH*=6TSUXM5%U{2Gi(t3k`AB#NH_~Al}7Iuaj zSNWjJoG3^ieFQjLbE6=C^$xieo-D-mfGS|^5=P;W; zRuOUKKGQJ(fw9OvEY?|T1hWwCBM&?igkm9^ zbGy3fgVr$7`h--uDqmYof*l2z`9;?HnEU=% ztaBS22Vf%+*#u}l3{Zgg&i9@!v)hy2qJe@}KrL!aboZ4M43W>_5CQ z6YK9I0}Y=4bLZ9HNvE$$3o}XPk9C;V5(7Jrt@|FY!VW|s47JI21@lFda4C~vFXz*? zjj;}IcM-MEy7%%HwoQM15dtb8c%<^HE(ALMtO1~rljU#-y z*UnESh_FmmYBX(!K{Jvl=n}TfIc?1Aak=#(f2DXeMBW&cJ@I_6s!AF?HH^$Uhv8xy zmp4bza2o7b-$nWO(B-|spz4c{|YGhxN@q`z~`#C+@nPu8o2YUH!4W4uz4JQYP zPt)C`!NkD%-}`i(pDlAGqB-cHDsXwk>Jds%*+K@I_6}LP>jjuy?Ht2u(B=6?Avsi~ z-8jnQ@_4CZ1To`xzY68xO8WaFRt|nRu~wOYj9q)~*7w+@CI8A)1=r_#d3htcRWWOWZ;9q;F0b4}=_@{yX&4LfhtFGWLF8MFeoW5Kl z#(H4>Ag3rat@c!rwSkbZx0-C1e;UR^52deQ7h_JJPOmOY+jMlrha$238h_a)rua__ zbJNRzwNMHW4@%-)sg_3jct9%`QUMnPe47S6Rlnfdzoc4<6>X?xgvNdM*_RJF9Z2Gp ztR%tKR!XCtwYp?)M4Xl+@pb@+b?S0eHw!sCy||=GaAon%o93>(Gx!>Sb8L=dG6AoY zH+5Z`UhM5KcADusk0co$ayhJoZeFo&32msQI;>V>xaB$%*mmv&kow){+&Cx~Kp~`a z+#L-5Av=a~+!_39+R4j&63cw}9?ghQ?PsJrh%PiX>CJJEYYDc(Tph&SJ*PkQbZzrTr`XZ$?jHWAfp@`6n?P9Sc5h zn9HZf;OV`X>|ZE}u)1SUlw2AMZ3!kt6Pgy~V~Chhtox%tB+@GkJGE6A)X4;u#u{P1 zkNtrgCxm=;z6ULux68O3Icw5VbbN%Gc|l&eylHY=X8?tYq45txT!3(^hvl=n{bL#1 z4XC@pWekWe2s7|H)>(X=zY^Jh7O@KGq(8tZ1S5+@f(Is@9OKQp;{gQ!juZxBKH@%< z1b{HW=FX=-ta%((g1c7>(X|e(Y>Zg*3F88!G=RT{vx}Cin0)J0!7F%n-GnKZCZlW3 zpb*)lkCafsb*z0gxaXm{mAvCrt;>hzEF&JLzF{~S#K$#8{|ctF(BQ2KbW06Vd>zY& zJk5gg4g)ZLg}i*T&SNWBic~pX-NO$(q6^bFh#}0?XOFRyBYp(Z8!EVfb$0Xz1A&+j zt%yXU2D&kFa}?#y zFgqN5e0R2}3(jcr$p@*zrUAH)ef3YIq7?68>=%XkaO5K>qNjEvBeIegsD2f%_<&&a z0mB4f?_vBsa;PjIRz>K9b(F+a+_XOAIK>A9v6;sN)y3wg$Y5j9V{IO2GC&A481bk2 zt(ef$(YKS8M$R+L{|{?#0uSZ(KaRJnRFX<%E7D>uWXTMbq^v1R5?ZX0Z7efpm`e6g zglr>Ak$vCB5V9|mbu2S>2E$keV~pYds9U$&?Q=i(^Zoz4yqG!9^PKZO@BMww`+XiI zE@wWnY$b8ny3*CTa~oc*us>30D?lhJ@p`#I8QlCaC%yyCWOlUzFwanKA$$m0R~6}r z-s|jzR-~p4uW|>k525nl3fe#X)f?!B6J|CW+|y?N9~?`JcJnn+b*aNTDO^Rtk6L5s zqbZbb+~Ws%zG&<3;*hGNl=EUu?(o^s*+~@xIYABhWu*6l@d_Ac7}tdf zFR=eGy!Rd_LzJ`MGuqFkcfn4CD+@Zf^@1c3j$8DYfSdl24>z-+HbiU&nx6?%+~` z#mIp|1{H=6xRd+fRlwV|oYcHDj?-T27h(zq$12g;lYC>?wpy+%^UC8~c1y)2dM>1E z6XiMPm7@(eV7*Qy_e%Q;pMf57?3nPLmoMa0N~gj&#v;K1`j1;k4D6^#DK+5EYEX@Q zDYoaFNL(izMzv)HBz2>^^TOEU!!cNF?>S#R@;bd{^(|W-AGv3L9^bcJt)tSqff}Lx zp(^tIphVrq!i+Fm9Ey}UKQP@fT;>NVOQCivAQ>Fw<9=?xr?Xp?ziH+$rMueU<*@{(l+QYj$G6>5*?!>q8j?#@ zAYq7uJPgJVnJ#-?vMs8GY!eQt-5*O1vV4BU(`=|Z8KtQ9*!m3wJ4j9wpS^gJ=Y;Lt zsNZ=bnVAM--FF`E?wG_{U1r@C!ul$hw_#U38zQ?RX68|_J@3OY{?WS}Itl7I^?`vO zO&xu$W?ij{A4d$?!)Hj5JSz(+2#tNC{YGNjlU`oByZqw9t!sAL&9^>`?1*M4u?x}a zA?|x1gL}5rU-ITKb~@3k`-n5DW^$gZA=aNZ5?Kp9>h;KRPy1R;KSCV5vpn_^>1pq) zt(Az&Eo^J-Wi-okzrm>Z&lvqzivwTFQvN7W5t9bWQT{KB0|0AENp7si{8XWG{+zVL zzh%v7!V`_Qn@4sG>=xiX+2ej=v2W|C8_0PP`6_ZOUGN=RzVG{ z#eFHo6TE-^SP+#sEQl#r_0EVImO$r*s{Gi$eo$^mi7C5STKd*#P8nyp;3CZs$GkvJ)LkPH>|tQu zMvDM*qDu}?lb14D>k<4ji&TWP+>uaXrFFTWW5#G-*-yj%82`tI-Q$a-{F88w3K0ix z$|$V%5BPpu|K&sAM2Ass&mM%E3@js==f2AXWo!a2ydvdRM(@Uy^`GkK-R~5uMt}0* zF(2iSS?P|^dlLn_a*>C|1nX&;Qw4*gjDH*;<6y}`Z*Klgh9t`CXLS7n#kWhk9ZKz3 z3U@|3CRxr))6CGJQt&~BB9Eq>YiwXrhlk`ep8${diqw@RT-?n--4(iSnC@H3gKi}W zFpm?61%v#P8ozdLT7<}|6KqFBn{@}f8oow(>B9B$;{(k*m|gGiv{bd(mx>gyH0gRX zEQcG{MiKvrG+Z)fj3F5u4~lV}jnb_*Ty%?!>mLpJw?que6%*?UU*O(h?W2{s}Y zWt%zdmNM!zv9xpbG+V^EsW5t|9PfH=$C{ui%B)^Jg*!e998*O@`*9(O8f zwY%Jg@{_a|We&z8!gS{iwTfuwGH%d zeEawQzO~$dP83)G`G8MAEU^EK*9}f2A)(p~NxwJ_!wDB{#-_ZuOseBTxH*6Wo3Q6{1J&3o^E#&?O7SCF34#!w3w#GrP)}o7GAIa2((7PJCK}$ zcIOSB{V~%_$_Y{6LEup&#Ue$T;J{HaVo#?m#rEz$o!A{3x&I71c8T!fI!BjEuWrJe z%NoOfGjnhXCVB`l+~~jm>!Yl!H;yAWmDuGt{YD3VRYbqq$^RvL!!K*^w#z%We)M~u zzxVF(*xe3iWzUt9NghwLj5IqdR*!q>0BnkBG;!y`?%nJ36;j zZpuR7H}epe{>eK2D_t%n2AoIp|67+|0Cc(7FS=Y3X!-i{-`CZr@lR}ewE$h7xjBAn z!6da%cyoS!R?WEyP(;=Gdd!24h&D6pdMBn!%ZH{h@9T=vcTJs_6v~?X+=WL6pWE=D9|3EGTjp7WC+YEDAOTdqM2} zT(^9sn4^VYj6$L$a$bripjioDZv*ZRp9StO9tL`WZFB`2>sH<9^XKco+#^0n^MQ+0 z&OL=tx$4nDHQz5d2*1J(K^)(s{9pl#t^s-3hNJ$wb zGVXcaT!y_O8iV2M0JOlC9Tr3i_yKXanfM?8x?>H{Oxc+I-=?*3{jc}@=;Oyz-&)A> zbBKblA7DmyVV80k-!V!_-N5K~RBd1#*7+6de_%Y)hYTIlbN&Met5X|*{I=zL7cfaS zpu?|EP=?R`fl_YNd?UmK71?G&&63%Dy2Jf8fRir}11qXMy40Xw3*Gj`WZg2c)cqSF zxq@8!mICNjpc(aPGrG(mW3`(?eq)BZ?dH}8U0Q8~?3j_#(QQaGn~{R5_vTOm^!$MA zH*kE%;RXkNn=dSqK7*n(Laxzil~zj~mMB;=5jNbG(mlnG1rY~m!#IXpsk}?2IQA$= zULHO}3kQa31d!8o6+p!gl>Ic-as0H&q-^asraT28*h+qbZa--91IK@-nQ0tAg1;dB z>j-~u`3;#?q4l;UP}`fHWmwpl3>e!qkxf6cuM@{i4<|Z+af;~TBv1E_QJ6GIo~L*U zIoc7apkv^LUMz89-EMQq}`UM`M>f#j*7(`G0(WLZ}6U2>nz zK{xIZ89i(tFrQzf=5LamoNo1Lo^Ci7F7409aHMTK3=+ zNFIh|DnV{=izFK`>ztpo9I+h~s9HOteA#&yJ_t};IpO00X5uZ;-d}&NU+iX}8gli@ z_C8oX^y0>C&vSuVh6bOD)SXq($Qikby&krq62wzcpXcxn;1G&5yYC{n=Y=#1hx3!n zIHA_rK24@R=0i}Gk_Cc@^2k-b03yH)QF(gnZ&uu_`CP1dwY97DA_>lFY&DQJ2G=nY z_BUp8q3ZLf253J`6J<$SL%Q9%s=q_i+JbxH$^_i7EV_mRks8zXaGJ>5+G90E-0F5U zFt@25ihZh9Lglba*HT*LP`dj4XJq}Akzwu(D&?|R9it^f$)^2y(FkG6-o>~7YZhx;gq>S{SG51tl>O=ri2aeC z_%C8iu0Op2bN9oN-@Ji$VUmCG28I+50N#Kf?lA9(qHPNglhZ>d^WJt|=eQmI-UMA7 zS@fR8gGkvMduiLN4={dgs?{6I&G5PEzOU}U=-w#EezEGt@oi(S2KTS6Zqlu@6#R`l zAo;U9@cpDYsq^RmtzC`=&ZbVNWnC*RtDx4sG=@kJ_BzCk=lV3aTx*Gblb8G2p7zvi zc8}Orx%*Djt1DWL+LtR|;xrw4ILwPljM0htm>Kz1Xd)rR&mh!Pml~)Ga|uZ~sqWzB zScYEFU7)DI5Z1{}Wi#tb!_oE6qi;<4P6<3)KZvT|&0%IczSC>1HBSMPveJ~+si-v+ z=;x%3)@->M7@6x0c#OTVli^OsT6I))4_$2DU2uG@mfpdUom5d3Ziu1@XVp==q`5fI zp!s=j4&pPdX+?4JHAGCYyw|xd# z%9nN{So(HV{VBCJ*aM~m)}Z#)Z4da*FDmU4cvXjV2Y<2*+7T~-vr8A!OjFWOQ0id* zsQskov670BROdAl$n#feZu45!!FLp1;*&v67vE16%4_9fteCRn17U0(s<2)y6OBLx zf;EhKre@3F1gH7LH2pI%a-}ZVL?m|?CQRvmAjqXCR&?=tnV~nieF|#U{e}%kmhdee zb+R&~8LQ}@4i8wg#%ZZa`8-iwyQ{TJ(nH7YHHTOgp&EEGnqoOKTgdSQokq4kjF02R z@WM-MW;@jG?X|i3c}kMG&Qwj=ny!>59J%A;QD6D>+@%vj&sZR)1N{x9abCpI;U>nN z1|0ddd=L{#LrtdBK*bO3a)iDzIHLQ`_raATOlvjEJGa{!Zn_g~Qd_de{N}=44N^_A z6>Q>W2fcEKdiE)zrllXq6Pd6!k|aDLbp4`oBW!!c_(j|ZXhecXoOj25&XR_f?ILGp zCEbf3@7*;g$kXGL==Sb@_?4Y|{EbT=BI5X0>}PP(pC6}6kGq@^BV-K@@E}qKznCyN zX4->O`S%1j-sjeBiqG7)vrYWHy^;Hw(XC4P1~nLBvxc0HvKT7|@03I5nB72dwj9tRda>HT%v=W3>?7j9s0EiP(Jqr=JXcb=DQ-m6q=T_yU+ZRv``oZ%vB9$(kG^uDX|w- z^!ECHcy)Z@aJYrY4TlrRtqO}rO6ZUCvVx-03l5E@`UY#gxM{azyh?zBU6SvzT*c<# z5lvUb2kzED6Tdyz?yhq^=HKsEr$cWzkt!oHiM5vy2#UK}V-r>G3*+>_2T5fj>s44NasTLrm`$*jm5Tf1WxKd5 zW+yK?`3$8aKcF)z_i;$A6I z<@E~!nvwLRDt|85OY)p2&gs_ZEo90Zf{4yNjbJa+e4bX(Dt$<{vgJ$OmMu$%!nAza z$q$lNGBz)4+AqD`WW1P?x%E`hW;!Vc{VDZ&v(I+HTWeqT(X*|yf-d{e7LFuo+Q;{L zND9Oyc0PJ%W0t@k!lZwkO_l{co1%Qcda&aAgCbH%4pjlxfR^p2h0 z8>Ox=_AM0;7ZVG%y&z0&tV-U})?2#{gg|F^D z=dE3_xgx?e;q2fGC;cqW7nwcps=XEsPd?RQef!jVYLffi5ujA7P@Aja-HMk5hvdP1 z-1l8Ni#}N2C!(TZ>l$C=4z{4`-oOm1PCcj?=hgA#GnstuHR=7@PGJo3DP<>pCXTCz zG?#t$4D#M*ZkK+Ms&IpOm7TmSilf6syEB~)3QA0BhvqE};#-Au0?OTNEhF+IFHh_i z=pZA{YLiY@aR}fl(R+c^RpF?+XG$)WWxjEG_WZt}E%Om8*_}G!yn_-YCY77Kq0Dio z%e@gk%kMchhn4a|Omzcss+QHuTQuwE_U<%}f3|u3%uchuvk_Qd%do18gORp4+^cDR zQ|ln%-Qz50h7a5kcy^|JNcWN_8q^x@WG7B>j(Ks*bmiK+7`gj=!w<(>cBH9Ge!4r_ z^mOOLdvkZR7|fH0Pjyy%oHhM$;~7fG`l`v2E5i1rGA<3YTMv}3-wfM2j&LiNy4Wnn7kO67=JxwfD!yJq5v2^wL*gd&iIG;% zr-ic9EJAMl7JapuqvKm87urTVO6BaY`|OoHA->oA^H5DXF|X#f`;BBiZtsNq4>#TJ z{NQG2w>@%*H|He9Y#>AXlY9Yai)IB6s#?W#o3iwwM(8H`(VLU$Q!vxkheg&RdsRdC z3wfyP8&A-rJB3vFp53lru-IdBB$GWDKE%RY5#z2+p0u_74FBTGeaCKU$+eDsPEi2- z#maw10rY0mi}NP8_IbP>3kF%-=RaAX_AYI^?8zdBmmzMae2%>gZAyzsFEqToocZ|9 z*0g%vkS{Z>dos^OZnnrRQ8wL=dS^js8-A9Y@OF!i>FwjRY?+ng&xlWO4q77vij6~` zg`RUg-=ncN!Pi7XLGZ!;L(f6?7hDLG`QrGU+*MD8UOjv)K*MXfy_;Y131&TQxi!$2 z|Hgi{{EByBmKUa`(7{hz_QPE&E<7Pl`gqcwM;8>II_|R*+J*AWiM^AYrWERTzA-A( zb8$Dh?a5+Afeq%dFJX`@I3GTVfSY&6#DjLJJSydYChiFnV3^ zg3a8CXblU`0@2Ila(Z7;bmX%P=kaslirPb>mck0M@E4||zdVF}Fo&Qrbjb%^to14520X(MW zhgKCWF3M}PTVxh&NAKfQ;fXwTwREum^A<1hor3QZG9q5`7f2~Qcl=j}UGm>=bE)~G_f?$Rqew+~)EHksMr=9xGD++C z-m|LSyLyrp^fm7f%-rC)p2uX=Z^^}1+rCrId;k7(?{}e^LF3&UY_WdbVENHMd#!Zhn(l>Ji-C+XavC zM^r>z8(wu*cBz%dDt46hNZm%cbqk|DJw7qLa6@JgV|1xr{iw{<`K?FJiL?D?<`U`vt$AK>gloE~fh}moBzic^j!InLh6}zY}Y! zV)e*#jrX*Lw{c6i5c@*3*`FjFwF@ju=`}i`NGeqaTt^MoWwO&WLaa z_zhKg&CU)-J|rcE+4w$`)iIcct|xc7uwf>tLLz&ms#c{|FQEFctkVsl$ZI58It0cC*r^Nb^-uo-b0w+7PFB-uK z2{BbFLknTq=t2p#KzJbCe_3DED{Wpmqr*PGpbHsQv)yXYk;;%)^-?n{tpEa?ac0UO z(NJToGSScnI=E=N%HH;1xaed4cJyF4ZrE)YlCh9ObK^VQ^+YI1qcdZe!y zX!e_In-kBdGGtIoTCt&YT3T?;igMZG3>P!Sko>dRGFh_t%#WSxBV7rTJs%4Kx$0_& zN0Il=KFDj(nG%LL3ob3`Isb><5E2I^0<|w0N3vgkJcm!xZJ8TcjCjKof*Y~TMMz9) zAN&+SiC~DKQ%gKCS{4*ym(ZgKnrRNEWVhkBe|FxQui% zOIZ+Z!74YOaU!E>W%bV;hkx9~`s-HTFIy%jUUJ;HkYHE+;-%{aDd|NMxsm@Dy%Oe% zMW`9Aq|BESRi=co4&{ux!83AWH95!t)0vER-AIfl<{X%-F%XH_-R5l55bqF`EG!Z9 z_yoHmb(8YQNT~)C z!xQ`8Hl+T=j8^h7)&2>`TBRRdsV9P;P9D+*t0DI7CXt?klekJtva6=rY%Hl;D(D+5xq9W2w8cYgZqjX#dvvFYCcHW%S6@uWXG#3 zO2MwSm2PhhIt{ZvpJH-%PIiu!)UMmc{}n?6#moZw_^>d7+i7j2&Z?UE``sfQ>KPA2RW+Cz+)Z*`(x#eH)^Px9(7BERJYk;+x+B?vGwsmL zG2}r{E@pL*aL}#R_sL|Psp<7z9(C-s^`^B~I2`pZo%mZ~&wrS*cQzTn_Z7Y{VAZrX zf-;5fJBn^-=^ye#!ghq6z?KSzrJ=j&oK%}z+ z7tP{@LWUY;r06vCja5G$4omhFO!l7ihDGhBTNo?D=#Lp0d2akEh{cZRMH0;!BkrV| zURENIfpBO0YTGRInw=|~@_RGS7afeb&Y~Ll7ixDuty$4}66ROh z%8*+3IfA{0F}Cy}caHB#4)b{wjybD`uD|gbzMAe|{UtHUZ_9%JW`Xl%)*;2q+qQSS zKmNr0a@_8sdu#{hu0>*YkPDCjpTnP-Kj1wf88lLGF8CZrZ>`bJx=PgpE4CdsMg1jD z^|pU*wW{7h0`-tQuN8XfO58g0+ULf*dkq+nKyorEPC>Th*?H75M@Ysg^LNwPs4@9VA{xzH1%ZQMxO0 z`m^0j+R=_AEoR~=Uu|nnD@KYcOuyyt8mMFb*b|N ze}nncH2tJACkHNlSzixS&2~@gm#cHmSEtfSq0fe1`5|oNi-R4U4zm7~1NSunulRLL zd`t(VWxzQy*D`fbBsbFL2#^g8q}2Y?AIu~20VZc6UNPd+a37m{4LdVJ&J!FVpa zG6r!vWf(0V#ZPvZeyw;^gH%ms)eMU%Jk76xL5r1Y-~;pD;);lFbbD+)tt0_R>Agml z!eRr1>|AlTsOJzoL*#baMr{R<{)_=YAvm&}2FAAhIyTVuXG-NCpGxlmP#RFr0fhkQ zG47%7x47Dd#i0kt-f$@K&&EF=4h3?I(~G?RY$&AG0$Jll=n;YfKH84uSL79ZHn4TZ zfj@soX@@NNxPY`<<2cy*$QT{YxNtg~D|jdQoE&P_y4_ZPxE zU5w$3&9N-D&M6B(Zzs=4ppAN6jjpYb4ULgCi?mkznMWA2nYux;FM4fJvhLVq$hxZm zn5D#9r17c>+Qa>R6m`Z|MeAZt~xBr#sN6nrwEE*loSJO&^Yc}w0{ZsrZ03qFK8NW!?EDU$GmXAn#(xY2>u z$@ltX&8OvHWsTr5$7m;wtQxX&!Jwy+_4WCzo!Q;* zN_EC&B@!uA+{0pq&M0ZnEooIhB$|VLhdKR=W=3fW5Nro zZBSgN^?6WhV)i_reJ)0_j?q<>MRM)m89?Tnz$b zwvgHc7LQjc(ipItiHCcFhq)6pCTwTi;bfsF6(~m*TcA}FJi|TEr$XHV?*o6rc-7E& z;pM=G0t^MeyL1iBBT|gxS zAi)LpR6FN3%%(sz7nyCT2#|9$K5@-lCGEUBOrrH(M8WiEp(22l|3*i9P4u?bWDnwzRvq@gjmucLv>xtCzVN}imxE;72xZYqW*Xe8S5A7x!j z;g#0`$*m|=<;!7;9i}o-C|gWl8~bE#2U<7Dc~k~mL-Qz9$wT9SN-A4jj~Z82a(C}} zugTE>bQ8G=1?5ZRPq;)yn^bv+DRx22K@Tep(M z2kfLK4U4Vs#-nwo$-|`*&gaLu8|^ayQkS61PEs#I@q3HIazvXkJGh-8I5{Pzk4lhT|&yCr$ya8CoAB{D}qM4X&ML!Lv zce_f`Z{aDkg_e%&+m}8DCRP$|L$H7dwHUnVf{-u6sqZ|1(%)W#*RD6ndQf~`kzlfD z6L@PNTEyAJi9kjr)3+Jg;zrZc)7!wzNNIW*YU&jEF3Os1$)q?@g_{bCNtbvSvkax6 znbZi0v(M>K8U^7l&EPKsKOY)YNbd9;EnhC1NA}_N)HW0PR-e<(f-*Z{L9z9CinHwQ|AcYvfqBczCgn6 zlo@)*F$~~gsdhdlsk#Vf?xkBMyAzu`K7F<}cv5v0zN1GigHdoF>F+pOLYrM+zm0ax zH%oB(FteBiG&;?4`pC}Jf3fE0H?A;U>62q#!WlDR&+Ax6^UwGYI{K|wo32Ymq&SPy zIgU=;2WLRUDHj#o-^b1o+SiVe1_H>PuRn$r#nM(^muF8U$$o#U#lwqqQcIlxVc=dm z&)V$lQ)^q%)7k2$rQZ$`WgVXTZCx$cGb0j82b`;eB@o)Ov3T)4nfHHG41Mc^(73)? z(>#3cccJa?b~a?h|9zycCabP+dC!8yz&^QG8Cklv!?z2i4n((|V%=`ediTmR=z$g| zQLwudUqji^K%d>JCl_X~H-1(=P@;38_O6gtyikSW#H;tFu@{(081E!s&x0I?%$YkQ zq#DKy!WXPHdAVXAjX%90rP46bL}-=0a^{7C%>HMrpV~%AGzV2n3sg5`Xm-hE8N)3a zlgQ}AINm?Q62h8&kJ-jyN)YaXAb^CG6W|iLW4p@0xeR{g`jF-TAW?`L; zoZfK*{{nP!jsv_@c`5Bk+-4%zDfP_hs>DwNKG2Y z-cg|9$}{Y81Zxn!#$$m1V1N*i%S9t z;{TTd9~rUZQZj!#{o{Xm1q}wT(mqkuovHGB^I3^!N2!LI`NeXr)&uT02qi zdj{u5CTMC-x-SvAHjn9J_{zZrU&JupBT+LiY~uq_@`w;3Om46IbPQ3$=*BdDJeruR zmjXspym}cde zhcWYb2-f-TFn*p?p6{7F?bzvcqA3ML?V_035TIQOSiPfDF+^A;l$=476%LC!Oq4Q& z)#lKMD%PRRLqIaJ?pvE7PB(~CP|$RqbCSb)m1)RyEf;dDAUlBLz8_^l|A%5=>5f}M#C3;MwlMRd=ZoEH^kn)#Te zeR(|hCtld(ytsNTxRD}}*21LEq~Gn|_I@Q*js>#1x@E=z5-Ue(q+9}PNCoH0amZ6Z zK`SM);Z~={q@a=;S@%EB<RsWW8d@i0^b!DIMe>f$^TDX z{Q1;Be**2ndf8R92@TqwH|e#UxoH89)iXJZ)`h;|k3t)+Zw`H;Z7D01FEFvDdF+CU zs}4YLl-FnxlyyG}#+^wF%-*PU_%;h?vdl3>V73EbCGCi>geSVWh@fAJVnHmekikQ| zzD;R0CjkK_uNu*E$fZ2u1_2tFFHo#c^^T z4|A0)Ysn%O>TE#wXdbii)?H?u){>3NsGu{!N-q*ADOeD-hN3Q4z8p@RgYJ`UR+?;wbm!fMk}ClI zAXYkSLiHRYV6+rMNL1lNsqp|(VTeYBLv_07Ox?(e# za%Br8jZ9krI_aTOu3L9Nyumikb}m4LO-3W67DwDYvttA{j*6$f$bto!$ppnOT7!71 z!;#HZgAeQTaec9^OILF0#gH7>RuYXv)4nOGCwbYBVq~0K%C>GjY;&+c7xr{ATDwFS z8uh+4dueObpv*`Ow)aHpQJa^ga!S25+A#`)i|>XBAeK%@7a;L105$+=Ifi}U1<-l~ zM#Dx9gcJLqZc!|N^wdJ2fRx40MfCwfkOiP;wvlojgzW_MG$)G4^aXv9%ZC^M!V=#@ zo8hPok8X~8Fz#2x1NpkW6u*#mN)B!vkWTld)p*iW(^(~8AY*{!eBI6rKygh2tSybc z@^Yn-Vhbjong;0DwL#-XKr@?TnY~JYQEE2`3yi%PE)sRFUK|@XE>kKHPocrMi;0EG zQnMu$vxx0rT%kRHLTW$6pzga4hoqo}aq|FAWf9-%OQGTcokZ?|sCfQrNyoy8gRs76 zRcpEHVG|EaN5YBaE^;(!HV|fUWRxAyLO9AC(~O4oX{N}!IRd85q}VR#6cRNFNc`!< z&?gxfg=wZUeUx6I*p9_{8wrhNPy^!V6I#g>N7Xce9})I`No+#ngUt;8hB>OMV_?1+ zB>LjO-^~$<=HK*rbq97V<_{=g+fWc)F{5e-Szt4Wg1CXgn&{U+asf&FI!*R_{AXty zAx0pnk|5RGRY1+thhe*Zpy0bH_*ISnNB#cG1w{2yh2j=~DxXly8#W{o*4$A|0TJc2 zfqFPBg|m2h&&p%euxvD`NDQk&DOn9MhFaHA*tw3Bx}d^BK|6mjX7qXPy0gwD!~X8*aqQuiG~KGEC_&ktWMQ9Q(HrBM9}iT_z3|okM10hwcL&>2nwdd z%{qwLeo`m-H(Z8^wGh#>uXEdz>zgJ=%y=x#(n>d+o}HkuK!B@YP*ej^3LxoIlnU`X zm;LJXZFmc?kZ=b~r7?fgY$DEfjoXZz#H^|j@gi_vzoi65}j zcbk&ClgFA`^%TyubU+Z!Hg^-noosLyz9i^5>UR|CT}+wjPNO!;)~2I#^)hVRNtjbj z$-T)RfG03W@SSy-RGRF`)OEb|msE@C6Az|MCsNmRiN)BCo@l$9^nRO8pgC+R@?uJZ zsn0j4y}czGmt3;TK5!VHqzyKeVPG0kwRf_4Ivtz=NK?)XwBi3pv0L&TM> zMsE@3tD`te2E~(%CU*W@sml6#-#n5JJ9g|hG27R`>6a~SvGf1_n2SUHSx>f|ICvGQ zcyPAi-Ogk0Hr;*tRA8qhogFe)fM5+}?+#6so#@;ax-`}PXz4x0)YH7;YH022xzDey zcu!YDhuKe=+Bd#_RABt%mNRs!mTPCp37b^vS)_7>k`QoO7OPcu|Rus7t1H(8RbRqi_Fh|Ge>-x=M_A@_sEJ-pQH)@lvC_xBw~=AP!+f&8jc!9%(Tnk)K{iQ^S)N#Y*7Mq^i?ZFyXmIoR_4~2xXS+}oT zPkTnRsnEDP2D{TD#e)cU%lnXIBd1Ehn;U6P7%?vT`Sw$SjpS+CXi$vSLe*l#+F;U* z%^JH@kFM5T^=QSkN@)BOl!m~{&^0xWj-~7G37L+H1ZJG2urKjF%vxWtBrHI>XF0Ii zks-4KE6kBBM;p1;8K)2U=s+E#TR=Ia;lly4hXhVvx@<0%=C2}^l6(f$GFa5Tn)s@f z>-2<1nXKOGKx>Qd-FHdvqs;o^z7}>L+E6K!jmd5(Q(r*ID1NQmjt7cufu@`|sE8@L zt%xB__z>wMj0Gu;)0TO{3FXo56@YcF8K;elA}8@`OX#{5M&Gg|L+7;G+^lV==QWUj z%*}_*18s)cWkqG-obr?YHrNU5U>WR+g2Cy7>Phk>EPFUveHs{rGn#l=>IR+C2)Mg2 zYsTs|sc@K^yBP{+vGzON_WwUOcilPoLGp_!pX{fV}RQLuo1iZkphW&iGJ{bv2)pK#ebTJ%r&(x!nPh-%y^ zJj2jKcK*he5xqfyn8gzSs57(_&~H&%Gf49almku%U;uFFkJ1P>3WcGI#LE2@X(N|+ z0--8r8c}P0ea%U%uM4w+pw?Fw@bm-TsCf;Cg;MGKIZ6~`=JL|9b-*)G0=cQn&VQND zd*CMtl=aWgQ96k6z#kj00-)p>aO%DSAWSVGOjw!r_6D9V0OB79Y;4qa85Z&>0RQf? zOdg^D`+cG`TsgcHb!xGsU)Z{hE^Fai9mJju1VV7?7tb~T2+wiP_Mw;Rwuvb{D z=Vm3T?taw(p2g~dm7Fy^6sd00IfpYTlg$Gvv?1}R9DU}%79-ZOpkZi;P7wpDBt2Z3)cDpN5z<-dIw;LBheVC#=Mg)F|dFR%4eaPZY_^FaqM{)RMZP~ZsU|7 z$y6dyMjh*c+llEj@FTX(;!}x32lk;X=SqTSX7NeH*4@DQ1_3w-Y_kNwY*R()*vMMQ zs$(LAc)Xj8cZ*WB0;WboQFrTgQ3kAcLm{@rce8nJCPqJ|ei zOG~ZYE5ytPU}O|3QO(y#4m(^K(85F?$J)7Y`?#)j@Gos}oeaJf1guALek9i2Ko71~ zUzovSAqg8|b;qa%(undCg#ohs8Koi^1Rawx#Ex!{2Xs>wu^31By3!!U{p)ILzpfVT zkp!Uc4Y7R18Kpp+lZ?^;OseGoP3!zm_GQk6%znUdY}a4;yw{?l`G(=tiy;G9*+h73 ztIbVJQcLe=If4Mvp}Wh#@z90dz%jo*!|1dzPV{}pEP2cSpo70^eBuO(eQiV&!0Bt^ z%+ma|rmvR$FMkT!DI*RC0aZOs>%JQQBV_x-gvjV%&DahD`f>*<=1{`?h6$`0qK&(W z{%O{(VcnY(F~%FL0yq@sUF6n?bMc1R{Y^a!-<^WliyG$UpMicJK+o4tH2d9L>GPY{ zY+7|2v_ReLIf_o{zHAB&IsL||nylOt9!qSRTK}q2If=_O{2DOmYZpMX!=gdJ+FSSP zU%>wn2maQ|{P(MQ?|)tf36J)D;>z-k;I97CkQLB&!6fuIDXM>F=m|n9bGY& zaFT+e@C1l*I4^mFpnx!7hRqzH0gV(mQR>T}j3jm@5_t_+2>WB^dYWGeI{sr>z2UkU zOv(9R-a=d`6s3OP(U*-=d4S_>G`~m1?GbH$I@KA5Ma$!ZquzOIiHEe`C(+z;KgQ{V zwfXHnoi0)`JMK`>mbNILmL#CqM2%NFlvfdDxB#|Wq|sxNxDBEGHP;Y;-5KDitiQ?R=C-{z zHwj<5bV$DIrQdp0|GeTILu37Ha)LhFot~WAfrE-{CE+KI30?K9Js@)z6nJv_=o&gc?wqfPe~! zNC`@lUZsVOf`uww5_&HQJp>YRmb&-5pS{m>zIR{WpOfn%Yq92BV~%l;v1ZAfYu*#F z+XksGr;=BQoNsZcdeo}?=qG2W8ElqUxHGIdd&&D;2DA)0jmoj%*|6W*gnySy>}h@! z8yNka8`jF_-TtJ1I_2HsvuV`)4;Ph~D}y)pKkz)u93DpN!=uUb)-?m{<9ObT_KOCb zfp)=b`v{fkjBzOJBC?L z$Zdf>l_9I?%X&(t{QROGvU@h;H?AA(oELP}pd@B3L&H1_$jKh|%rq-5_*=QK!o%BL_=#0z=NKb*Yfo#(;W45_fL zT}(dtK3=M5cVQurt!MWj=Ik?lc?z!V7`#ZxvH_1R#)Tcj*pXG|Pf!N4V~Hc)LVRtR z$)VdKoqa8@9H4L3Vase%906I+n%teQB_;?9PM=9TW}MjZIWlAp!=xj+IEnPC6j#W1 z$)(fl+t4S{st62sa)8iFOVViPQk>om*eg*i6W3|#4Z(hRmR^{OWn|$)1D(M9>G{@N z$WKn@KAG%(7?OP^zqozUh5OkZEr?d6L|cVo;A=DQVk0AxNiu$icq%4;?UPw|h-+%| ztv)6BZF}_(?^*J7+2t=d-Sv`_zOnRh#oLdP&Moz9Sa!`osHKlv_as3-T|C zBZg(}K*amCc-Lp$y^aj$c0c}*mb$AyH+bgP6vi-Msa2WS`7QzgvYBekhMZ~D+%VCpu)^eYFm#B56N zS7KA4#S%MrT)o(hQR`=nH{)~;hFrL~Lf z7FT-vlc_LUSxaYo$v1^!HcZF!@B=!fKqm8Lzpx9eCJe0i26z+1SeRcn4g@F^oU&;; zhFpYp@O@ibtk;sCS^UO)&h;cDDs_DL1xbW%zT^eF6b4P@_YXnZ`~nx}`D$`ikZ+{X zh-)Bvq!uQ6wfFPcX6er1y@gpRrY(1#SMpEp^YPBuo=OQNQmVT@oG6~d3m1;8+m!(rRk;>zTgx-3i7j|y92_i0BJ9!_7I125Y~l z*5{pd6wz-(we+~w_=uEJd`VU=r8=RSNM0Gt^ZL;yLDR0^Ei~yVjyxYO9pdiZbRb~7 zZdJ{~pq+Y={B~0mqojl9Ojqu!E^SxZSMv`nX36S$#S>+1T0A&zVM4)*%|Vb zj>_H1o6u;!%1a@nJ#hVqt@6e9zC8Qd4fWEfpOZPy87*!o zYf^&;SnFdU8cN0xB7)nj-Ot2y+>_2 zdXTi=GEX9ON3+EHNEN7M zyJiy`$Ff6iS5p@M(6ZT#buXO5E5!6}lsU&3j~%K-vG3UDJX4pnzIy2UC!V!eOO^xU zq8WAuV+Ira45Qv6Iv%AuH97T+3EHTK%~xp|w>j4Nd1&?drtmIR#%!>3iM*xy5O7W_ zO89}ml&}?dsxP-=t6Hu9NqO=pim<+vsjfclT@Fz!A2t>x%H+|06e=U{>0Dz$ut}#9 zai86~nk$#WQm2Nfw7akDW4}Fd81S->M!R~hrMHFd&aPgv8U<1@!`f53)k3vYW*~~O zOk(uAB9R-Y)RPUrCb`#7dbX9XaW67lcQjOJqw~7hhkFxuemr6&=`i4MF0Nb4)q*Hl z$W$|v*b`S$jTq9Hs*4L&*E&AD&ygN|&r4;xAf7pcGBbaT`{6;a>(_}XR?j8qDNj|r?I(W!ran)CP5J7_y!J^OMRf~Np(Zi&JH!^(zXsLO&&}+|3y{2@ zS!Y_ZzYEuBGpJ21k)k~**6I=-ege7xVXv@NjO(Hb!8F1hg!V-relm*j8o6v`?vd~~ zKv!z44(HCuz%MW*Wsc$o>Grcy+PlD1K18_GINhu=la6(b{upLi9+1bYryl&KxC3+b zu_0M$Fn@;s4Tcq|GSAAB9~@|yM*MBlHXVU{nGDLrQpc0Rpz#X9Z;cfnr+nys4ot>= zEbly{WfiIIc|2s&Iq-4OovL|eg!1-DFm~%?&sbU+SEFNlFUZfenE!L3`wrgs$%)aE za>&HcNM4)wn*qDT!z(ii*29gy=Sc32D10h?QRW{w{4PuJdr z1Ux2rlN?*JKcga581=&o-4<`j^zD#u-cCj6xke|OeZJkNQ`*;&h94`Pi&;(v#wL0P z(DQ1|8l(4%n}nHKU!in~OH0(M;=j@=--i#^7Ii9=Wd-|$%FS+a6 z&tght329zEqR>x|CmH66`e7a7wG#22ofFFx91%3KscFNp-U13~bGbEHviyyYl*R@p zA0Kw!Znj&#U*NW#IXYTZZk_&_uBD>BCz2!Az3Tpga>{&8oA5awwlSBLb(7fbg$6~& zR%*DU&OP_k2)coFi_}|tsG?~Ifu#4DTKV6M3#9<p$I^KmPgr@8bgfL{(3K6fXus zi09hw>_KDsynEv>50}4^AH75`ba?$v%FtOD?>ph(^boY7DrLxE8`HB z*{;z!$$f$=#k+Xl-6X>6((QA(hdM3Rk#LjQ@)Pybtj}(dx4!0odATQeNx(r>s*F0f zr6?Pb+8wk8ETz?%({V*dTBUn)vN&QD43|ff<+qkSGumEv|7g&hcwPV&#cM}j&3q>l zHeHOt-L8@>zMX&W(V|f+Zow~dK_G#a&p?g$Zy=HUD^K8mwPIFQUhaRkQU1Svs}@+O zs{qXL{@a|QjEtooRlgrXAn z=jGoK0*M}kRWF0*)KJ+xP^Ez7>w>?bJHLfkD*1qxb!OdBHZC1ON9;s`aC4O4<4#8C z|MAotXC-PIK80*i`^*T%`CwL@@P~fBiXvc*xuT$P)z$g#Wnkm{v!mH^*s{_HxLZdh z+PL-YiDF7JOddKOk6IZC|Kz&{7CVG(>dz`>XM>)=>g&UoLQj%A3`Y{_A)|4H-QU9 z`d|_u2lS+i5z2Y`bvF{!t^`~F>^eC}S^Q<8;Ve{zc>ecUAor8sW`~{p$_U&X4u<}L zLa;HZ;};Y#HT)8A%YA4I(eGRS!1KI6`LC%V(9nOTQfxaH#GlL@jXca2OLNO!$adg* z@sR-^d)B=_alP>VZ-@w~vVS?ixebc=qvbz}{APHA`k%hx!z|KK z{VVT3{I=NtkCRB)@ToU>)Zckjsoa?PDF|1$sr^TlP_W`JE03Y#uPfWPJVXZp@3$jb zn=HfKJzCNjLE-=|l~ypGNfo3;1)ty>u@`e7=0un@_89AfAq)7n&yE*tAv|u&D1$XK z;_905zVRR&87f>gv}I~5yfqB~zBgP0*GLSw<1c~bG_U=R^xq9ItO5yM|F`%Xyomeo z_@6;3O%4|SD_nmOwj+HDe4FE+x9W8zlk)yuERUFSf294D5P48*bR*33)U|zJJX(|c zkiaz;p|UX)zHkT%A$?EyXNf>SJpVq3|Nhx~_cXi(h%W>_^t704<%kjpJLDh+k^nl} z^QjKpHMJ3A)6Vpqg;jcfnf|;0f$c1IJNMw7kpbhj7zsl;c_?Q6YcRc>f<0E~^^h3w^XGZleXuJ| z;?N`JXCUn8Q=AXHiDGP}2bq}(7t4eyS8STPl)Q(BGhN;R4#oSccL}00K^Ez#qB&GfICwy#C-r(&!qf2`BjY?&;l@JMVT<3}y|DzV!@ZA9z0n zp^uD%64nYE?KUJH(!zV4S-b&FC>dik3A+a=3zXnQeB*=^Y1d#ME+44!9KdS+ zIlqi3JOZ8sg_GUVT2h(2U`wG=0~;tT9^`P^Iamw3KR~d=Hh{voyFK-X2OD67L=Z0O zq$&JxoekW^v<3ZuxvJ;gbL3?cOzQ%3onKLYoisDE`!Z^-A@amrBQXNq$-M3jqjs`Y%)bHnhp9z`i>vaz`nP z&PPx!V=&F1ipQbMq5Zl2JWV~2IT@zK@ey_+q;;Ip2I_v8#jFX`I1Z;ekBd)>iMs}^ z@ciP%$LEeSV4&rZ^!Ruar0uRTc@)g@vfE9|z!zc$G zd>imi-I63u4`GPaY>8UG^G=4x^wZ#T{FZ!#ek@hNdZXWRit(I=H$iojU_M38d+yw&UFKP^%!Si`(R>Rh^T9@498;jQ9rrZAj4+Y z&G9%R4x|in;3ntA>CGgwgM70WNavhApcfA&H$D*0I|K1_Jh}0#J8#aJ4H{62tvVeo zOQKArucwwXjMgkN)X}J5ro(syr=tC3*bNkiEs33!ny%e3m zB@n}_uY??qe1U}9s2N0Adws*}_UDs^8Oh%aos!3YY4EK{4SAT|JIe$1#EVY6%==yo zLpDsCh9A4GyzKU51ub#B;C~2q_E~izpf8*nmIUs9@IooSy_p|ODT3D?3Gdm~-bWqs zs)T5~7r4GZdB#x-+h`8ExBI&MqU-dv^WStO=Ze}0hdaAYeNYo~+iFjcZmeqUf?1#> zZxR9ck$S(r{N>a8+|ju$QoK02pN5Ia)5F~E+}o8!PXHVo!8)Ilxp;iJ`yI%DAdA48 z95*ghkKa|j!%0B}?mkROO`?M@dKVjq*GXPK<|u=8ty?z*z~vj0KG58bB${{r0Idt% zs%%ppE9RlR3V84sfARP#1McHe(dDaTOwgEh?f#;@W=&sk8vUhkKtdg?t~qVG8CJq; z2>h21L*A2V_n3JFi{}Lc|Gx)i=Op2-lFvKW0S~>*qgKo8Jxy{zu^rsnA@F!zq5|g~ z7>8}#Go1+1hMLK$A9K8e{fsdzf@%e+%mDUCf*-1lVOW*-sE>@55K*vQZ1GyLjd_7S zxT(vKYpB>F|O{(*glOAE)-*4i9>Fe1e7l}Uq7KNm-q$^N`O6M zdbE%V87L>Dp80)53Yw0evp#?=(8j11<~#QLEG8;*x1Y{G6C zJl%14OOe3isrC{_)$Mjjv~-yYeBwH!XG^NKsWcPXyz4-+#76~gcCFfYc^2A?sSODQ zvXMFhjFihVF5o%2Vjd4)6)%O0;FE$>;-@=#Zzw2Lfu5cM>Fu%T z^Y|}Z=ierQKq?>8MI0}{Q20ira&cf*$6S3pBl8JpljXvp%7jifmaHCT2j+<|(tnrF ztiimK611S>=ih?;w8Knw%k+x^y(OEY%M#Qi&xXTUYYX*JfY`~T)3MKs1pBMAA zmxl3yqcI#e!?$vu^4i63nnO!IwvxN{t|>fW-gDrp88;|N><(sq z-%OKtA~eCy5PcAND4XBXvUvf|AIF$IDz$Pb%x6`!D}|wjZ3hTtu%e_UuQyfpXS1%my6e^2xstkJX<>R>UUDAK`(av-6#a zN}%uh*Rj&CVeUVNiT`sN-1ka}?wRX9#t?mdjDKET|0FeyiD8|6KWr)fTB_s|hi=R6 zr~6>_ru%0;me2$X^B|mgs?f7*uLBq38lYWQ<{SD$GIc%gZf`NJqrgd(OM7 zoSQIQJMZzj@|X=faP|s1#kPSkd|s_}(Xj52d5`QYpdmE-v@P(&xK%mZfDk8>BXt0~ zI#`+6l(d84biaS;N8tN%7{zyXpua2A74ZNSo^9ZY|II4;`|CO1YVY30GgsaxK(BMq z9QKCd{M*vcQYqJ;?r#@i^hl^@iB3o}!Tc?F91aHl5pm0|tzG$LgGAf4b5Z}2!lZEv zh6OV`kp22MzZm`g+2%i1AxO#mgJ|Qw*in!5`6UKTbt)}IcFL|XGklbYodUUzK>a1x zA6#XYt!_wjf8&5_@t6zm8JMZ1n7MR*oOZ8%>g$nPmQ>WSlK8>=i*EBmk7F-$ZQaxj zY0Tcw-<+GILaAMGdbw7c7w*csEr!@fG1zeqrp75g`8X~<%O~eGi$P=8;*RchiRU^G zP!3%Tr?Lx}_#5l4{o?le59_3_UHiv21poU_g#Gs{J(qgUK_@CNEPc*%97CN-+Fg*U zqMf87=@t1zJ9&ED^=jNChmo!izeu~<82gm|VV}!lV)am@f+NUu%sRbdXY*9ji&`dnY; z*cSS*hnBxXcT%HA>;-ahQ@)R%q~{rRZL=Mj>u0lcuF17SbdHw%@s#UMhL*<~%@?h= zl;^{nN_#3P23Pqrp6g)N3(*0z#RrE~voZT?9yqv{n}d9m0G@IYymnYv+`jLOEnMI3 zjKzk9Su5UE6fPN-){wqQ!mRg5>pBkydu2D7PlT@%f0xm63RJb8_F~>UQP>XZllVmu z$4^h|Fh|d@C-(`-+*2vWkyppD2c8Do)NXdEeS&e!@f{f&Z^S>Ht81eX#H`U>3)5kL+ye?{2I}ds4 z9&m|9YwOx?GUNmk0>_@uBNIWRb@JBjnWE2vSbYqWhxI$E>M1CwcCa|AH@wg8&|3xi z3R7_>1@PQ4;*OWepwKd%7oqO$uH&}rV2o$gd_@I*##TT7O!pCD{{tqvJO9>KR|yl| zm8YDMEUW|i7Y35{9~$fy80-o>@gh0#kFRu0m8RRK*R0gBx2IA1?&7*U%6*$(_Pf~` z{o;9QvH^>Tpi8VhO{#-7Ks>1=2x>?+@Omf#$Ja_flla|^6yC+@->-*Wx^{-Ixx7BXG}Kk=JIlDSX34Bg;JvIQf7a~AdIIX>m`-rl96x@-JNc%F}oyXe^slS@z= zNpdhtls3(eg`bSAEwwI}7Sif#sn)#X_%7t8$~_U87ASG%h`{ib&S#$AN#VUN>Rv@L z#PaO80en^>!F^$g?bVF|Xg5iE6{m?&w0793K{=10qa7a( z1a>C{k7A$XC<)kcP_{@Aa~0(Z_E}t_arO1+d7fHF!lzz{cX_g)Cmt*6v@kV9`XDbs zpr-8Lsr{1=Y!doc-2*A1+l+N_~r8*#U)Dy3eqUol#EeDdyA#$#P- ze-6w8o~-ri{;QMKY=d&jOYg2cY6ah*mll5}QCw}1eSz}Imn>^mofH-wA;xd>;ez_z z?;fN{;I_|+z%V42)=I%*`_EnH@2O;dbPE3J3Gan*)522^V$V5 z_YEZVdxg9RBQ_0p4c7t>4Jpzb?++=_u%)qc8e}e_#Dn+hUaej_)GD|nU~b{k^Z9%7 zZOP|5#F463S?v8;UA|m-_mDbHm*l#l){_QgRJ+M1Taf6rk#iYSd7@86G}SXK7%I-f zsctrU4J2KD!_+_Z{_HME43+4yA-Rx5et}D(T&cF;S9I# zm8US9Dwdf^`H67hOgV9;R2AQ1y;R8`K79Gg1=ZaLUZ6Zdu&i&|F5cR`8J9CeFw>Ivie#X6t6NvIEduegrN00xa(IW~g=r^fp z$!;&3v&bLcQUbz?$ z)&#wNn+m)pSVj6Q%G=Cob?8cNl-{#dGbo+0Hy<%7!_fQ_^CznNHJvqcG)Ylz_hq|i z3k@5+iJM*aTcbgq6ri&CtkH{A8Ru>h(Ng#9ich5(K1Li>>FU2!w(6T9I@&nN<|a8i zO9M&!orN=32y}B=qko77HlqUIP8Tze#tA&#QP!=S!IZ}@3_Z$0nkS+%6w`QsN4?CRMuUj!%g7{W@|bx^gyxRF_WGmeuQ3 zZkKQdUVao*o?1;f{6^rLC)9g*mUCZOq`sw0^Szud_f}Nhq7WCy5w?`Dn*2l6CJG(< zx|(L`oAi_TqMh*2=E_>`vl_fE|ABW(nEgCn-*$UX(evyqxuKGoJ>DGph?D>GoM^#w z@&_W`0xL;OffV*5R+Hb!o~_KixxJIFax&@_GZK$Ft~sI&Tp;KOGd`a-Ww2pg6lJ%aU(b0{oI2@BOKH;z>ctH(C z{3yjA*IEpETP&d6ZH8<^YE@o*EhK|>nP^$eYG-)b{}ysk+r}nauw>R+vR%Q|bLUOw zwEbiAgv>{AsM2X?RmRlrrkKa$)#p{iziI!dkZe$~blbwr$q3#Y`5qMqeZzqzrNnFQ zeAD-8ZRn4Lzju))-1pMwd_k>ZBbgLMO>tCD)PocI5tlDle7E*ONsB~Y6U1!Jk~mh9 zM|8cY>dkHM@=6EWj&SjA{2k-69Kl4kS)P(-kEluM?W&9+0-+fKMKV*$$_)JKFP0T= zT-jfF`qK2;Rcdt(>+4Euv{l863c?9Ut```WQ%}yIYte>Q7Q&)~^EE7Wjaw84@9n_x z?&^nY+^{L8fRg<-Ob?6f411hvAHwm z`Dm5QwkWdFS0a+^OYov=t-DFwkeyazOvZ(dD-OIe4nG?bF|9ia$l{;UE*Js?9A1SU)s#MqipZ)q*&w{TU{Yn;Cr5AROZ9#a3;td znbjv)*XugWDOCGrp)RQ`ylG#9oLJMU#+9+~EAh~}b;Qojfn$3>2*F_j(ad_YT|+kX zct9dFsA9R9)^obk9COlW#@Z8(n5dYRUZc${jgZ_DA8wNLzEX&v17#O9nt!wdrt1}DCu_9D@<&tjoR;ignj+LJWL0FlHS9j$VC#@__{>hzw$Y8N z4&Mt(3Pa%tb+MdekQ4)}2iG*7_rtgwQamje-j}HiGD%{D%2~ttKj^z&QGDS{V#MqA zjrY_I1tov^k^BeIBj#FO_m|#pn;uc<2U_!A>#pZgpxTZd%0Na{VBI!pqr6eawM>Y-?&f3K1d**qV>^Wh)dUg| zblkw-jdT8nFHuV2AFL+-#VAhS;1`b!)AT;8e4(bMxX!+v)*b3AFKRBu7~T6Y8vB7} z*6`$diOTU~SZHhqDbsTznOBN_FXWf>-$+xN8#i(m-Lh1R15^6?KEBh@#s{>m!iz&3 zmA4@|kBvs>-i&BON~q^P)Nt+4k{mdPBdw4r&w;f)dmtomqdM_H!TI0{xP5)I&&vGA zEktSkqmnY7wD+@Dhh;WP%lZ@6Rt)XPrEkV)H#&?rV`Dx^@lhH=X$%!#{tXOLe}&<{ zTK*uT@GlnW-{=1~8-_{C@hV9F2aEJIU@^sizx=`7nCmM?^n7fYx7>v-NBitglf2Y; z!t8}lx=(1w&%IUOPzytv!V4sL48fQz=*u4DrYzBa`&r8#O9yR^S&qd%`bc#MzuRI1 zHA=YC{yk;$^djF7dTO|#G831xmMh)V^kMOE!E=F@$h{h8p`E^B-js*NZig4)Rw|M` z2n76_Sp;sz=WrVB4c*L{XYVM2JcY5{G7U2?w7HRLJ1W#8)_V?WJEpT+8)ViCB_wOc z!w1l2dsk7T=%b%K(k-~ai5YXK?T5Cl0fHo1dH6&^x`da>d!%!?(}y6_qWx0qT7~EH zZTrgVBUfYOB_{lp&J2uYaC*(bLm`c;k+U(~$`gEDo~84w^WP?v{h}Tb!TD$_E*+>@ zCC^&4=FwCH9M%)Q>`bS|og<%62##g3B3y=mpx!Nh4JM)rk0;sdkBVNz;#m3FHZgSuBlFe@zCo4Kpz5X$l$*OyUThlAC z6oS?~tAaAJ<`a`xk#2IUd+|YgL+W~VLx#>5(N?e>jDIkkAc8V-=k5?q%mt5-%g;et zuPBUpG8gExSqXEg_6kQQGI$Mhi8Y$Y-m9uis)?O5`msjD-a7mg?~?*tmA0eu@Fc8K zBusfmOizRBzZwOIG&bR3T;-0H(9G|idC^<)9q!u%#AELL!l4_U2b^CYEmJ=woP@5v zZwT*fSD4M1jkt8ozN(x~as-;Q%|cskq%+M03_v{ELM?sQ-pk+fDz?^Q#6QmQw`<@0 zNK~}akkYlwg>I1OOvJU#>Q5;#Bfg;{@h0UG?Y}PE9X_G);=*p1IE7JrEMV|$CoL}& z{e&zIF29-GB1-qULD^^`a?dNy(~oicdIE6esupD?Yk}l2F>vHA5dD1=D7cE-n|7zP z{*x@b9GC225Hy!JZTziMq8%9kfZ&&M+|S@ue8VsAc4*#x(;PqO=Z5Lt`n5!l3+=?j zGg~6fQ}1_{k5bSJ4G-`YnTH__{#$z^ZMcdFk{0t{4?^oUk$3<;CDlIN_TaJ#&J#?=IW<(SEvL`cE3d^avfC&}yj!DpSyUO+(OS|WpF`;@zQ zFm}0Nn&Iv2mKhL2zfeQUe(_ati_isdusPWK?%BXC1LTtkR(8d0Y`vLXV+g+M@AuWy zhce9fu78_tyTc~ZIU*e-j}2QE2!N83v?zrie18 zbMMn_q44coECBW}7q=a>mb6cz{ov>yiSmDZ{RVSgmSoMf<1=xeO<@?!t zoSbnoW)k#fCO~No^F^Xvj6CoN|73kTk00kG@#MEuz(RM(&_0at-P71K;}mBb%#~^1 zA9^Oy-Z~m>3tiuKq*j^04B`TeGkz(L;sD7GNowoZ}gx0(NkjDN+z|44;q z=M=kK)=sn!D2@f>`V#%S+`nX(Y{&hW_N46h42Zxkr_VqEhyb{!D zq=cgDDQut%^3{=TX>gUM8GnK^^giRx#@5>Ogg(@;0-KnnK5I3)8(ckdHh6HbImIMfNAwcXlpNVzf`yM4JsIXwKcCp#3-Red?4Xo~x)SX!&_6Kzj$0rv?(%>{>m^bkVu>qa z#@%qJVF@nWdo1y<8ic(S_x^VT6YVSK*arJ5{feAEv==l?kz-Un{rD#3o7rL9kM9z} z9Kb^W2#fsXDL?7o1uJbG7(_VRpZ$qvs++tgVGe4Bl^9W-zM(ebtq(M)i-U2yj= z|1YJ_sNp@}Y6}hDi32yWfVC8Xn}_Gxtxu40?FRu$M4mxFlp!l3Aos-73x*jg`zncu zVwWo0+ck*VEeMxgY#u@d+SjByoO^7E)-t-<6rRuaRE!|#WhW#b&H=FqIQ7rhpLW@f z-^}@N$8C9}$aP|>KR96c=yWZt?aCy&SO6&a3H>>qzHY#-`}%?6-hF!`-n z=9AT;O}znh-l%ID)B(0Tzl#n&k=a&)a1X81$=9_hWlNr^SvzBQaTWO~Xy?5u>>T7? z0V8a7Y-(Lk-cQ;Py6QUVWK{h|rc>!UbQKhi-n-UIZ;fmf4OMEjC%iQpaAr>UT74fC zjzjttB=4_|=rg=gAG!b+5y|bZWC~2_P)!2SjfE@_p=qn@%Dy^req|RPF|+> z1oO_GxlYB}`L6N!Fe9|cobx~WJ1@)HaLx;xB`!bFQ?Z-T4w_H#IhCGgEOQp^nc*}C zc&kfK@!vds2%%dh^=r5@`?D;yv!}Ly5yCJ)AhLOwa_w*Wk-s~0|1oXwKdg-ZMR%Gz zo|v9Uovv#rn)kq@zfZ3(C!{8d^@XtROR>D?Aod3sFJ?Ky<=ZyuV!jqQ+RKup@GX5A zNCKu}sB%2_fY`}RitX>TlKX#8W++Ji_fs|grq;=uGE8CU$}e0=b-UxRxFD^nAD5$| zK71lw=yXR`+(=>RMSedE>ES8TpA2Cw{^*>QA-&qCQ1-K{yX3X`Q~x^SGNpseaJkVP zgN!XJ(OO;ihcmZ2ofr2X4Qd@OHe%^{%rxYfYAwm|_6;s=%>m|-!q$bQhDtJae!$2utZk@7S znQqwQ5}QZgTbJT^LDN&h7(anpU1euzI zz2ie|#db0}$FE0O=8H_7KQbITOU?E6MUmBpAd;Iw`+9#SjZk|t!F2n668>h3^9Qb*9f-?VwWpb9` zGOaDt7dFJ6(?EuL*pQ>< z?v$?hk3v}Tf=WkfhuoE0ovbUks#o9gl~!1VrFsMrwW68)mCI@8W~d?L_MoJZkyV#y z*4mPGewY&*ooyAb;#i3$+hc?ASND6`fLHdHBwrgJ+|lvO(rDk9O~+zlWroOV~LP^-Z0QGw4imn4yC&p2X`<0HmBr2R(bAAi(vvgJ(D^iW zPSAzi&zRyk>yG-iS=vw!d4suTiNcUgw1h_t1*cO+DFr&Hg{@sm^2N8bCmBzk6o<;M z*efe$%||$VYI=9yX>PHom`$KQcv_Y}K$*>+pV>C6JbjD%YbX zI9yfs!wm-Fh%U3mDxJM1rC1c-%G2K+JEn;2XZUg)OxkdiJ z65y@nVsa?8CUl)Qz}@8dJrNE@T~-eK=kEGQYRF3#6%@2d!V5EL?S<(Hqi?gAL%lmC z`LhAP1i+dxi(m&hQ-J$u5vl;?Rk?RN^h6GO=A;(B6HJF8N3k{9_JgjVinL$hhBInizCZQ=&fRY*wiOrF%_B4rT5=)gJCqKQpr2CKk96_Vz4n zF=rB#wBFW%B*CO%8xNCCmrxUf17?d@&Duq!uT7YU-Y~}G+5N3Q+$|yV;(t!y00{`7 z*&~1;sU6Vnx7FML9|_*yHno}dpS)109QfS7F)qmglul>O62Ez3yfGsH@5_j_Fp|9y zj%w@p%`|{C2iUUY0n-{6Q8gh>WM3%Dce0A{1t+Nf}2if zMfy*t+tqaB2Iw7wJmg|>WQG%4{d#t?WgBSA^!bU4Y@6{2~mi6Eh1VQc1?Vdn3 zflvUe5S>;~*ol)Na!fcV5Md25&vcyEW!U$c4aZnxDolh_PzFh>4XE3!bKa0y>tdSY znspTf7qZ{lA$QEWSU3=Ize!grWUj(QQ3X|w1m{b4;78;kMSing=0JR#qPl6$jFu7v zV~_=%VP+dVb2h7LE!hNI>leH$!Zb1DSxDmP&>4Qwyfm2Lhj!t_(5s-hk@D~ZP4pRl zKzR`lrvtd`bP=9Y%$R%^+wp)2a!me!NU3!Ke6+dL4a0=-flSp@A|%Xe$hCx=jCSVFeWurL_GyG}**L+MO>u>ADEV#S1Ne@@+g+$nx3^Txj1 zkEIGP^g@pPY!jpyV~By;G#uFS=&x6Jd7^Xf0v9)Cdp|%rt&2jE!Zv5TRXk(iB>+Vm z?dn4{N5tvU!2o<@V@e58?njWE#E&!>E>tFm%L3il2j*^JIc%6<%;gYIOo+Dk=S*LA zdcX@5Y9Stlp2{FbO7Z@W``tivV*HRdBC)Xu;34}C_bv^aE!_8h6o8P07fs1;o9<@ipRx&|r=OzW z2R^WmS;dFrs@-h?$+~Te!I^`zpio?MMf8P@wuime$ssqxw$HA4tivlS6OfOf3qX}| zbMvwJNyYwqxy{(J`S9Bh_q$iSYgOdAGzVb-UzXrv^B9X>?^&2Ny6DPAm@%sSZP3S{ zTc(}H=&Ai{#p{`Yyr9+G%|NI>%bnWiKhFcUZ0F>Ze?+zDc>zuZ3xKq{kD z5wR?xp%)u$^xV?`aGfyr+o3)eA&{mzgtgaT)BWa(n#z#Prx-+7?R`U0-}D(24TcTX zuw~Y)-N@xL3j~<$m5C zfck&D-|Nrbh|UgqtLhytMEN>&;43f0)Q9#wq5;Km)Zd6Y=y(!5NWbkh*yucEjz8`E zhKxV6_L5)DahQF+ivB5;BT<9#f{lT`1cj-g__sXB_~3?8;P|u!M9m-)E{rgdWJIsx^ZSt3X&{+MZQbMs0VM=0DrEgk zJpq8A)i3=FB@B_)L5M-p&qJc&Ak?rIaOg#)?KHwSa^NANsE|eRM&R!RVG2+oHXu}g ztl8WJ5;t>N^%YQTznduSWV1DJ3lGFPNE&FlYrWwQ z>uyhfWu;gt1VO)opWi!KpAK9+=>@GMn}r*1kc>WSFWAqpp{j}oKzsrG z2f6|f1@xjORNn$TnGK(VtNHFIJOJ~`3CfSL)}5tFhD)Qe=e#G@#EqNMSwcxw zP|e7^uLwBMS9EAhaK*THf*M2v5!$}yi%s&~iyl9$3w=DJdaYsbdV6bIOlarCFDfq8 zw>|jJ#h&d~{Ek!ci46|X63`D&`&wrou27R?ooU^gqxe_I3f$(8(yX260ciE(kLNCK^P8hGGi`=o;^doKd`Xz^4y-8gd1+yh+?0Hl=kpo5Yr`_NJ0PEFgKtWGNRT~ z_g>m5KdN|8P$FQDo!i;NTf(<8yEdaw72kP)^~T>oriM|gqB0r;5+j?1kdNkc2gbDg z8~D0a{~y-g0xFJXUDyo-2@oK`-2x%F`{3^GFi0S{yADo*J3)dwL4vym2s#jgyAE!H z>z!nO`<#9L{oQ-cx$9>w2By2ZyXuvyx4Wy}=b^z`|K_}><)|$++ySr zjSsW-QDU=&gIGd%l~E#PyS>a#tk0bN!r^M^Qluoy#raRBCW~2`UZILs>uF*M;dKbT z&MvTqAqn@$ZWG#x#7_O#l-db0k8H(aNFYSF(0k#txH3ZCHfFC81NpSJlJt7%PiVM{ zzE~X2bNu+|=mm=%z2B43UQy^oSsY~DKR~xe%Nuvm_m;rkRawO}Sr^Ae%p80E-DNb; z5)6_~+TQDZ3k$amuf>NBrQbct9~9)@eUMW^{Xl_!ApgJ0(EgeGTfn)C+wh@A|C<99 zPv;0^p>XYk`}lpc&p{gQ#Rs6t<9Gp#BKvu}c>=<77b&2~u02jrIBx6)Qt|4w4F7^>#5isSsPF3eKruFR07y*~P24X1wW#Y0<%G2|v*n#c8x%d8a1N?tHp=0~s;KicY={$gfaRXu9XbZVKGI+=AO;HM{At%9GQ}Wk$39(FBUl*t5 z=8knP)3+vLndO<^;(pAn%an7{$P*39eHgOQbiOp02_u$}zdPDm z6%)e1@G!n}Up#Cb$0^IjQ#F-(syh@8_q*};e_~2ORpHR~l&t7->gQB6+F%VI>#AWC zt_tsr#Wqm+su}3s78B5aig}0?ZT@!>`I~GA?5zKs1MuHJ7O2qCkKO0S@Xt1iZ{+Wi z7-+ds%fW1&CCiEIV{Ra=L!1|~#8NbkdarB;QBVdwBv6?Tr)2`iCCR>`Nvcs_#78Y| zo+M_d*(sPy8m4AE+#Ep!RA&LHCLDxe6>3{1d659-j=Md^(QWdFCD0me+7^cA5mpfh z)Z}vo#LLk86-v*IsZ6Y^Nfn+;7xiP<8|pZFAJ^^>liKazp|`J)4sH~9Z+ofevRp>r zT4#TTJTJhx!O0n$TDsdBpK?ok>p!cP(VXkFr1imv++7H!2}q`BH^TG{rzt^so-Qw9~zm=8|$MHWy$-Ss6E`h4B-Mupizp zhFYuZkHPdAN?BS`9+ys~yD?XF^B0z%wC8tD!OL&uNn~_*hG=Z-Tt$jsil-(`%0#%EpR(ID4cb0NSVX zYB*B%uISx4xP}k7h(6Avvt3*wiOyRlO;7UsRldX8?_E|VZ%UC#hrWWgcL%GU!!_kU zx5BrVZWWnkV*#f=TNCm5XuBwJ$0AOCNr{xlU-A&{7)n`Vs8rNy$)|3%GCGnTe^48O zAQ4+KoN#dV^K#M_9^VLl)@Ls^SilGo>N_R)x%b!sfK)^E1>T2of-x^cBTp92p^M<> zrcI}Z9U84XuAQp&xjx}NQh9Y}wjf>!;58GABbY&yl){OVhNp*QNk-nj0GGyEl3^L$ z4*LLY*iSlA5#K!YRYGgZyW3+=bs z6Zk_SI4YTUl#BX^EH8z4Tk>e(HI5tu0xZYh5EP=4M?cBW5@q=b^R2vJBmI zaig2w$q&5eMt8O-T(^qSS{;1P3QEhvvjLI1OQDp`&D}lTvva#G5BtNooc61L}^F+wbGqW907IDevc8VT_L zkziSu3J@B&EvVz?%q7Bw4E1!9>&`bX)n$O}0v9D?`+AJmh;K&;Ux#xm2@P#ELTBqn zL*%MaH?jE!m=|qBj?{ypq&)G*5IstS`6Wo+w%9s(i<{L2x1X$gH zZp(NQ!3%DhNRx}5KhkV+SG#p#$}FH4pMD_nyQ|}8m#=^iB0LitzIl)!Qb%A6|H+Rq zsux_Ye4-U_`8|REHq1<(u5x!HJ)0F;qg0D;+*XYIx-C;AiCu-m>XMy$B?T_$2ifU~ zG*L0ip@@8UQ~_Wl??S%D|Hm;=&94SELa5Ktb`$z)sH|&5qOl}}oP)BD6Rikg-1@L_ z4kI=zgDG{={rRlAp9i|SoVZfiCf3Z=OHn$N61`_K-SN*vjMzgJ-pO^1@nhY;3eizb z`j8)j_&j`zJomA++qYQ4T>gF}X=No&CHTE<(g`-OY^bBD8)ILA;msPAOQ^3%;=T;F zTBPXlP83w|mzf|O)J5&}USqv7qwl%f=(D&O6zi?df*KKdqhu`TT~`8nFXb(sXyVO#h?s0{))<8si7IxC_^ym)7xouR1+O!TXfXQ zt76wGX#-|U`Wi{*)r{le5K(XCy^D~!(1w3I)!kFey%&vv#Zu)Et_sj2(qX8IwH~#2 z`CZL$FCu@)zV01$BhMa^(VLegS+D#i{PM~_q5i5@xs|)pX8hzYZX@#SnFCb2?lDPd zep75p8%b10PjZg@g#8*5^-;Jck~qNU;-Wg0^V zAI-_t#eYUwkM8uC)5_T+S&Gi-Ip`1@PpDrsvI%6TodCt)W7*3ufAX>y|JjLB@h zxTGSS)CE#A#Sn{^%14YHyW(>n@ExAt2%R_BqSDViW?HG4Z~mloKGhYMbi$phVHJbg zEf*x!V%|C$9nqb)$ow*jp(SF$ZHc*f8`Z=+L*lbtw)`CwlCtSq4hqGgh}u-*KZNHq zOyS=p{^sqhI8ZyI&R7q5iyd2UwekYhusSohp67z(bWt! zGuXSsH+x{-;^u%}Vt69VAYDec_DbTXXP7_;8Bwow0Sija7S)S^(JBDEL7+BrJ4M4J z1-8m3lWr{p1(9=^t3?4@{~o!f{)js;zID0@;KYb?ve;5McTmioGz_)IW~@@^x>A_u zOUA9}Vo=iJ3w`I}yX@jYmx>ku^{ch4NkYMo*3H4|0vfK!eW1L;y8WaeeF-(Ddi#T6 zIs4MGYfJU`UtG`?*=om=03!KNzeKC9k3_0dA4X_Jo+F$juUR9$=znq=nKHdC&IY0= z$cgbxDyeXb#fm>!C~l}>y|sPGb-#&;i0F->=D|%{Md^?AW7+S<@k*HTU9-jCg&hCu zY(xK72qwHduoob|XZro?3lJ`j|M8-(!zFb61sK7^*lze|0kM3_gq>W@Lb|&Wgj+wM zh|&=LbJ#ju_mbH7(-UptCZ_FNfH~uW^deZ&E$)+b=h$e*d)`5YdRrWMoQV$`|GxEE+ z3^xL0_#?ht3s+V{x)Nh~Q$cmeoUE;C0d{yi-;Oovc-XdYR78Hjyx_UAR*|~BW~j$(f(Lga^9}Du+fbje{p7g+JFxUb47qd@@Z4f{Hxhdyd|$+a+-}9 zgLA`L9;GlAh{+{nX_RbVD)%?eA)pF5qj}4A+T?M>Whcv{R@eCrG0~?7V!*IUr5*Z8#-??%7x0Lk%O*ZlWGwC}lvG)G@&4DC|1IgRg(_Zbt9|;y&&(9a<6ySIS zh{J!gGBDgaChQxC2X_;vPuSb}l+;H_x536(?za}#;lTb!e_+3V)_D1Dn5M+%tEEUHJ43Xe^$2d0u<+u4O`_un8*DDmum zPMy%#mm!3)Jifv-{wK{m6#Wdqs|q7@U;|Iecj<}EluC)L668*O$yb1x?LRK_` z_9QS!F7u?M!ypHk();Lptlr?i+U>dTmu@RTEm%=F zCfxtj+X%fP>cRde8%us;L(;$*wAg4w-F2@y@p8w#c4S-&;rP;C@XitB$Nb>M z6GQyl-dQ|5;@vm(!R#=6sB$fGjn=LijQv^uMYx!huhEy*VPaW%jql_Jdg##(`yhvM zWqCD#U`U_81mINzv4kOENGc&4VxHU0Qd#-$v%D&!A+E{4t&UG0#!E=|9Mk@Ae!tdT z+YK{3?Jk$AyEND%(v<~1=dIRA7>s}b-0qO?pW#RJa28qf;6(?GjuZ+=fMgsrKN;@b zP`6w#@-;sk6$HKK+um))yjcx^@f8KLL5aTi3j>`(@|V*^N6lVJXY}?=&X1X5V=#A> z1dIFe3+ypN+p;hx!8tSdE5DI#d5|?L*AZzd=&L)SIEYZ- z&sXCL0WH1$lhgZw0ha@?j}6RF+eRJ%D^PVj9{?5@fG_BP>nsSSy>vKflgrAl*dNWF z8RhW=l7i=KuJ0oO~;G}!vLV2Ej|eC8BSJi-Yt;}gWabN zna=NVvb;KGT|mhG&d#*5S(T{(GAlv|=Gh2TG7G$ty4S&013dFoeMp1gp7|~pyMV1C znY{p=FykrT@ba9GCmr^srSt1ZhS^J11SDQN)T(%N85d>ES0~FWYt_rO*zieRW9V>2 zMk(*Y?0Aof{@i!M4b8Jd7e?$W9l7?uI_4M>e){m~jt9fL9I#>u?rUtb@DnKHk%cj8 z%)QA3sH~>NC6AR|cwwj%?LDvifC=N+6_|{3gdM0H!4I^V5I*l^x)4a87g^x23vO0E z=^raQ*ffCsg~@@*7d#2<4uPcmjA{2=8Q4w&RusTX$5qES=duj4?i=A%QmHUYlL0?4 z%$sSOj*Lw*d)@a_e7RO-*tI3W>wtm$0<-eFlhbN~vxNbk-210)7Ax89wls%Q(?#K_ zyRqw>(9BkeV`zc+;9ByhOdtCXy(em*N-NM&=OeK25db*+5d(>xmdnmoI{}mmLk;A?h|LVEUYnXG z$2}1<``~RD`ezinqtv=-#smf2-O$}^_B_G}shiCwS#lkqjj@9}o&(UfS6Wu2gO`tR zC?r5ImNfo52YAh!OxU=NISYBf>BKh?Favub5x*7y$=LK-`^{HCdvyzwH@_!Y;_&}B zbpslG0chffp~PmopgD?vgWV`!zxmz$GRuP?Q%*b+#-Dv{JQzLj)d58|sS^u%C%Th! z(hzgP2Z@VE$GUX$0%9EIZ^?#i6!L4+HcNe)<{|Uh(5$;h5g2dEh7G3Z|L9Nt-~WTe z0b&hQl3-*16+i*jyZoaA%1VXEv9!Wak!Y6y%gKKkmaCc|-3;-`r;8vN0MuaFrKS7F z067d*sjzeD^3TreCVeg%W*RVH_6`$_@S+2@STPTR4T-_Ohr}7^?u=1ihHPtmrjV6e z>Ta}CdfKDI#GQR!yon_j2#MuMd~lB%M!DoSgZcJlL3ZWOUKXv|>|&nZ-)Ads;tXI3qGiyw?v^e>|k1{k_ZLl4tjw+oW%QzkuFTX=qL{|pxLoMc6E z|4YID2oq9|{^(K)ixjepzI6JqV=c4^UY)IIKYcioPVzaeHAbxNxN!!u)^BbFA^l z9Sm>upK@U7f?n^_{ZY`WGmPE?aNQ5Ri-ysBM8I?F*RZkSuQoP2gRvnKF5#@c1~8

*$BF?LX8Xi7khT2?SXgiI;Tde`!pK5i_Ussej<*TRfT4aFiI8zY zVRARpOi1_I1Eu$JGp8=-%^vcvzfk zp~W-}-En+<{4>kN4jXrD2Ks%acgS5}3=GA=kJS+b{|Q)O6LpIT7l}esW&5V`BU+Na z5r9WG56o?fpsmFZe(kO)ti$1Pnm1D4Glz99A@vm^gz|eZDHuElr;fqKvgR1r0I=QM zf9&y}g88@qp!HbVost*Ci;L`9-|YX$D(xfb6E!0TZ%R7bq^ z=t<&0pI=6gr_z7M?)+iRRllI`kb8JZcCqA^%-)iaV{rg}NXJxa>^{rCT{A})Te{7Y%1tTF5vH>}h6RHU z5AqbE6VQt%Ritn=84IFvw!s!GLr*pCa^Tr{MRN@e$qRFYdeHmE0~SxU4#nh-;0(Qf zMoq??m;WS(#`8gT^=do@CMmcE!;zo7Pevvw+y;Sl+p=-A$W4`->LB&iWTD&Kr!ni}W@S5w`KCY<%|NR)a>0OHlfRJ;*WV;WVQ1s|i_iMs zza?nbm4~rZUUytoy{Zn64`vH@P!oMCg7*WyRfn3`^*P_ji0J$Bg`QcpD(Wj+$V~Bd zf&`b{)$_qzlLY#NzN)eg`WjGM^X&;XLk&D`!qCAsw>qaf!kZBd_u5gRt(k$|+ygRg znY@4HQGUo#21nhgSLu%razVM6;RysB>M9r;$I5nW*8zeaHL#qt@$4> zU@}mi)xco4>g^u#<2P8LB4Rt$-|JIX(TPaz$22aLAuFAPIy?j86sp3oA_OFtXAxSi1`C{cchZZ9`Y9@?7hH<4(^uLyd zkv5M_hvbNToY8r&eP zsw#!T1xJ84lF}X@tEzV7*aO2|+tUjq)e+JwI4z^;SRH_`x3xkI=Ze!R`kS7ZYI0{N|O|78Sxd072t< z`T>)@Gg46TBYVclu0JsUBIozxPo)wJKVW#$BSLGv4)e=)R%9xb`Dx`1h4PM<7ZNh)0viGk{l+cUVql>H5~d6-T;4{KYI7*BY@6cSV?bh0R> zn@skv@Tgv9*6|)9A$STODAJiuMg{XEt&@okLSDlQ5>sl0L~qCUq)B^013t6G)@;X5 z0~S9czuads`#^9fNKjhFaBO>&DkQXKV_}wI=i_KNuowUGoz_#aTv|VQG(LhAMd6Mk z{_3mGBBj{M&4-zI7+9Z$j#QpWnT8aO*y=f6%)ai&eO1|oIYMk7+5noR z)tZ_KKGAY6eCnrqqUD>kV5ZnylO_0bN5RCSiV${Y=Xo}Q@E+ukbZVSi1khl%cMNIF z>#}}(uE3RShb2p&gSP3`GW3XEKaH^53OU?A_Wtr-n*`7O4O(uZ+BvC)UJ^o zCJctWZ^;JOh$fjI8TQ%Iqsj6GOS@F?bu0+gMbJ3FCpPMhmqwISrO$52xWA2deqCbe zCTOMj56|peCpNJ zo%CeDd};X;YsJ_ISM=O8=&Zbl)k|32b{ex2$(F*tlfv?SMvbg1xzxDOueyjLyI{E#N z<0a!sR>$Ax zoS`M?`%C2ThA6^p4Y_w~NdmhPAe;W=$&bT;xb3ee0jgJ1`@>V2vTOTEO{b?c%*+Aco9Eeb-C8G9URmG>gi&Br}KEV4ctp~b5yE?4zvW3$T_dr zmDMZ=w{R!vWpukGMSGQ7#b*TEOWRb%8~CuZQq`Bs8Gp6wA<{MlJoVxW9R34YCsV*cebtRa#J?mjaj>);FQl7Vm(73MNFJb>a0DTC8u%y*C#hPr3f1FfWj zqEqJ;U|bccH(|JzrE%MR`eg41lJC8Yy1s#08El@i%l&fn8w&JSMO|JKCw!H23@q_q z!u|okJM8L}o4;CczPE67eWqSy{6o(s@JrR_b^^@RNBlZBudRQNUU=C59ht6dYOU^S zhsvTv!Nv+($}eT*?BYtn%gc$%@>lG~#>d0=KV2FzrEBOw+=%5jQ>Bw5;&i9NJh&Fu zi&iWtwAWwhvp?&$`s)R%tg0zZLz15Jm(7_K4u*vZy>A}1ct?`tA*AsZvSVqO7|nxs zJ$PeIug~Mrx_hzx8)TSqnAdj;DO_o4Bt`V8v=OF@S;VATl@ZmC3Rh2+#C#YLdTEA2 zYeT|bxvI3iCJSGg94dY7Y^NP zJ=Pp*#sD^@n9oTrYZal-aNEoeO<(BhtF2L9s3C+gN-1@zsC0|Q<&-n{p3JXPO5m29 z7z92?Kpc1>T8Ah$f)RuxBzsU865l6XXROv;B75T950Gu5ASM(;sM12jghQa-ks?c? zj*=|!{oop3B7#W7TGK@Y_dG&Up3A47SI#}iK1sBxX-Mrh1!omYy-O)Df0fB~HxB_j z{N^!JVqLe?#GZ zo@se1{i#74)MX^XX?WQMWZgB#ED5w@@iB(;L4OWeeLTbTyoyXVLEk!>b8Qk%$Cs9D zzB8+R2O{_Tj6XAc_povIn*Y{8s{M>Dg+G17!Wzr3UHd5d!z8qJ%~GqdsBCv;qTK0e z;#}ORU^t>O;%n;XB2bt|AR4ayCmI%c#!}V9_#N#zgZ^&XC`ryRoaj*fMHHlqZZGj9 z`}{}u1{;JK-ikT6e}>#-SoBI6(k4g-aZZeVn?+RN7_4)z>|SgaK&321if!k zjz;FuXKQBceJavZDS0-NjKyGLb?qkLjivXcwX0UNl~IcFaK#zhBLzqMFHMoDd^5!# znyBnAtTx@)iKa3WOz@vA2$PSnnKg4{wK0&}+iGAZVTMCbUDb|Kk$v=U?`BSYA0JoD zZ=Ae^gn5Ehn+Y*Mf}k^tr(V$JK;o(Gqp=6%VerZ!tKEnD@y_vn=NuB7?LLOF?4ORV zq?A4CzI`lY2SP?aDh$aF00h|eBll`yJF)(NEG2t>0}IWm*dk0LF+?lVQjgP0YMaLSZWW5VJ@)l`%GdAaT6wQ0%0`H}ek>U~;;Rtn z67eu^E0NTsP)1TVDyde+eN7(kYvU4oe;~J`sW~BGiC)}?uvefE-6RurD)t7$tzm5( zK&H#4ls*8($N(isW%h16ELG?hrpU}|!Cxs;TXC7G4adFqnd~Vw|G~K7n8X_OJQn+N zTC+ONPW)u$)G5u^ajFgmCb2_H-F_?ozTNTMo&x()xPf6A8ycDgO|=3Jx@H(245@C4CE+`K|&y}Zf>0nHQQc?aePZ?(j z>~j%~0lru)F1Vc59vFBQysHVf6;;=t{?YM%^JwdxL*)Hw>Vy2EYdwje|Lwxb%}xN) z-GW1c@1C$1+w7>olFi;~^`2)10&=9_TX!u})`L!eK`w`W=ms>9P!U40Jft!>0IZ_Quco5?58}5x)!GlE~3>>8lEc^MA zo*xi7iK@fbcOmLjB90`Sa`UXe`+)sX&ydOVy}5m*hq&9}qHn>YZ-x#-p+HKXb!3J> z?Y`dAZoMaig+afUnM3czof>M z$9wq!A45ueizD+)4_=4U^?3=2Bc%O$lriU&WK1c^6|PmWVW#(l5R_2iv+;EsoE7v+?x%qD+SgLEfo+kbXB#zKdNji9h*@1Jyx>L}MPaAS0KD(8w@Pkt(yGI*I(HXWc-6xEjjS|p_j4G@r+#xx1`iSE)DHm@il9CK}|B7`tQ!-py z5UqQf0X_-((=gWF{TQ+&>li4rWtw|rN;^$&mz0uYAB!MPalQv!xp(pe*B;3^xmV~o zn(JnDzLMN@c=hTfUV9e%Q`dzl{~L7NGNFr- z4D)zYrwwQDL()?7e0tm?P<4@aI+ z6L{uUR=gn4_;_w!&y;~w+M7$#^Mm2OT8*7a(Xc^Zd6(=&f1Q@(W6Sdj9ldNvy}SI= zxrzJcf;j5@xqC-imaY^(N=Z@n6tMXL7dwkKqg?^PTF$B13-Q&TymAPW6x)^OF4Ja& zs(W(ZOdDA=tmSZUUD!al8RF(;;bR(;q&l`WFVOT7737lk*5o`2=inM&?N;8ZAfI&q zT1tp}qzY82^u%aHd7@W8Dv!76Mo&X8UJY|!10cRmLPWxR#zr}>cq{HGl_j5Hx2wtRqX$FYrhrI`G^A|Ro;sO^fz}xZtD0cRj|Nl`fSVni zSd^L7J@_!iv%RAg58+wBpj$$`SmdJpJL(r^t52EOrL{Mj!yi?#q@)=o-eGf^1}$~n z##>2fw!KHXZiSz!_%a)WELGn9Tx{w583Nu;`x=~NH&p=Evp76uBX)I)JZ#Z|ljbA} z63CqF)~g@zuWxL)0JSsOyXT!2so6DvB{vySYWUnn*>%gri6rL0`)_0zKXlh2xnI$l zbwAD#sPg-{E$hjGcfTPsVqXUMzGDK1CqMs^FJbl4{VFVgJV`x6d}r5z=X*0BiDCX3Q%dD*0`R!A>HbjkU91(oUWFGMaI8&Vu&+DPJiig~Dz; z(C8VTM~VlqP|s&>o>*JI7~%cAeXC z+drH>semc#4nI;sc%Wj3yX;_%;-b33;4i-6*`ve4{yo;<^aQkot+}?x+w~yW5$Bw@j^L-aM9M_>Qw<*_cSKj4ave)iojAT@v+e6ISg&r4?b2u zR38VMbGROwUJjZVm_bQSZM2X2>w=eq3O2a!Z2hhFJc>A(51w{J2jwF#K@^o0NUf}2 z2=MVu*POEK-hQg`hJ={NmOQvGX&n0V&MJ~+|Cn~!bVhfReV6@&V$PALNkwgrs>`>`H2Tstv=@Y)M{5!t2F%XIKQ1;4W=#ErVHG3ol-Up zy}tL(xG3VA1yT%av#%qA8of;r&9rWg?Q|rMc@-Rm_|C)X`f@USBl=sOobR0uE8-`O zNAa`sAaXelU>bOw%__y9H#opSfe}kJ>*yz`@RMyPM#EWLRH^oW?3k3^x5cdJv9K&4 zEDJ}3W$5FwG4j@}zLkL{TH^FnWM#y~QgrhGu1yg6x?pdP`)rWt)1v{AQDDz6S7vFC zVEpPLB;5_Kuxe0!jciSrcJrAvoF2XP*aqpwc3jK$tAc^YxZB6L7{#|C+LUjW%_D>sceu)#CNL^+71VWh z)-QAYvBEaAh5(12S3a6f3o*5176ij`4X4JxwiA7}LP9loV;p7bkoh@kBenn3(F=g8 zrCcpM(BsbLOolLRjWb`U#CtEehmYD4gT;N*D{k$$u_x+ zm$L-bA7woDPV`ekKJ%t1suLE(z-5Mu4W2jEl;0gTAG3r|iaXfMw6wfRpbY7nID*}B ze?PkDOdMqMk5p-pkl*+eIouUw?}ojZ{_~)FZ&-d!2d?Rjup7Z^a;Zg(#upPusv-G$ zk#(KVUyCqFsT`N%v5v>Flz19_i5>XKr7`c0#tr{PVkP5Do@K`?C^K2lB0GPImUpb7rY+z3LTeSsCqaYHca4Q$T^WtU zNq9|bU^St!a@2s5<%!#bs0)Lk|k(>2ja?AYBGADAd|6c${a!~N{{GLVnFHrw?pkQ#pMmXJ3b{naG=@ibMt$ZJutZ%l`?AgZNt|T=>8V^U#p#}D?lwI> z>uLzdkY=tm{UttT>y1Ffzhtn0AtBZ&P2}yx-SbGE6jS%V;`X4iTyg+C*Fy2{2wB=r zdLkd9@ZC;eTiZ&4@nPE~N!w1&4rmyicrLVOk$V68mV^X4SI zGc_m`xhj;W^Q7ju9;YqS)2g+d+2@$~XW?nFXbr#7)$Rr7j%DG@POsuUhv8m3xkRD^ zKJd=Q#BA2I%0&3E-16*h^NxZD^f)V^M0VVFdSQiBbMn(BUCSRfV}(&JGlBYvtWOY6 zc(b9?AGn}nQ$2m5&FMjWA?Pj3)0Q=6WCgl-Q~Xf>gX_7?@wMaq)`w}M)9Y&T`tS0d z59ZVIDeX8clfZHVPp@;VHbx(&PQy>I#wM#vH+Z7J8m$>RgGvAlYeQfG0pVcHEt``+ zVTaIUw8I0#V85^et*NW#Bk~QNhu3w)ZRwY&`@l$^@^?V3KZlsW_HOmbpZ1_tiMjnb zN7W_lyudabmiNGfUip)nE!+lgy5CaQmi}`@qb@Kg6{RY%iwXJuPOsDeXH>IDct^SKQxp+IfE2>)=$dMAG_&Ogg!cV z@He-sRSz{5>rmdxUkOq;1HPkG+4YLIk$BPpmDOQ9B;!9bRsYbhWy-s&1NB zW4Sowx^={UQb_eaMn)_5OgksKoa~5r+o{}vnWg39MCC#UW^hhl(@o-~N%I!0_-Ryh zXS01#XZ=e)!rMxI%t)Q(4pMcvvbum$f$86SZITtC0i)bfqo87QprJ=f!rsq31_sjhST zrxICD#}{-!q3uYathSb?Ll}kmKH9qNS163|6Wi`P8oO7cJ#VC~`|!s+&x-hyy{Deq zrcuqu`o=QHq(E#!fToz$Eii-YPaC1mnaPXjvbJB6AjK+jV0%uduP0yIO70}^)Kloo zudvkEjWt)mQ_qB_c-aM3e>kpZC_T)-+FFi&>*==zrOLB(5C5)Q1I!!2f2lw-_#duG zf9zG^`^wS{7^F=H940T&`t-*je{1vxs1}>Ls+pI%zV-dLb^qlZQ`xw1%RRwS_%c-dnL65&W|g{FJU_ z2OR$N(!vu@u>*4(E&nRmY)}d$YX?NLh1$DJ7)^?stt6yQAJ8VQNbT%SXfdGeWoT zIrvaGSWzsNJ{t?t?%_o37yX8GGuUp{6lyFWXEF)oRB?l8&`c3k=uJpz7_nag!pq`|KWJK=(xk&S*{tk=uhZf}p;JP_9Xv_aO5;W1S3v|BiD@IHJt zEN1gdzUHT5m7%4g*OsYu)l7GI|84u_PAP`x=(IDB-gDprs6ydr==PD87~~{>*7C<` zK3CzLQE7Tb3oU*ZhKAnI1G(G(a+S|jm+N021`ckqTxR*K|8f)PRNIF5kUi03+9b<-$TZjbIJO}dqsNHz<^gFbf1$(L@{|SWn|BOPvU?OiL@xhgi z<>$gR72fzA4^jPIy6u!;QsvPGm?+cv5)>LXsIldLdYCptcmI9bB9ALgwI&gq@|GR1 zgyTl5Co%p0s_6Q%z)fy83!Oud*scmhzaqwCm`xNaLVLLgs`W>2w@9ClJeo~EKd(*d}*&OFv27m5+f0PHV3JyR2 zhaNlCp(-eOt++||At9Nx@S_mxq|vT}3m>GaNlBRg_AXO_J64X@u)r;`7gQ_!NjL?ZwvSS@IR~5nMPov1o6xV`C*VklvjmdOlsQV zr+a6z;CszYI~+lhLL127&+JF%bfGLExtE)3qh78TR4t$z`W7F5Qa?W~!^rma!d8po z+JRjb4q%lsi0ot?lu)n3svAK{{^ZW@L&73pQJH#n;CkdnjaBtltkZDA`~oM1?Bs)C zvO6&>QnT*1@t&5O`2b;Qi`pu`CwV<46_11rp3+6*|Ami<%wYkYztH7XBv) zY;foa4zyJEceBF&gy8?@YI=^pqVWI0oH?ebZ|E=xn0Q)bpiVHZ zM5u0DJQk=>VZWQXb;8x0s-MUuUG8EB{y+qPYlnFp$3%x-)tcb5VluxN`N(o5?Q--t z_jSB~vmBm-9k%s%nNIiyO zBC8=|9eaPP;h~whgj0BN>mBcTuIk~ZIm$vVba&5?&y8Ci#hMXb9IYOuU;C}itgtR4 zd}-UL&!&606(oFr0j+A14Rh}Px|xB6EC}H7xKbN3oxym&B9t}IAh4czgz_X!k{6LH zlzGW{fB1!4a;fpcNnUv#PTDSG08TE3DL9 z8+%!n?T}1OJrKNpmE|zcsEiqDOV*1e_W_yrMzPrOejY+c>@ns}Ev?}uOS^4V7R}H0 z`7Is{1%x)HI*wL21%cpo_=%;{yaywW`x*nKp;l_ETtFJ#@%S%kweKfZB+rTUO|51+ z>^xMxV}Y6EamMjjjLl{Stum)hp|qoAyANf)1)C#jdyp5F*T>B?R1#LZ=9*J8f=?ix zMRym{=vP?z3nXsj>kCd7EPbH1L#3cEQ9^J-#r+X1YHW)%b(4Cf*lCCSs<*7r$)*=z z^={MI-J@!_##3_n3iXp0hMc#+DZlD&A<WATvDUCtorjv4(t$j%(TOW}y#GJScb~ z3&UIt;UQ|?ZyLWQoE0}MTT2Z3+0L}t^$r`+0hV7{Q-D~CjjZ-V^@`p-5Zijl_t$Ax z!swQF!zS<~F@b~9%LD+tBq{xDq)K|m?J}EZ;)nsK#6+hK??ub??_T;F#$y>7ejqhR zROkLc+KI#B|H7n{Z*8d(lW>)kg56g{l*~aT`{Y&ih+Iu{#9&I-f{9qISEWq8i=;eH@UK{16E zhU~54dwfJe3fB)$DJ(KbL&F{|>o2k0#DdUKQkD0}!i%J_uM$XwXxyjOtrB-QpNWyT z-!^p>e6b9X^C^9o5HZ9Bx1}UsE?NEy?Nsi$KlUj5=j74q@M!Lh54GO5S2-qP^{nvG z0~Mm`zCPmDTI_v>N(1#WI9B)=RMxs%iFqe>RK7eYD|rrK<+wNc>r0Y}q|%I7c5qx4 zli>={SKu!eCreno$R_YVIp-+^i#txcIKe~T%4~d_`Ox^hqm*++VM}{Cny-<6(RK)O zv*q@`80Y!OpP@s`;A%$>9MH@sJVRu49i+%g<$>la8?C$hn-%KF6mX7+eZ|1Kk%oxN z(zHd-8?QFx@ed~Vb*pEbN19!wk)F&K9! zHMquYiUbob-D{?mnH1DiU-OWLhr1YK4@53i zzzeg=F-u4`#oL2cPb3Aq@MxuyewvqOMCaMxhVx2VO}s(4#xdnH$4|?+aj9{#u23HV zANVf~X;DWfoSKh~Z>;NneWhvZE!+ z|7X6f5|-aZ9iF!(JPX6NM*V3t=V2s*L~GC2kSw3>%u74;B1$JD)Q(nC2i__QndVF~ zmL*~YI!tX4wJ2+WE1~ELD1^e0eODX&0(D1=K*l&+9kX=O#>mpzH*NhR$W9&%kxFfZ ztH=|@xF$M6LO44SY8ZH_HPSH?DG3cwrdwCoz*Kxk7)~5-6g(tRze*V&H;R_@b3mr3 zYd0uhc?B;HNnvko3pcuA0^1Hl%+N|mE_$To#u4#wZxF!5@rrq}uat2o1MW1b>g_o0 zH*w3nrywzL<9s^Y87&oF=aX%TB{nKm z-bEt-h8e1mbp9XS-ZCn#W^2^N-5r7j4cfR92<{q!H}3B47Bs=#Ew~43+}$m>yStyx zd-iwk9{21q?*4Uu!01|QRjraatE;N#Gv{hc_QBu=(a$@%h4t_%q8&uPOXGrgZ>AkG`A@Q}-gJstMLQtSSKtp@vY=SP|! z2yXt?L1TET3nhPQdLdh3u?FvbD@}e2AbjnGwF)IR?I^|;k(_%Ijf--W+Vqe`>DvY zxnvjiFCA7cdt68tWr>e0RUb3oH%Z?ct7IY2Utrf*F#*OPB^Lj6TL^l(CC3l55K0fN z_fpmClk3PY&w_?3ui5;SVwh>pCLE>&ZKQ+2aOVy+o!WxPd*;BWY0yrHM~OaEt`i3I zVoJg}H>2Db9@mG}-Ki@9POdmlMtgH+|E*(|rZiMrLdHZid%%{yMwtvld+9;fTJ<~I zR)BsYNsRw${qi0xun4u*(st_G$lXk&1w_Sn&iqJDBTmGMtBf%9Wz!s5-C6=BzZLna zqHh;lb9gHjClkx8Dmc`&H(#oSErwY9BWk>67Q*ls4bqgQ=11$KbAmrSiS+0Y9|>=` z`yGi!La^S2CqYj?e2gXb`2r+X+!p>2iECB47nCiPTdllwLf{OO@U+T6GyU)@k(G3BxIiSW!a z^-ndagI@eW2zgaztN2+fkXt|?43Eq=Mu~c*2+j!Bd>Fzo@3*Zd1uyLQz~lxoGx`_f z%^6X*6+Z+`@^@<(5);1|CZ+c8Q6_H{Kanmc@#%86r@2JeXs-FE21uNShlDCH@hWK# ze6^JG)RQBiD-(8rrM$# z5-5gDPL>WoW!C!(_NVlM#NOXwtQ`nHj zGw}s~=C0adc;)o?ecJNSN6OPfDi(s7QtK?g4A$y6?NXfr<_{;mDQ;hfiHKwz6EV64 z&7F@Umu|4DW@Hp08qP<|oUA0t9M0mEv;5kDbxB1}m^5_BVwYAlsSg@j_xZoD{XlF%_+*$^DT;FP!g&H5C=ter51;@6=53zCTz(*A+do;76 ze(tEFEe}nTf5kd2Vf2tbf9oc;&PTYraJU$QJjSel(OC9J;^v$@pSRGGo6)y}!TOA> zQD`7nPKfic8g`^p6f{m1o>L`3M$KMQS78CqI%=}4%4Zc#seMJ~T4@x1>ME(jYwkdz zS&uS@oEOQ&YUP$iH!?cwgg^>`g>4ifSS~X$H0M19%cM}=I{hI${SXij_aDmN3W`j1 z{j#m%`n<`stn!SSxLD4#Lewgu1dTXgzc)#| zEdTDn<9}>qc>nvKU*?E3?Un~0S35*JK7Bj;Q|C0n5?OEVZycKk`zJbp>8CuTR$c|< zCzX-hEn`2Q4qhpTU$Hwog-j?ZvpVZOXnO0quNO&!;sHp~yIW@l0m`}f1aE^PQy~-t z21R&o%Y9+HtzpO7rE|5;=*=g?rmMC+lf_n$+nIFd$MjJG^|zdw^QWA1Bedz(-nuSe zS9Mm{5)1uzna%4xS8Mh11m0|~G@zwdB-3w+(P1YockLkVhQlh0nit2{%d;vWP@yW8 zRBtve5i7L{g{wgiT^euFck{$E4@a@DkdMXVH?h6Gs&{6cPIMTQ;-0e|TcL(60p!ty zC(YVD5qhjW`1Z3%Lxd6VS)td`zGXRB(%my{9fkoDlOG8rGJZ0CQ0z8@9;>Zw26HddeGy$Nz##8jW^(Y*_vB_(UjDLX*HvBp;3>(e)_Z-8 z*j0?;FjyiQb5lI0=IO+(1r-T^nt~WVEBX@~fdz*=__A1j&aRaC$o1>+Ddi;Vtktb3jkx>l) zLu+~MzR&_r7j?92BXX;LRZ`xHta~QKMM-pnjAuG1mnSm>av)WiY&$6|{B5h*!^5;c z#FbeV@-w~>p?!=LBpH2c)$;}TcxBTbPU66YaINzav92{-1U-`G#2DMP`?;%HH>IT8 z5nIg!W~I?*aWr|&K7jY)eXT5|F%t9R@0G8+f8flvb>d`hMf_O}aT9OOh+vLV3&Z?3 zfI-lc`jBsh->0WWY!C^$gz9R!4GJp9L(pVkr?rZO!zLjk0+9OZmvb<2tsP`B0Afns zalA>Cm3Z--Lcn#bk|e;YXlNJ_sZeMaWJ(%D5Kkvkny{6^!*zhiSGod-p5UK@Z-aF? zSx<&AA7YF%DCs;rcS11wc?f1RvRh^SVMMnOP@08Vv5{E$L;*-72|2uBg>WTxph-mv zO=0FviydCV>RwZZ%&EnR zuCh`lK*JVPkPC>NbV@40CcGj1^R`X_&%`iH4^huFS_Xuqs@W<$*-&V`&56D7m<1VA zt%G9DJa7Ur#KcW*DTa@c*U(e0#bPInI z5B^M?)PReS)mhzC6wXXxM%i2F8sFoHw~6-Q@OpIXbBaO+pS_R$*z&Z*z-zVLw_F7K zj;OH>j}hu7oUI-~CkmhQ$qB4k)85$w z{g_Kt_YosgrQdZbhWUtU?p-YIU=t+{D|n0ja?};@g`o*w0i}G>>R3|U*t&j+_|bu- z&@UJaSedDG>j8&=m6y5n*0S}=e)AvHSSWH54lpzwe}Fy@-|?+doDx_UpLf)4@rk$S zTqNfr@73Qyo~NE%CHNm=;-Dysyo+c(9lB{{vXwRxswt-URjR94$IN*9@b|T$kqQQR zVB80Zg(ZyU;Y|u<$-^B*1e2k$P-ykzvL-X&WpbdLlv$VR9I1P05=ts6#shW;qy?AX z9-*WV?Bou4a}$C)HxYCjp8IdxJwN!udj;VOd4ApM3>SEUw)z~D_Lk>l zpNrWo{qwDzFKXM$sEaaw*J!DNQnHAeIgIc7l}Grp>v4U~P8hGEopb8>6HoKG&Xz?_ zRU~0Pl$m1hD800$Lw?&B=WKSAQMfmRjA6-YIXgzKJvRKi3luqx*_SL>awHhY;}3Mh z#*RuwM!rKfmr=_U#Vg;KYvzP+4v38uL&dGEp5-eaLUSm_Wo(H)DHW~5ctUfUKw3ak z<1O5r%k17)u{0b-eL|CS+ip8 z#YF!`aB7`Q4sJBR8-Q8Ef^6?)^+^LFWT?%AkH35rP6ev;y0VfzbWo_IY2f-xXOJ1;>t+cUE%fBX2tKQLZd&V#csPV&L;{tOJvnWvQVWV*J#0gs+%s z3bLq)ydk3B^aNTZR3-*{Bx15fJa0Z5s!$f%NHE0_g&Qjbf|ToEF3%O&1@$(3gY<9T z3&4M6`VMd;gZU|GAIGKIebeei?0;kHnqOI$v-;6&S~NNH7T;CK+LJ*r2Y;z+3uURU z8u_fVXJPN8wzO!^zr9*QT19zdPlCy>)LB5)i8U#s#OrsJU1$*2;OaQ{DlswPrpC1w?ix_M)1f;g z-56?Pprpm8ly5ecAzMB( zyPs_J(`ZRC$fkqv?EB|bm%08><3V`RKOEU*r1CRSun^s1#p)trfxb0EYMV?!MiLn% zG7fi?q%9q<4!qgns3l5Z#8gi##GIn^bc1f zY$UcdS`|-mAt>+paXI-AJo?LZT^mK+z(O^n=>u9}e^Jx>MLypSjWNixK2cFlKmrQz zj`ni99^`$suIkLq${u-twaE~9{$fUH(EClz9WuXSE7S7;HQv9jwP&T5^NbeE#T=Z68GhOTHiRqitJLVbDMomun$XQJqwDDTg$!1T;p=~L~ z!llZ)*)M?~wp527uf43HEuH0Z>YDYhGDYN~+s-Dg_ox57a`}$1%tLO+GRHdOKj?Me z$UW-xO-jH7OgPv*Q0VWkT2{*{R;OUd!C)~0`%p|I2*)qikrN~}nOxSXH0o?V%$(Vf z><1Di>o!A~uyp(|PWimn5Be|}jLR*}5U0>P8iN4snFcAP{`#9+7dH}a0<9|9rd&D( z5F1u#vxkfpNQfo#!-vwPlPu-*1+YzzrSxm}StF$!1%~WQy9|aWKfFU2=Zj&r0SJDd z{;`vq)U0PF>CF;%?TJeOdv!nNx>(3oKaNsuRO-9Vs4qx(09hiGC))<)n==#+EtZIA z%Vx2oT?Jh_ay5iBucRXlkW`2!UDE;bb4KmMn6%p7nMsM5h{6>M{k1$2-@Jg)DW6QC zgDu@c6-H$)d-H}qV|*5*9?W+rX)H?0G5s821q>nd$kCp1baYcPEm{WX7>+!IzAE`) zRUT~~_f!Dv5`s}P+|B4>=|7{N6Kr~{z z(|=i6l@bYw6wCMOd(s`oQw_O~(Fy#>x^)r|_X~H^!+dCJIK4_yO`W@VXw#mGhqq>_ z{H(OnP3^ihJyPhMypO(GzfgBopd5VCr&Uwi#eTMXwC8(kb!%ta1da}S%?6tX^9eU)Wz8BV()2Ho~J=oswtV ztXhkvaD=KTT0d{)6HhE;d*;#?$~GZtK?1vdRt5vWKNr9F2a~zmXGUV4WnQ zFMWlKl3+_z_%Zf^=k=0*5{zd|)xGe9_rmOX6I4ASn5h^}W@zd#vmDZP^L@IfyOSkL zalnq9TYiR>ld-4y!XPOt%!fri&y=koJ%NL-N)NA{-B(N%ud1B6nKup9wE5%W`*qaU zZ`Jlvaihf7MT0I%|<{Q%GC8 zd#-HrS!-;y36(5hKm+DfYalaX$0)7n%IFEs>-xZR)!Mb(35^K`GZdCC%Mwlk?TM!# z%gW?c?JR;VF* z&6*7x8(Z-AulIz4pWJh)dV$9HmPj>umOHt z>?pBM=Q$GY(D@FJy~c<^X4yB|Nm+8ro7{Fez*DT+(1^6@qKoL>rT)^`m?{vwVwRNT z;3fBm7Q0IxX@^zcH0e_$W(S}KSey;WCX>WtWD)v^?T}ZfiXk1Q)LJIVIE2dNq9S)i zFVJWssV(U1cmmItPifN#KUAxHhU0&6RIgf)=gUH>Gc8E%Iq-}cJ}zB{b~-|1nlB5Q z+#EsJy|Snesn`_*D)SoOZu&#~Qb3rzO*qP(Y}3|)H_%2vKoM5R=BSk|ZT4E^)+VYv z*4}ZRLkS}r;v(RvwmyC5`{E}ZxFZ)ZEROZovV3>yGXqIKR(7yZ|0`|%cx{qVcHhcbiI7xv>49&ogi)OC}2!u$x z`a!$?Sn+*UxA|kKPdXYKA8r<76}xa#SX4rb<-TARNXq9Z z!?)`UpR+o1&8LzzDC+WqZ@~Sd6SQ;z9CVDN-;rzMtxapc^`s;r)M=_A8YPDCbs~_| zD)=$b!bsFV_Wt7KlGalK*oCdRab+uVb<>c)-*Wbua}~F(7rXO%1})h;Jmc`U@*0bm z_KE6@5z9A^9Ws3yx;g+D^jctHnjMa`YR>2RuQQ5sW zI;eNFqjDRf`Yz0%5iv;2(CH>08nCJ|3BP#3RH9Lm?Y{PL;vD72R>5^FjK~u!9(#RaQjGcks zV!8c)f)wL4qsh`sN}efI?}PRa+=k+<<9_m!L%;TCZN!aJ^!kFXn(aulX5OB4@9@(} zXkKsTInk#qu_*QTHdvMi5$f}8uD-=eH{r7#CiGe2d#UeA<4ywoB1Cm}A%s#$ywAwo z%wYk70nSt(a!#0;v>LVZ;B*$UvIO`v)WdHQ4%sow=*1b3Q@eIBp07-YrsKA0jp*FE z&h@j74@}s)TDFu|Nr4JilVYd-IILlFXWzgxUXYgyek84F(Z;UZucg(&f#|jRoxX?)D?nI#6zbVftM=N&9p$r4WU`6nVEeUBjv;uyA53xWbb4 zS?SQ6B|d#GwL+v*O6!Rt*pDj#)L#;Z4YGe>*lU>*_1HJyy%7JBS}Za+_2&JofW#`{ z$+T(s<2m>LaCQ9f9C%CL(I8Nh_27zctp()27 zTh0xS4olyTO=nJ8oAz<#sb>agBz}9RN%RaTBT4<#6KnmU`W5jR+f@%EpNAlM!N{^Z zZj}Mj>5sUwlMM=Kw&q(fBqQGpN__{1mWL#Bq zq+=B*k_}_59LH$Z>?{vujXpqGe*R_RJTI!7;itXW^2)8G@`iMnI+pE$oXz@JW?z0; zzaKf(%yR@ZjerJ_N>zMt%rI!-lI{K;twMMFnbGXqN0N_kY3$RK4p&@H%XL^h4kSn0 zOg4d!GahPDyu~Ca#sy1@QTTojzAsWtrmndiqY?3K!WlCuoH4yrR`Wq5{l)#L~j z=nAG?dLkrL6BZs?6XG3@7G8SA=0!51F=N8d&J9AqhqM($yC)8B|AMmOeilyQ zn&?$kp@#hXi3#`DSgo!YzPOLTmfrhQTQ6N#Zl5JumoNrYcjp=*odFc*=^GrP*WpPT za}~ei-!bo-FG**-ZZoaXg!|%&l`ukmuw2x+_0G(&=(vuZPk>VMkN?4e8!zjB*r~Ac z{P)Ia@`|l!-7J5J9MYMO78wQ924YHxTw+|kN)2?SvtbAq?*-@uPjEtK-nE!wFo88I zS}^s4OhmstuP6q^qpX#!k5Xr4#>a#P3_Q@E;hWt2zn5oh@F8x?tH5vSQ^vGWgELfcR9)GngYeSggtR+7A= z)1tej`m)oLlDn-C)IC_SD(~zJqKB&T{m1Rvznic9|HtESvXJw#{Y8QPoyXw2w8PiWxjZJ5M-)LLqDb`P&P{*S%w zE2%3sh6W}ug7ZJIgTSA>2Y%!SC1bZ zTV6butl93Ttz_=B#!DV=w=It+$o#auI7d!E=55cUVRHq$^_UB&g83`c$G>$~DnLn= z6ND3h(}IV`+xj&1c%8O_y3SewzlO(gz@b3(N+tR}_e@4xy^#0Y)#2^g^%WB_N>%VK zmowMW#noX$c-<~u3pz+8VL8rY9R<3_%DRe$1*DR=Z2D;Z0iXxuuRCx8Z0=52`E^0C zfTj~>SOGmF#mOYMSYYhs^Y!9DSTa5!^XdpCRGkmd!+>5TNbn5#my&=@T9$MWR{Cpr zuhiUz5r**3#f0Sv`^kSU?mNO`piv@$Epm@E&M>+##s5K7a1UHCHVE>;{Z)mwi-!KL zEcy2`Nn5vX;DS5w=$IPJVH}G!8;SafsQhEb3+r*XgC52dh7X+cK-Fb&t6FlMZ_N&& z72zdW-LD}#ngtG=>gQgA-5dfO1cInkOOf5QI|2*A}e4a)wYP(}q&@703mCE@FvfQ>`L-GqHap^acwI=HUL$Z!ZV8 zW1$+@&G5}FSbYu{s_qT=@JIG(ZTorA=hQdDzRh z;>};3dA};$tZI{CboO}n$=;V;SCg7n6-C*LU?;hL;wNqU19WYA8&?Ep9_qwe@0Dx-mRfxcyrrrQ_U;v{@ zT5So#e>0Y-pZv|J04TivoAD60<8Q|Kk&AAtU&H^>5vzBMq7sb1hT6zgh9J{r&2wM8 zth(V_x}_?O^P^DW6sSG*QY2o|znIZ@+qBwuu&n(ZOQWQbqFnb)PM}*bK2X^b{MHi? zN7CQ1`CC?*8*lXf-&Xxi)!vNQ!~;I{|8FI%%8dOm@K1jW{H<~`aDSEb(8+-Cz{(W}aBnyNB0Lfg&ssAnNBXDnIbE`1`(o;NAlGeK&WvyMN1= z9}(~(1nRDV#{a$`)ic!&(1Tnh&=&Io7ObbF<7Ifz1<{NwuLrntY&UH0;F9VaZMfi} zzGGnD|!mt-Y`{Rds}ZNR=%$U>kLOW z{0tYK?HSA+^V(xVmq#<{1wh641&@Y@i7ngHImWu{yxTJ`;%`ELVujWt?R9@t!5fOg zGP{lin`dxq1@~Seygs~HlUlQMzQ}ChKC-ec38a{Pzs3jhy0__&joTHu?`|euU8|oA zh?s)iYJ;2It~o^}nmHEK_LpKDZsT)$y!~?u4Fp3@N8`JTR>k;WT(FnH9}I%H3jL!l z-RZd9yML_8Xu&tS8NQD+E|}_pVYnc5f@G(v=(UZIWauhk=5yfrD|m{yZ2MZ2JbYK{ z2r4%RBpwyUwUtu=7#3xBJxg=|3}Bh)9*Ip(+%o}})ju+PT^ra|E(RVO06y?rQ8F0q zf(9_ml?kON0yYzKlM|0QcOd=1b93Ub)$h#PGqZrl_uVr_0iTgE74^l&A6`h&pF6%z zybx`X+4tXJ0eL-2n86PxCtv;V#AkrX39dnNvQd=*-bwTUuF+$eS=ARjmTj<73r){% z;iBvf$tu5jCD>Idu}A_VuGhsgy68C@J zwR4_8e3b_T%g4eD3^zw`T(t|iH1h{@hbcUj3te#YhjRuI(T6StE{UF*a37|QHI^-J z`I^)0II6Z$%jS=97NvklcHgLhw4tUrRWnxS^mlxp1t{HJH0d=P7FcU`fU9`Fzn!_b zCt!J%Z1e5>X+w)ByouwhA{sdEG^f5+@y7$ghcQMcch_J10TginyR_UxQ1o>G***L7 zWjzG_H&=Y#S#1i@h8%!_p8e(H2Vsdz$IsqpZ80dO>SRe=*M*Lr;C}eY{F3QtH^d!47Oo1z1L3i z9cQTLq#1REnPal|SA1;7^q1RLx6^aImto=mXQ9V4_zuD7OM&qpey{jv75e{1Qds`w z$?v~FQb;0mF=4(Sz=TKobD^qqV!)CrViDZXuTZm0_?DHV+2)0WCn{X{u`*bZ<%#ZN z{tOFm(uhcAJCs4Kt0%^HlLhqV3d8)QvB=Wo{ge!))jWl|;zesS4FAw>m(F?DPgQ;B z6d-sB&4g=U{hw<8?~V%p$FPc%6!HXp@o!pMI65j-EtaXO0{FT!M2dGIQI2js zxf=9!UuEo0GgmG|qA#Ua?Yef9B#~A5y=^ja=ci`u8pierp)sV;_jdD)p_DIox7^>G zbb#XW29MC*h6Rq#E%3Yc-8Q?~;?s@OMP`UvS<0)YLpU7*>PLy-k&$RJx^Q1#Jv5uI z;&mN|eLbkHzcrLuHU>-R^@IeSj4=%Z<8X$u>T@^JEK(z0&#$?0MOe41IJBvV<&qGt8+UQJ@zPYLEji~C42Ca5jNxW6*Hpbl*(PJ~W{fz5s#J4Po!B(%D_h$u z?9hBxPkgn2bcMBgXt$PhI(OmDEytuUW^-F^ZK9d)5oQ*G`LKx$N!L3$6Rkz7p1U<7 zddyYzlS{wp-4;EiN#KbN=WEEuH9Bj$?cDl!8RviHRJJcC(N?0EHJM8^8&3y}X%8_j zd_oS2JQDC~I4pB*+|gWG(CfzK{vlgjC+nx2wd&vSWxZ6uxocD>)U)%>=TN9U$`iC5 zEU}Q>>atX+toXLJ3n%%RY@^?|v{5I^g~v$SCjIwYo0gtL^o$Bm)!<;S&}o{<0g|3z zbHxY`y9_$%$s8^jBzZRhT2@$c<@3e{mN%|9x)%K{ZNd?jq&^%vttdYy1R{5F^b;KI z$_AGc(11jBI9<=3{TzYa^)zLJYy9zel&hWkM>+~K$69Z-&H$aaDLO0{E5g#_A zr_`nk-?mJ3xQYXs=r)cBy9I|vu}Me*&z)@;eM_0Enzk+TJ`5B(7dTwdSR%*Lxf4M= zJM_5Uu91KD-UEsGvJ1w{ zcj`@^Q&;*5gA!u*aE$d>IDGPp6?l?#9ro?A2K|y=KH1|O2QhaB%|D?$O2;VkR%eH2 zDNVyv`iVsA3#lI@0>M${F`%Uk>5W?| zU?h%pXk&OF(HgB{jlu=?A)Lz`j8z8(4KyvPLA151Y_%&>cszI3GJKX%o&4N;VGm8` z30pO;u@To56Eea|FgQeH@sWUxRC?QTk{bdR3yl9zonlXMSWwfsdwTj+5t@=AbV9H) zl@^{oBclyuS=wo6=z=dC{P@nK`8Smwf z#@3~__FDvsqX+vS5QG~OsKSEA(U4h0aK=_Ky}i14*e0l)^*=cU@FD;a^gvJ6TGt1zOWyUhNwZG`=ut2L{nN|5khnZklQq0TrWrZ ztg=Ol+4z5I^#JGhRiMz-Xsd&T)-R!dd3^e+0PyY>AKlx{_~uyD{2KbrpMvhbO`jW; z306`1=QQj{J|@sB4xS3;utQw5n?|cR#HlCSq5QW+`EOREKO7@J8ACpAIdDyq%7E4Y znZv*McR2Hx2#R|=1lR!=Myl$%jxg@m2RO+e{h(<3E@_6wvZT(JwrA-Au30J(g5tcS zowryS)Ou{l_1fF-;l-f_l}$MZ8EGwZ)vEToUg#K+Hw2rBPwdDpc|cQSk~;$}f69w> zS}R(5ip-k$O1O20prl&-J2H3~I;swgJ2OlmB~s#jp@WcLA_r*IDOA|9LXO4_j^B${ zYWpCmf}ooGNH7RmhB;j;xV|uT^T>XM0UDJhtvNomJQ^63YMny~;rZ-DpUOQCRm za(H8SiCX_IM%q-t+Q2VYZ^4f(<7Aqz8?hgzgp(M<8 zIw@uOt^w_o>idPKCO`G+PkZ|E*7mGcA)xERUDHn-S^-W2(Q{$Apvp#4I!Jlj zjfJgr_}BJlMKu!&!P^<4lkW|klS>|pC`9bgdBWTlU1FeX_VbTshz7jgG_5(D4L|Y) z_v!iHl2HxBe_d`XerT`7E@zzhwrOq7!JYGCG8AFbNbj~m=XUVGDVr{%(#i=Tkhn?T z5sg|#RR2y&d~?wVmD9!AIxVc!6-lU&4_TiRtzzh|OnrrbRl- zV~2C})j{6DD7Bt~ZtAlE8*vU$D0#?z^*Gkkv9A+bS@0bKzG6X*-=JkXBIkA<_cxkA z0|4i z=eNci6V4hN9v4F#5Jo47Jk~^ON)FBU^Cd^lr)FUjhKaeh@+9L#sb&dr0p6uL>u{5W zRu^$072#udLfq82(W6DZDhPtXGYOcn4@j@ooWUfDCU97;>|A>(k;$t+*o)$e#orpq zN!L0Je9t5r%q1JlbEG#HvWk6KmQK&(J!R2`4ok>}C7K*xN}}{>JS7(WZi1HJ-&mRF zE&C@!{DXhuS0m^sS;Wa)4ko^#&#IXX{)Yu5`+tbj!OjA<*Z(gMVl%aDek}K6zWH{D zyecKoE1=v*%hK3G6I2JliO0^v^Z1{l3xu0t=1PSWoMrf}r)Fx#F(|hbu;?=;=jyzU6m9|YmjM{LRkB%8Vedwu{=@hAgGCk3qnqvOG zrTTB4*{N+$N}I2r8+4`D?I|LMtJG$`jx7)jRfuH_szJTO*b^6Pl~2$1e={iy3o_H8 zg(3*66WHe2B-D!>yhMv<6JfsCE&&+$O2&)I#)r+b)-zanuKzBEQzUtEZt}5o4SgC(g+RFpY37l^ zof^FNEk^@WijML%rniMRN+}C`E*K))9k22)S@-bM{hY6xK++b(O6HBQoATp5MH-3$|#1&VWV%vVdG%>hb4ctZ!5W0DDJEFB<$KJg}5>Q zt3%w;BshgVlrgxdq+~9sNDgR;dZ#phkJ?|vj1g4+|rw?GM_Zjr{Qw^=F3-}84{H+m7R6aT*0tE)^TSnsi4QA zck;j%Rti9;_CnS+Sk|R9YCF!3HSn|i!Fle!)F@UBKh{G~d@yo2NQ_dSld~|Osu<*J z{^Njrm=D)yMctmpidIPx>I>^ke)O%i32YzLcLG#-;?~loZ@q+J^doLwimGRE8;@Hk zH($>Kb0msSBV&rgk`Qgeu$Ku`crQPWkLmXu^3=hmRFy$b#X2L#*hVG)m=ex!cs4(x zv{k;(!n0XP+;x}@Npm)E97Kqdhs{%0wO!(*VTHpv>m$VU6cvrl$W)!kjQ^eaNWhr; zGpIJsi&)WRYO0tH;-2HnXE>{8*E(1xUZFsbn2)G$Le`5P^nKM)LXabqx+&t8PfK&P zjLh5hzE|{{IvBcO80}DnE)Qt&$BWel?$+C=f5ycSJv2c=z8S9_RKg37a5WUZ_>|6s zVXR7k6gA%f~nezRPz^ITVtuG1R4xv#%AQH0xM3Cu8pA)79gHms17a!33+N|;FVgQ@pM zAA87JxaG`Bx+gI1)a3{XDtCC|7nxpeWdTHaZgR=}&rYkPw>0<^2bZq%K3A{q?_+x& z0VKir%@2@z7aA5}gWQq(~pS#|CP7sGb9_Wl-IcVjp$pejH zy+@IxeQ6p_(wT3c4%%AoCZtigjvYk*8yI4Quy~r0G!BtBUP6CkIEi;y_|Y=4pTJ~1 zrk*8}pYiy|R|k5zL2sf-dEu5ZYc=VzuWbA>19ZkOOL-I0ag0CZjOb>Zq|es`S6*GY zJ)`Axy*a{TP!DXpnx1c{HBUDv;c#GUZu*!6%k@o|CmoLF!dl-~+FaBm?c0WXk>aYx z=+aUK{-0@ z4|cVbO9}IfD*GIx5n+%D4O<_G126A_o6g4N7HlL&VY6!|7SM;?d{I&onFwuO zv+v%Fy9m44)P2M~tI;Sw>7t<6J>%JRc2!JyQ?fvPyVrjpa863|mXg$)bq(hPNrYR* z>lLmw%%5qS=f8zV>N3JcxAG|;@%@sU%!Hn~s6lfbh=-hBy$gvQAI@$&N1tyB>z$EU#q%00 z%SY4-g|)V6=xk}~gZ}3*YBm}no6`UC{Nf3;r!uY_pq1E|i`;SwG0_FX2vdMAY+PP0 zL@U$toMca>8b#zCD9;t4F{Azrfa{-4_z*i=;*+a zYE)Ki|BYke{!#4XYI`NpV=8o(-ash#jd1V{t1`IZ9l6_W)!S=$%0cn7xrjR-GI|zw2)t2|Ds|Qz9hdkq`Wg zr=b1krwo;#h@#p{u%DjqAIZc?@ifQ!T)u_5lIvHJ6F@@S48%&+sOv}&z7Ws~J43V5 z)J4s&`!Jvlw1DEAy7eJFV2_@>*Iwi39bXGaNYTqgpB%ld+dsK_5#KI)4wc;IHQnY> zyf@^8m4U-&vebiqZKYN}!eImEs?qP*rwMrDVx#0ciP%AQikas=Z07*?mUEu?h{6sy z9)wmAGn2$Pe^}K0ucr76r0e#R&H++o`)}jOhejvm`peMI8P@i?&l&>XTz2`W+7Ys+ zy*40uXY1p+OGB*2h*kDrvr3AuOinUuqZLotQVwFXYh;TaN@;3DV)X7R&P{V!-6&78e{UU>6=DE8&=k~h>VxWcmD?|_HBev9aS zRs-8;pzuOo`0N4aM)M^xOB?jPEv%sX89NGx`-0CCwNfpEmqPAFbsu|8p|YxE7ez9n z1a_C0`1t|6A;KdSL_F!LQ!}JJ>Mba}+z<73L@l}2@Hpv*U(Uh*3F-Oh>G%rtuST06 zJ`nAT(*m{7jh`U>5z(qsCWC-LT8CKk(&BWv1KZL&=SFas;PH=emch; zJZgk&QLI31obtgidxp}>8c2<5WpS-MUOyapkcEe0oUh%zC3JmyvXpcjHH!Vpbw~1a zX@-_O%AB?<^W%`=AmFafU9>D?9An;A5j5pi%PFX1u01R)o#M%y!`M)+?`W;rjIdEK z>OC9MmH*Ck$QfLg+k~C^9oFi7yB#zqffF88$kvs#$At;rF6YLi**P>3tbF(S&Lj5V z8&ug>3xo|L*2_(&J~Nx5ZmIN=5`9uu8mY$W4zD)vBkX9~xHpVU9wiay{z4+o-%2W( zFPJaB0PS@cI{vf;6$yOD{*!^c{;q5Ga6K2$+3Fy#|GkmBi{BD@*{y3S@<7o1TK+&+ zXQ0du+NkxQ?%`V?+{};&T&3_sjY)0|q-l>JYNK{c)lb=}rc+o!hYtfo98Nx0`pfpq zbjD~$Qhj;0Ra20`o4!)$cNqE~#y-)q)9VJu$A?hNQ+}eYnx{5`hJ~;&Wu9%mXa z_J_CRnORHMc9z3ieu+|W0#(4+6!uVb(VPtlsh@?Nkc2S}Fhffejw55@8{w6qM5_aMr;l?6n~E=dn9f8mLzK7QQC|((wnT(* zuC-Wk-oJOjC`wcDwU@+jFdx*|=;8%oE}$bKcsN>AI(KMo=$Yt{JEVw_^JJ9{SfibL zC7XW$9ggd|bQgZqXhL>{!ct*v|W04ZygJc zR{SCPE^vT0%S??E0#z!5szEt&*xhi0`u+}?{d@oFKkNiJ{{y4T&ceg<-|hrQlqMVj>S2aPo?lr42H$=eu zq>}f^*6w+&`)Qyj{oHr(T9;%YqZ|6bDqw2;{=RuIem`{888F*zRF4KKQe4=$peW;=HVZZOi-Zre3SDTXvBHYU>lKue9Ipmy(}L6(7Ar zW$jL`5Td~j#x1=NtY2=?&oUBi2i=4U`T_~1b<|>>$qDKet)l-C4p{c zyt&xD@mw zW?KYK)a7vo$dK#Kbh8KGVt&fox2Kvm!OYZ`Id2vo@tRStFPWP~ez|9X_?BsV01(z@ zixY~3PdVt_9#fPcQ=me_8Ucmir&^}r$Dtvo!rs%xHaF)GE4*TS?TzCRL|-wDId`k) zGhfhdohl;ay~~R+Z_OINEf`?^q%Xf7dW$ybn^9lyupyH4wN(I~s$2?Li4E(d=ghn_XWlvY&YAmse|^tW zRlTdLYRg)Ct^KQAYp+!b_x<9XEg3nmbf%9CO^6swS%tshvyq#XwidY2H5as46- z7W5g+ZT;rAs$W5L4%Bh@vGBe-N8fBi2~%gFugv(9-9E>;kFuhQ(XeD6RCWpl=&62H zq_L@pr`K?n@XxefJft#UCxK_GO_+t-@3-i%P}6eO5AW>qGA|>l>0{#g)Mn#5uyvF1 zmdL?-aK?@`{I#eOZdH|7>u1Sg6WtKRl1$D{nCDSaqG)BCZTLtQ!j8ZWi^;04UB+Ai zJ^44J8FehUlhq=dHeX)j`I&?~I;@p&sb4`MI&+?=)EH|m4IK&`%Nl778z>6%=D!srBw z3*X3FTrNYz$mblBKp-dhm?VV+iA?+B*2t1|#A~IOR~D=2ZPyt*=~q9#K2yEw_2_$r zoHM`ew??p6-BYen`<5(x*g+$5lrS4Ov93rqwA8#f5>iyO#yD(B7J>$IE!^klc>bJB z$;`h-)a3NTR)d40!c$;&O1RUpNX4^Q`28NFD}o`lIZ6$3TxCXNYcT14qRWxNmkU%s zi}PKS=+5>#o;%tf33bP(j-oXeK|qR|=iaZCdW$|*tK}rg%aVSgk>QfctR#@u6QYma zs!h3@U`>L&%!V6?!iZwUw&#?WSP?O=@?{g~dY(T9uh@P>b-|`u&(y0rW+3j7+UfQI zi$^q3dF-{2a{NXeg3`0%U>CkrBbX;na`R~z>AKbnGI?uL@TvZFh(n=vi}mZExRkzU z13v5d5@qkOD3r!&F8e^g=qN6(9HhSX~}zc#F|W|fD_aQ?ZFOAW&_8? z!Z)x}OW0GeRxI*yYcN~jH# zSL(Hab>VekO_Hje(v**?rQDA?93smc*Jr6l#VRAs?l~TlpDyWZRMDh6=&7-GBaEs$ z-9Z-Hfq491r(BhbXZYKA21qq8Z|nH*ScdAMp(J-J^Yj$EjLfK?kZXl`lb%-~4!zwq zt0MOOw6)`5+nG`WL*jBJC`%~GkT`8+gp7|#cGS`B163WIo_he&=meAwY4hjRuKOxb zK35ymFuqr`)q2Ag-^w$idX&kL6n#}eR2;TtY0t;;mCr|5KT-%)?esck z-(=l{5bQMVYiZE2Iia;f(Qi0hO{hKP-B0{K5m1K(6Of5(oZxySkLLzFABa(M5$1}a zA6-xhjTG%^tmex^w{9&CslZY*>QK?!`elU3Oc~2#Zpfbods;R@kTF1hP9sNvcup7D z`7D$r{XKzsSY!Q8J8h)#^aD@hNoyrV=^oKS#pZK0=M9fXR-70PoERd?fRNW30Z}wB zuas#~T1nIlEJIR#?dX#5W7rEiv`~sSKk>ZT6~#O|@lT|}e@n{+BjkKnUES|$DMdaJ zu(#4&tt`{$WB6G6qL?+|ov!+u!~FF{yPF|+-1^n$tPBPDfk!iM1iUe>GfB$NJ*Vp2 zhZ@jMzo%T9=74!@1%zP5%P9BJrGVusa&V9yEO|0zvT>$P|iHPLAh%uhA ze#J2iHp=iCEhE=4Etj77>&*A^#+CUGw$aGtJHK??)t4N9UQhK@|IlS%|Nb*%{8P~m z_Bn$SGIOngXo`NhiTXNrD+A-;TkSnoPCbYHPnCua@ZKjHPBCJeGH;jRX{OaMsNw1h zWt63NH48fvV%x&|UzJ#@!24rmKH>7HUqO#Lcq>FCzPbd_y05+Kgl|k%K?K8o1l!*% z{UTg7V>(xm@2w*nZ|xixWP;HA-^;n$>dy7g=!Up&e3PEA)5t4qe>K0*V9=njCQvY? zCd)%hGOYDggW43wBJG12`lYJz-vlF(E$XVvYnJYM6ao_o)(gd{vKp`s%ts z)s+}mR;TG)OjGLWVaC-ay^pgFDMMYa09pSm@zK3}aaUEc2$uYmZ~P!JpB>g$`$$a3Oej%|3}I^h{Zp!5I)G6JTB_H4vxR4)qtS26R}|kfG&0Wu4tM;b3o=ligNb#H zq&^QGpV>X@!ZqZ~v3AYa>U$-aBYrm#?q*9~O7(uThsBa)N)3P8ol3DMVGn$14da zt2<`S`oMrPCi!uS&mrCTU7ok3;FX?`{~R&vN#~#_S>Ch4d#35&6!7%8uPg8l*6*@?m#NrUC#B%G|cg4hbwk^-lcs zRUg55X;~%cy7i@JeOa9<_1rf7FY^P=nU&7n0CW`z@Fc+_Ll!zRfRCG*_3Vne1H#CK z$zyCCgoYcIp1h-&*potLPeG4QthM<85G9#H_>P%|-0S+&5-H3AUmNX)i6oyYBMpI> zfK!>LF?7MOEpXWw4|ew$UGSnkTp6Un>3`{xt)j1qPhwv*<$L_eo2fkNi`}>JbAeI$ z18j+B#1(;2(hAgQn5(NM1_(X`YdCp7%f5=E#>B$8?CdnaA>`AyY_crY@+=ECU1Q>w z+F<|t3kHDej}b8bW@`TDK{a{*8B~*PmTEu zN{vk-KnXwIhsT$h&|9F_^-V=IKZ1_-E0x&G3wH9iNGFbZcwGwbM)dC-yR=DVqQ3i+ z*Tj~8ph(VVY(pnVH)6Ewk$zd4NrCnJWd;UnN(d6dH(JAyo&*zw^-;r#0gcYBx>#RF zoC;&aHH|Vl4@Nf04Dy+k_HTPH5g4rz@Sll)Bd^gXYuM(7 zn!wc7HVB6kh4z9}Uo3jSzlICuZnFS7JD!!Fcaxo-B}jR^s*xYQ$u>4rJtM_Lq^hpT{n4#1b9B5D z)?Lwkzy4gyjv^Lc?Ahu4^P|W#t*_6fBH+U$dTEp+wWx$OoqEA4qvA zc1}UMf4WXKzbQC8^*Gz9Xqqgz+4635oA+pkPbCwBlKs#pKE@ryr}6 z`xN2e5uJmX+Ukb@zekZqb9)stkTM0fg8NFgY|HnbK)Pood3(N+fspzScU|kU)4c@V zKfvR`1o$1mkC-iz{i`O?lWXytMRyZ7KMHKqYFm*qa%{6~H_~|?n=lmpwlVvhvuw1Z z8f%!p<7vhdoAOvUQ6_evLb>Np=3=Ke9!KM?Y3J2z3am=Y;#gOIR?G0dVAtsf&lsm* z(E*=d+Yy9>_42qNgx|QZq&n*g)V@pE9a>rGJssLTz^JWFJLq<-zdh3SV6c_v^-JUY zML;Y7kbU>T)pR7VOSOu>F-;ir+}qX>I5pL*Kly-z1pz0eDHM`J;$PMS1jzJk|O)Isd798gBAm)1oHYU)m9FiG}hC za2Zxqw0;rVC#@jUp|nQ4nB@#Jr!`jZ9b{v(b#9qCGxsvxeb_go)LiH*l3@TF?!U6< zsWyO5$bQS(F3mEFtM9}5oH`I17323!vhgL2LJgw%3Tf&?b@L-C>-U$#>n)nASZFlM z3qNr63D$EV(49Sl;;%|Px(kiL=!waV=1jP zBp00tR#oaPLvBgx#8M}5)4Cc9dXdAX! zaSo25wIN?GB8{9D+Jwj6(ZCpD3RRiWpiqnbIlfnf!%LqNLidxK#t8g2d*YiBjcRfgRqC;?b#a*a|SJyxN^OQgQ?S%+*FI2jyVli+MGiik%=Ss&}PY zAx?w0Vh-Kd1>aZ=+nBRxcHVi9X~8>2wLBDc=Gs0!6&fFF?I5BCBR8jt5E;BpcZLB` zYW?+4EKkld%)3+LH^|(GDb5O*VK?q0{3ExJ(#~Uz#^ptZAn`NiI)YXi8@A63?rSM< zW}9CGC6yp~Q#)8|1V%vA=K6e_#Orr=3&k3|2YDA-7ej#?`L{bfL z=hq73R&*(HstW>&v$N(>y3GIGFK)jLdM{^%0vsjnXU`0N}q#RD&Czs#$ z-NhWn9$flvkseNR#{I^-eN%sv=YDtyl6y+Mz0xQ9>Ziy8&PdyD%VtUYs#i&kIqjwwM{$z*FzPnBk?|(#Izh$PpV^uc5Thx!0W6c?t0e)qq}?jj_zA13di_l#^gZ zOFE{VMyM1A2ds^oA7wt$kG&D=+>#)3sOIc5_2?SN&@1)jX&O5&_vWoz% z;WUcW(wZgLeD8@W?3Z59T_zZc$n*qTjXp=kS=3|t9=|#O+)VDjhKv{1{%%Q9a;jC5 zic$Eff z^~C7h?+Cmc`;md>u|fr00;zJ-<7LLPpQbojLo7HUX3m(L?dy7CDJi_Wxt&dLS-x?m!4<~Zi3Bg?v)ySnIXv$k4pNFz00 z$2!4X0${8Ea4zY6p7iZnaI1o6-kg)%WqORf%o_Il`UPJC^ z0S)Ns;YhDXV5!)jiNRnewFjlli1(TMacf5@eoop6Q_Uc>HITXt>^lo@8~D)CM6p zGw|7rNed0SpD4)@A0$6z^kP69ThuS>tEIX=*+f@K3~GYBvP#5XeFP3h-vS{9%O8S5Tzh>z5k zAklRn_0=>wr4xBcVBf!wo{dc)CnETgA+UILm)E!`_AH@AziYdA7KCKkRL!w8CPvb& zs*1UHqNs*Fp)%!)HGwFE&yHnj5JzCoC*|Vm9L6Irm6~Th1(%BNkDBDSx3@6?VWL@= zz(QTMNxfGYxA;|otarbYX-71|x1+xuCnO^>oy#+_u4*m!B8F{YG4wcpB%wsO*l$-s zm53;RCQh`uplYW$;eG72=gYt^U<6#qAztw}6*LZ2$ue~%%Hy#SRyC;?Db!rV{d{f{ zP@G^UNK*&Yx#2~Bm;qfiVs>v=CD(TMVIaIRif#ihAvrG6*!<9&Wabj=LU_?z&lh$@;D{%qa z4%xNyXa1%Dh2pc%^iJ(a`zpqA@i@QA`(bUa1TKRbUzwGqH}96JJSr&56eSRLdfP{Y zX^ffNIYVhI+`K_wcD{N?i?2K`2BuQ_K5lXAr!eiBk(kb{&Mi6y&V=XBK(DXHD4SLv zZI5M|Dwk31;Ug^vYSiv=)kwn45KlTk3SHF+#Hd1r(5{a&)J#Klbae=j_jqDv26`D&f9-bL%MJO z22H)4+HPoGDKQ7y_VUA^7|8Bk<`XvC5(+*K{gI02q$G%Y&E?Gc>gLxHSx+!8K{b_?2~9>xd*`^G zL$0TdtlhUbubhC`fZ{l65$Z^BuP?CLMd9se*4_weZ5IXRxKUgw2En6`C>tiIXbz~) zaXc%U)ozFK@h_^fZFdcz?H_?`{F+d>bioML8NpxdQ#PhLO%kkRKc`bEy3zB`Yg=yL ze8!<eB_0$`;^Mr|x-rOqC&+|~!sWU4OqVtZVx6H5JS<$dF5w7$(bB%L zJDZFUUreNZj?G#3$pSbS$1x2lF@0@cz49TPgGinyLpxgotZIed%$s^gKn~;w=FJo3 z$lUnZ-@B#YUxZ`s%d=QS+Yg;V&h;@c?{-b0#aSl?<>;!XX}rIp_tJQ`@>2_#2UoaB zQG4Xsr|T3QcNSYhkZm1WDZfTnZdHy(kGD|@RY zU&rkAsd4hDd8QbJRpyDKqN#+}9r%L(gb3XKwGRjmZq|RqWc@uYX71>qXlyTTYYnuu zF}86e2cWQs+gjP$D+Bcnjmb4xq-~6hU5$;%S>78vn;RN`k`yC|oxlnx)s|O4i(++UBl|pQ(%JjDNFn>yxvHDkUmX-_~5d z=aX6$`|{vLD={X%l)PX#InuZP*)lWOoLcIoTgcHmP6qw%f?COCJMKq3ZnjlhTs?2g z_6mjp?7)IWw{`2srgM(@>x3a2$_qRW18>LBuCR>c;Y-28<8DnbI@S2mt-ZA+4fP(f42e`RBENrg)~@NKu#Fbhw(Xn-UEph9}U$ zc3lLf|IGoGA6~eeF=OWrAUc#ON=j6ad`eW_%OnindO^KR30u`d@@?;`U@Q^Cxu+V0 zEFi#q{_uLGWs_}Fk}X=IEvwIeTOiCiH;wmxpn&La&3%fDdgW^R8XU@3%CCJg8+f|$ z(v--_aG`7V<6^oH$8sBisT_>J3G&g)n@hc_Nh8>kTc{LO2*|zDR8QCwJ_$b*xLZjb z7r$h^gK+xPe&)TV?ch@%9dsbKC+S|T8p|Ob%mt8~IxTIw>EwLk_CZK9e^_jt13KUd zjke#sPA|TdTCYsGMi}DW=?a1QIfDy^Yjzzbh8GS~f-5s_h|l|DS+=1TB|?(-Y0cfW zj~LsgI+h*Y-c&oL%+>Kvq-aU2?=!qfBsw%b=$z|D47pzqczN>=Ex>A_PYIdx#M`?89$;5}*=e2A`lCQsJq9dYf4BkCG?agOP_nhN zK{7;>0&(LwHLV;<_Et=Ex|v?p$`mRUBD-P&G#s}a&s%oSat(FXckRZkSg`KMH0?>B zVh!cqTHT^%HMJabhLyfiTE1e<={1F33+&sGI)+|`F6)FR*uH6ME#D;)7QA{SUm5VC z{(}r~f#LrsTK^QEe@)?1F7oJ%@GN~?c0EaK_@4 zvTZi(l$2{|^Qk1~r~L!C(>wcOzm1a!mN`TTal16|HX8#uP?+*8^G43OIbeZN`5NKD zr}7bi=@l^1(_yc{xG0=-d2~rcvfDNF&}DI`H)T5R0tFS9x232$cz@`4N`dA!m_~=# znc4Mo@3mBmL;DFSNm@R(MM##pex0t(S>}(*LX}Df$zEY0eZG)OpD>;LmilkBgbnut z(9l7r6JFQSZB-k;Ycq;%36g8AzF8)3p`W{ol|nqj^s)%feL7yw4k+?kE*4BLrESVLn{WpWZ4(1&Wd-u&{b{J<>apX#YzX^8m`jFAI6+tsY@ zt^KY8JuQYLeE)ct{f(>hOjl%2RM-wQ1S*p#njL z>4#(O3n}*@6ByA;(N`};6GA3~OKxrrqK>QWcw8#wbgNK{rS`zXG;hJtL@S;P*K4(L z!@2A0X@iq4VpaHJIlEWYuRZ}6u}(b7Z`K&jNs!~#b<@zDQAdB+7gi!?yM3ptWZPV@ z?IpIoM&=i4J$*Fys8m(g6R5625+Y;rXDDHLXCXn^?RI~zEs%zodQ<7mb=2QpSEwWX zOHoz7Z}_Fms+g6;nn*@Ub!CbV$(1Jwg;6Y0%B1kgZu_UYS0%K{LisfBYfd+JZks%p z&Kl{>g;TFjPBcpLTZ(=nMe^7wMq_Mn@yqamjQMmTD|ahfGu;DD+mVxAUYLk^x+OY# z?@#;c*_S(Dn3Cx`OP-=vuO(a~{6>o3w`1aCvS&z0GJgVEen$@fd)+?A-xt3Bm38|e z8CY;^NG}V5znS^rv3~aL9)C&0N6(DD5WUO`*|U59;x%%4Bz3D_qQ9GgRrK3dJZ!gL zMp9c474c@yxdbaxS;j%UyLuo?_yu1Eo1|;!h*Sp75I9jbU2C^g^|Q*fD9G9MMZYTr ze%YnaMjjOIaw#!k%nbb81nT4)c?C!D$#eBjng#fi^;m4Y+<%)!`wPud32P4f#rn$M z@ugY@?XtTmy~2`gd>t@_`>5rf z!KiyM$Z_}f5;H`ZSfYFrASh0J!tShL+fqN!Yq5m9uDs`9W-oZ#y#|_UC97sbBjAty z9AGC1&H*&O(36N_6!i_vF1GeoOa5G8+pVKPD`Ng9MB)CUmxI6Lr0o9~Kn46(^dFp* zJ@JC$ug+;c0aVSo-vBCqjswY5aR-Wx71K~VH<$s(guwPB>hLkv*%sTtK!=>=^2d=d zW-&6&rL<3KDtN)H%K9JZ^!J30sV9180tM z#Y5lE31BDS_*!U(_X|^w`~Qef|8;=<1HAo{3pe0*zCiFh2}r6eh6CFF*9iO%ko+gP z{974!7AbicOVapHwGTceytevvof%yDymB$+X zG^+9QmlCd>ExX8pxzZ4cCa>o~+uJXO8gBWv!R(eGhws+{)M@AwrW8?$cDc*0Dnkoo zoLu4Wg_5^uTQOR+M7z%Lk8JRgJe@bmU zoyt9Vw8R{Fmr2V%z~<~pqMJ9z(7*#c}oYe0XKeh&^5sn_9lBt+UH^L>qeg3fXcPbW=AfdJf?0E= z6xr>mgkh00)uPM(p(L=CbUTJ z50it0$M*L>ieW}iOdoj&Uft9En;`#ZsXr_{oxZ=TZ-zJsX;J-)?sF4Pxt%f|)+DY5 zPHhM@686V8nynTMYHS{zn54&%3O~RAn(V*k)t!$47i@pyXU;`YKqNOzPAL2{X)Ws* z=ZAEm`Sr>!Hn8`TV9VRlJYmfsXlPr2TbKH>M%a|cbPT6se*d|2YfuoTF?v1lRI!CzY9Bi{#8 z$;yqo*dBAe%I*^7BI&bq#df~vb1b9mo;EcVul9Ga@T2kF&v>=0>z}gB z@6=|^%sr%YxoTXyY@3hHfA%-1{TsgQkm-2b{`HFCPawzd9s3=n=AxCFup|%W z#?2QadUx$9%l6C`rl${?APe1dn;;1}&4cOIg}d|lT1M>+yVR43h(>vX*sV#f_;RW5 z>%!At-b=JF` z4$U8r!hH_`u~xCU8ert+hS}G*aSV}&T5RrArgk)wXXp&?F70TJ9B5_Ka5>pKoKr5U5jcZWY%|b7d2&);TL;KFN*~t zbbqLy4)iYL-a>}v5OG3%?m*t3&vTiAOB_*5M|X1%R>q4xc1Q$+4}QoUZXPu8qEI4A z?=UroK`5yxwf98aZl5L^iiolb!&8}yXXD{D$Ub2=E9uGF_G7Au&f%-_E=^FrQYll1 z9!^XcwHVKv8#Y=QG&a`EDiq}v&!rcoD2MmZwM391!zB3{N{QJh9kW~Etu=`;ien#? zt64tS+?Z_Ix^yod#GDJKpc12j5{-p`Mf+W9ALRPxr6Q@{01$!0$_dXvd5#LCP25$< zkwazl4Ymsel?OY$F;0?&3-xLaT!vl36b${eGwoaRf?Pv~n--@&hs)++uE58mG2R$t z@c~3~<2Tc($GJ;hxskkC+OvH2nx$m|AqdhYVrk>5%CEnVV2ONlm1*XnZxpaqwX8%QA=q+<6R7(`*WMb&=4hnKKUY76KqmdWjz65A-WL>NcG!h>_& z9+LccG&;v0*z!QIRYBC6Lxhx09eZdP^;h$IV6c`kp4F+1256ZJDL1I5AFY zo3&2XFs^6PzSa?c_lgLeQJGFfQ4nu* zS@2MP9Y4gYt4;%4iMl5flTz@7E+F&!ZmGUcy3w5m*feb`;vv>4G!`qd^!a1sdco}B zR>5|>`;ILA>&S0hNY!}_f#D|PXK1#nwC;}1((|;a%7ZDDMs6#Ec}2)|Ukid)we8nf zqCY0x!LOT*65*6foo#$uqYh3q$g0w@+$);$X7mDIC{3z{LNSz8zWaB6D}?3$IL5J) zmv!ROf6q#iH`4Q7?a*CwPnw|FT$BHk_&3@YEwYhuN932R1Ttpho2j6024%%uSJKQl zqb2UdQB0Q`b`il_on>Ee4Bf1-;U*>&5oU4sz3Nxe{H9#-Q0ejnreta`jGpV~yI)`z zZ03zkg?1ur+e-ms-lKQt&K$|cTkl>sp-0(Fy2!sTo!IA0CSpYzrxu8#=}Q`nY*K>X z3;XPBO@t`TPyg9yUBpUhIi4doA53yW&mXO#6qEkZ2^vl#iT|v-Cf(_%&O4j1J!5Ti zW|;n4xi%FB#auXy5OUZoc(bIFoW^JCg<#vR3T^ zr~{`+SnO|J$4Y}Xzw|78pv-5UqT5G0g_YP0!qNY1-kuEUWZUdh$bZeIQ8StPg?I79 z%}CUX0c-3|`YvOywH!evv*msd!Q@?C)?zQo0QDS~Ad^GJTa9lQ32$9#XB4S;B;9d7 zuq0)!z0ni`Sc%+o)5Y9`mH;?nG^Ne93Bdz&*bq zx^=Mxf8n z==9qAchdqYb_i73jaf%TRnJ?s`#uv;ChOWW=YE%#EySz&h!yMjX1s^XWE<{yt|m_) zD%0~5)28BQVm=a-aSOJOQthTn20xht=Xm^VADx{jxF(krZceDfW&MY(Jx=X;f8?iK zzb(b zJ2*ZS-@FJ}Gmy}W&xskp5|em{(MCw?y;ja&Ez+;>kJKM~ib2{tyvpB>t-MCNWt?A) zr^oJRMz8c0JKyDN6!QueMp@+GO3y^ti(=|cp|4D80N-$-?G0XGKP^OodK<@pLdNN#|2J=j>Q*1B#VodI)=QSiz_YXJuXRbq@rF>jsT%EW1Kd2k=_+p z5zDd70o+Hed92m_yt^yjBgHa|@|0-L>Dd9>vuU~67!fZw1)7XC>a-TaG2!Zo+ch?` zpH$JRFD(<23I_1lYt05dDzNl6x4QkJn*L%A&<8$~59inPIdl{Hs z;7s5`a^N*8^oeaC!8VZe9c+HxoR25dM6D`>s7ugG(B7DwF07eO4PHLo7Ra}wr>PYguFuAFjb zm2tUdub)xR%+@aBoNvnugk}uQ=!HB;N`~}#$v&~`As6SNT-jUU|5ULsGE0${4p+C0 zAoKJ0w@u-j0bY!azL&KO4q{4luKSpED(@A7dX1NpB31yPP-a51_?lUE&+rpo?{0&} ziZ9!^DeHvJ+_a4jG{GWN=WR5tQhRWbAklqO0_;uFP@dQ8^cs$aUzSg(n|~EB1z5u6 z%8--nCO*g^7BmrSdu3FNoY%_C#TKpJKM^kndWS>)S`TqB*Qk`ZmiZA)n}++)Du8ee zDhhFhhL!7O_cN#Rq`fV}w|8Qsc8kB4{{iRfN^tdBFQ}x24e(BObmTBkTLG=T z`O6_DS^eZ|#7wUHOd3coDP#%7*Dy zsMXy)Lgusz>nc7F382vW5wk&-e|Ve@ia^aR>iNvlmoMv2 zhynN`G5*rDJU8c)U5J#ixv7~WIVT4@3d{c=yAc2(MwN*vMZ=Gzb6Ik2Ur#xy`2&n)z48ye0W^`^i>j$S_%&QZ1Y9hWmH@T zO0FsB(i?ytSV{1&fkd8MVgKfCnp%o9=|?95krf+;U645O+A9#Cg95`2AP9OqJ=7KS zyvljl7Pbc59oLSrJYbi!KWn&>*;Z|1P7Q>Row4W&-mT*M^gbp4uOKX^wP&hz2#Dr( z4~?W{oyx!~=%LTuq8jGo-Ek%(;Vkpx_1`WpO?_ob?Q(WIIGVa^4}vtc`C9^m@O%ow z#cN!}FqkHva$>41cTcE|cc1x{k#P?p^YJL8{T_FLDa=e<;q%*n(2ctljPNUDP^t-LMwr6}HHq+jr zIFHz8|F84=A9ns$Hjon(OIQ9xrgGxA?bW~Y`L9&!kD0sz1(K+CnL%HHsCWl{8e{#f zxBSIi@>a{(J&{2G;(q-vFaPC9a1^8)mge2+d9z3QMD#GobeiDsBV%eR$O$0asMovN z)SgSMdLYcP!CLBj*z!=|dg0@9)4j0jM(R^qswVMSK$BY6wfaG4LSs?Q)d|-{`MgT1 zAlMY)bnjsg;t~MSg6>9~1J6s63P6Pj{j!3e@58i$&yO^e5D(!U>JF$99cUdejNKoB zcx9xj2EQ@}co|*FN1+=o%L3i{S>lPnx&@x99bm#KD0ql%9jJ|B%{wsNx%v#$DTbW+ z3^Y_Ez6!*~P=gncmgNPP2195-U-$=rZ@MyoGh-R^fX{mv)2)DB=AcnNMRIT?u?Jg#^V^cO@PddeajnKI@pUc9ag;zs59l*&5)NwF78?evD-R zUBTi6_FI%L2?x%!#0O3ZtE^QJoZopJ?m$bt#)=0jY0m(fjc3q6M>=;D0amlp8qwmb zBg*)Ba!*xh#;BAcv1onDBVlc%PQ1211M+t%dO9Nk(h}cnubDg%Fqin?rfHOxq%RjT zCRG~sR75UGgCGWODFnMspSb0EQkw{aiw?y60Yo^Ok?3A`PJpZVL%*Jo&%n0I7orf>!@E91$G+PhCXozR^w1ZzQ$XC(1=M>8eD ze=xSFlEpnT3NzUN0lHiJM}w_+=STbd!DZ(oBt@V;ojBo_@lfN#`Hix3poZqn0og(8 zD-N)@#3n26mSQBYtmLAwmh}S4X9*>bGFunh`PC*2ZWZzHl&awj4%mo>Lqec)oqO(D z5Hu@z_B=GLR8!Z|`?mMwGW)poXi^#j6zyE+#_sW`d-&+e=3!rx*T6LiNuM_+mS5H{ zV8-ii4uJ5Vf`Wz&p}-T2{>JYNAnl#i`FS60t3=_3HVTYez9M2S782(<;cQa*9(}$z z$c}VJ4kWC*f0g8CfH;1vxZJ2d;eh^r(nkfePHWXrHRUE=$4z1yfEBFF`+GgiZEtyD z@UiK-7M0l9g_gKwfzWPze04v#iE^#1P#iy6JQN1>fHPR$w{PKhBk6Y500?9Tk8@)< zH9*TXc^v_0`>g0|M-_O)Yh|I38v)!@lAVfD$nFTNe2{k{hD2yic7mYoiFqCi(9_ZI*h3`;bYTQU}CS-ZY5ePqLvDrkUQi!JY;_*Ir+OxxmPi1QW=w(-V6HqdeU zd2JiQ{Llr5GJ=2aJ5MUB!O7Iwi=))W#efJlGL{kCm65x{ioUH_Xd1kK63tK2-M$9Q zgJkE3ua;?{sbQQ(%{UQTwQ#i_*Df!R4Crr|nDk?iU`%85C&ShfvJ#>zH3dB%J7<%1 z4`-u1ToPhU&AL~qiB-c??LT!Ew1wS*>Sq`9+iuT+05{NNG*6*MMI#9d%g$=F-nn85 zF$&15v94I$%|)F*-m3#l=|RIK^fQ`%ngbKy2oUJfpZldx*r9(vrsEGK{`^ozOZ6n` z@))%r_*O>JAlcR8p?kpY0d7CO1=6cY+Zf16H)t_; z^RnHCC!IQ}KkZIiia|D+nbkG_lFv5l1T;uyzXhCG)}QOsCtS!NEjU~~nYCI*{PZMe z`y`vek}0+Iyva`|-F%F3YXUmlFh#)LKCtEMZ>kZ<``RVssY?1oWFJ<5Tr78Y9EpeT zxrVVZ6uL50hc}_+%S$am<^s-`Aole#WP^1K+_u$RW(zJ&x%bfW{U^0~3t+TO4Ys7> z3%1ZP$ijF{?ChY<`$c0R?f>EJtHYw|+Py8nVxyGA42UQo(m6AfI)I9l(gNaubj{F> zU~C#ix^z@Rq(!=i?(Q0p?r!+DKF@pJIPW><`ObBH-#=WNJ$tXc*4^v3?sYF4S&1rQ zIq;3$+Njs~HF}Ez#3oC;ua1o-3=}NvuC6J}*EY(+)Fp=(a^O9@iwYA5cnJ_?32L*# zv50&~Di*!;vQSN_Nckp6)9g$zlA18ZVd69T4vbMVhA_yG1=&6-yGY1t@FR+Ug>F6s z^| z@U5BlDuQTs-f758ZOjq6MMpT={{K(W|*~ko}~B#c;4O z+d00xhtJMDEWv3HRT2;T!R2tkKLqhqqA8{G2ZcI?j6}dUCdnf=hiu?I6JXeaN$Y;J z#O$%NRT?~D825_sTppZTsQBvRS$1%CN}EyzpB&M7C2^HpOWf(*@= zj!rxlaT0u@cUtF)v>3!$*#%`?SUMB7uf%^J1vT}z;0J%>54TGAiU>u5*;BuLo7F7# zrA;9q9E9#2#VfX8SS4UV6Th2=TyQPq9xs9MuoRnEZ^8+{z6#gii)^}x3{`-7%0ANu zCn`hkHy_ed1BXa@n`DFB#tEc6j;;kG5>5E23QI%;^kHm5f4Pzgl-=*Ry|3*-CB_7- z#55{41VKY#S{UCz{APXs7+`M(sK8|B`DW)cC8f`TI7H|bP+%fclFju5xHhiLKE@|j z36W1kHaOAQu%-kGDX{2jLK1)Buh_$ zl~e^u?-QO$5zgC0?I~7?w!*X@)=hD@ghP2){zx?*AU74g0(@8fnMwFft{x z-_#p-{G4~ghAFI+NhjIediCynuVw!$fV^k2+c17I28UFd#5-fA&-3K>Rt zk1jpwE2=DfFP6LWa}Iy88*QAXmDB4=GpYay!`YI-$ zGy??+>>nY6o`9aNC!$Wm)!Fw1CfbNt1jrch2F4bc6!qzkpo^Tyhcx}w>Ur?Txk1Px z&;m3KL`6SI80TApRYWRRY^}o)ydj=&{R?`!8&g=|`bUU{t8@4jon(5Dk2bmih=59b z0BT784Z(%kuoP~DhqJv2ihyFtT>HpKTPXa?BwPf4e5Qsmd1&D1+J5U$wlXj|irH+T z4hG(T1avumi=Nhf@Ep*Xx7hEPAAN=v3^afT+2>ExP(zWGaw%k=2J?cW9rIyVz9Ux` z+A*iASYat{=fy(N&JOwWcl~%l(^^_aY-G1r^$gzwZ&dO5sg+qp8AR$q`Wu^_hD5Zp zeey@(1LXQ?d+=^3o#Jsd4p1zwi3yg7fj8u%NgN_y;~>^8LILnufMx(QK8Pp9adZO0 z0qEah|2Ku)nKmLHL|Lp7d)PfPiZNxx#Th=~h0KHck9XjxW8nQsSi^h5=6yz3&6u(W z%U$Sv7}zp%jfV-YL3X>=c49PN1t*(f}z9mi)t_$^xF1K0k2u>R`c&jN6rUts0z}a+dnhjQoO82S*L<`d7*j!xu&_z?f`uEQ#XB&&l{U{nVV@G1 zI%)qahsfN+X5WqCzza**+MmE7Ox;S0sRkcp6B%CPLgE;id(iCLiD#_CI2;_|N$n$@ zu$piix&h2UdNo441uu*HHM6P5o6$Gn_fTD9Kx0{o1k8%Of~f(&fEa>_rjEUG)C}+I zCz29h0K-I5<^}zP(mHL_6kJKSHD~PytJP-xkAivB@cwUU0Sn4FZKD4wa8~%!JCS zr3%6ZFeLgtl|)*C7hL^%u1&u1;8Z!VefOf$(&}!~sK$q?S-!#2{EilkKp{b7GxHM` z3w~oUn&|#xW*GWm!9G9Gln;!WyAq>~l$W*bJJ^*hf+cF-<2a2JkX}Km4XgR>YdkomOjE!>=7s^QPvI}-MTGM)iyRCV8UA287{_)i zu!k7`gVmnS^~$|0PsqXp*y4kwnCboC2uPcJreFTeI{r`7==fX$fDC>%Y%1FK%>TEm zcuNaSp4k`Q{(08dcmrtN+S=Qqos>I^wdS5*`|fxZ(qgU9>e@+v;{Ud)u>@+)P5jt^NZeTj1ZMWONAd{q<9fX#%q>rCBbKc{imAJPnX zN+ggdMRqQ&+yld1p9AhBk2eoS1DM*YF7dy+Ve?!ZcRi-gca<^2D+*@&vyWRVri(im zUhuk{ajnMX!$BhE9XcJKQ%9{ar61b&j_XEutaVt|DFH<6O=zMhVe=*hytw*F`pM==%y!JGj}=Hg@$eD>$A_cZCX7Xnx>L3P+vn zsz&!x^lZTG_lhd3475dYNvnGnpo=l#BJA;FK?4*l+dO%C;bU!c(SFehD>9z4JL)Ft zrZlmE&@`SkysA<9Ofl(~o#uM``tytWeWRsk1yf?^&-eL!c~)F!+EK0GvQtadadode zOg~Ec(!=`e>Z~1}yTT6{E*5gUy}eznB*7m2QSycqi-H@9%bI72I=HJCXGmk^d#_CD zIm*1rxHG2{*VsLga_Xo)KOo~(kl0ervR+Sae%#o_3@`P)*dqaxyVR*1$0wS`WpBOZ zUXIT74O*vBt&n+L8i)36(h8sCA3-$IC_PutmOtxPkn~we`{vuXafGAq{&^0FoZ<8Y zuf%(gKUy!8W_8)l-Bgh5m`?hZg|UK*h`pjeIyOSmR>Bx#w%$GY9F?l@`5l#vOTew| z#S6~&K3rMWkP4dGrKa7~D4Kt_)^s^Tu8bnJFvGDijkT|4dWqf#Utf(nq zsyN5|bi5aRarNg9tCY`D$mWJA4xEyloSanH`)6xCPi3PV+7d$R z5htXgcA4FyDEr+r8OcS}v@uohaW*qqm=-k!3b_0Hb%D?=r`3A}Pd%D9>qCyB=h@eB zf)CHhl_t|gzwYbXMi;TcrQ=`9-6}lt-0a!O)VN<`U}mJ`yy&%hbY4i8}PAafB?|iRVO*l zic#b`xCD{1UNC67!o~IF7Mko#w#INNQ`l?&yYZw1SIW00)vC(;G~rLpog})~Vlv0? zb0?fyR}@~W7TThd3mTlzq)74d6KY?Kuu3@HbwM zReO_9^lE-g^0qZeyU{en_&mLr=e}-uPSG=oyFYK`#@>3B;o|69_3pk|$W|@BjOi{P z-1?g|oLS9Z^X)4$dgW$#imGZ&eftA1-gpeZVirsMgWll{Ih!k=tgC1u!y98wbwm6; z?WY`yJWPx(9L6=B3tyV7l&3ny^29OtmX2ioBXUcfa=X`#CO#lVRGr58D)rkVg+tGy zBSNE(wKCQ_E2IJ0+Zf&0g;iSmtfRVs5Vp){U(9UKItl-$GeM8kAG)sNMmAI)th~_V z`D`5dvh$M`g#xwZhd^j8Kc84helgrwz-5ftd3Pf_-(Bd%UBTTSN$S<~mE?f8^`mCb z<8Lfh)Wfdxc>8^`v)(%;7joyv9v@u{a@RS1!dv0%dH%;)*<2GZ_0O}HUb>*)!|f^S za)G}@I>H)lA}Cy@SW0?h%-ES{ zs=N41M(lzk4Nd;HGY+4ar1T9)u!{q)eqw{$OZ`0qE255a4$zO_KUQLu@O+9@HYxbX zHzr?z^LJlKT^F7xVSL0?PUgbBx3g0pOYY#!?nC9mb1owIfQLzyhIi7l!h~S>Z++}P z=9B(^Tms~~3i`0WWCFKj2S*4g_y7qec8B$oHeS*S$e-}jLJ^L1(!qZ*zc^O_!t zO@HGEe}EbjwmToBJhrErsdk(7>=AB~AQddx;krec8-m-&v*MrNhhB;3^*mZ?rqF9@{1mddMzOtc!{WvShlW3Wut8BQ zqB!CpsS2a|47$rY!M^soIpE$;WfciGO3SFQZUTqi9EKe9+oI6(bmisojozXt^MPl3 zY7+3Dj%>PiLrnyeQp*mIhlVv2^imL*b8ogyk1Aq5eq%nv%(~g{m2T1_&u(lay|tD8 zUth!f8=mtUO8bfmb%vm|L)67_2x|ThtmmIElk14H3l-lEP64=?A>t;%f96&Zgew+0 zI(bl!h)ej>O1gRC-07qQ{VQvf#g2>wX#76eYcvp`^ag1gX`&z z#uEzTM+*nJHw#meXaTf$q?kBhtycpmQM@PXLeF8!Hu(gc++U$#TVXjfV>pa|aHx36u@1Q~B{BW&O|C$^*usXyFJ9ScQhXPC*@ez+ zEI~GX{M!;+W)NXpPZHwcR}c}KPf!Y$-WqptqS>4UDyOgQLjqt6zAHaIbS>5OaNQmr z6k}&D`^CXH>xj?iou;ol&>g!n1=}q`G9$6TEK9@3pC1t_y+wiR>pUV3JI+FX^g`}I z=H<}vZ!(!QuiW7EOH=5mPeGvJ3-q}ssDqtNCSd(sk{4kL_zb4a6T|APnJ2JeSb8`x zj|^a5%+?LM$4Y_Kgy4opjGzoy|FQ=dJ(f%dnU_C|WyTXWNXUV8V}5OiNRL#@Z37&xNEb@oMG-I$AohLuCE4TTBGm~+*bPJ2eCNURx|!ZpFntOM*UJ2o_OYdTe@Y^8Jm1@Q@KC_hru?``Q|}}bEh5Cf=xW( zw^4Imb1`Zq7B45(6W2CMFk(9J*blt<4*4c|D`vbYR%B_@NNM4fsadV!4dT8Y75JUQ zVF`Ps<%uRBn4t8)VZ$~_5LW0{Oi-&N0f}wbWFkz*z0v-GG{L=Gv{Rou$jGNFj9^S) z9mtZsEMZDiQ%Zpiq=MfHBM|{|`kl_1?L7xeo@LIlU%_b{Ts(W6 zPjn@{8J~=eyr2PmkD1PIy{z~CyOzYkhRN}vMt!*WcTc(}DWEsu5vplZ{o3PI7=zIQPjzwB(E!^}J`f8hGO&5W3RBy+Kr&W$21Pv@r;qbQ`Fbp`K$sLoy0F+O@J z)p2Zpl@UpME}>F$3+Scp$3B{9fwvSqFmlcZ62_G65V2Am0VA5Wn&LU4JI=tTetc3T zKHpI@diEoMsIQ$`L7Y?B6d=K^0JansVq8iGW}=NO!+3lC(ZmI))Upnf9#4SLEgVNK z+(DQ^1X+6E(CQ(t2@#?6E+rwu0fpw@pXr;0UM#ZMTp zW=p-Ar;LD4Wn_G+338WLgxW2x@yk<{e={NZK4&e<@1r_E^Ba z-3NCY$)_ncO>Jh#hw!rm42DZ=mUo%_Q(5XLKA;+(wUp^(VvU#pZ$SMY1QK|`geU;* zIHQ3GJ&Dp$a8a9VgE$bslhB6G=)@=6au7a{`4c!n^Nj6;7JSxRo0c6>SwxH)Hr^J6 z_KZU_|8g8*=ttAa*!*t%jNOcySSSW1gczD@8y!=oMsC=ls9{!iGZSsO=$C|cRHzq{ z8m8Ec7u5tw!5Eky3j9<LS);5MM#8g1I1Ip9#-=fRaZ@Ap|T9U4dr<1@3|)nXvPK z`JazL%5TU480G+ml|aH9!ZRm8LI%2k22#Dt{@IoTZDTTr(1{TS)*(iqmA(?X=Mg#+ zeC2pcXmEv4L-%Hs?NA%k(2#rV=mNU%7i;kq7-PWjERYb7R!Txw5M{xNOqdI+oe384 z1mXxBkntx>(pl?*Bg%htL5aG-wOZ6i+h!5gYza*W`o7W~GHG6PBo1&eQCD)u`rbLWqdD4wap z>x~i`{Q{jL4zvbW#v~d0lVuUKs4YiKC$Xzn-*(K; zs$2luJ38?DCBfau`t?bgyvarx5EH#>#=WhTVH;VR(}UM?<3I=bj4U>GVRR^y)F~p) z1?Mp_7;dg?V5dF&a%?#esf)@rvsfr=WPcp7P=g<>!jER4O=p$&>`kJ@sL6kaA4#ho z#P;j(y+GLB9E^b+5Q9ru%tIuC#xHjPlmu`l1V}LZ9u@1(WcXF%dL1Q){=bDG%9}%NJCi_E z1jvCHB$wyJCAf<_0@5IRjyXskYUlX+>hT+6Z7M&p{r49k9*d$b+Mz<3V7X$YI4wq`~T9sriS}=HPe(UcLtpEvByK;;C#R-O>S= z5#4xODWpD@koZExe5BLkZx4LWxbJLO?{2X4dacMJ(D~CQm13D8&Zr@_Q`{F;D6C}o zG`|U{llVw#A`@2=p$$uPw!}FV+|R#L)5sPtC$RzVAwX-Lo64xK;L(}f#!?Y*qy>@1+#6TMBJP6bMh!W3GZ{s*pSi)O#z1u; z;lTH15n@C1aB@uf8>Dsw^&}sqI$~kWqWv;fcwPA=AvFeorCY^_2|!~g0t9Pcp zgCFDbLhyk41iYr>jQxUg>UKnt?MyIkDA61&g=hrXAjo)PIe?cChftt?FJz6*!uR4( z8{Lp08d~DZoK=R>%N{~9Bs|EGNmS-SZKNTn{eeJoP=E)#(+ixo1l;|BYCp6fC#c_v z@5AQD^0uX+NBz+CPr<%7=tqluf){Q|X=e(8UfiI5H{Na$>s*sHJA|kAA`EePxB}H7 zgf?$NCZ@(7Zp2e!O2F_NR|BURf!@1A!;lR(u|NpT2MK($011p>c40xXFyxd( zix4`N@6*Jn(|~GU0!>vMe9B7Ss1KB3K<%$_my8n@=mpqBhKo ziYxH!~wRb`kF7;{P#c-e7DgB9#w-qH~p3dO~-iCDp+2NZ_N&^Kkr8ZgRx?VB*# zvRj{P(h1k9JB+m_Gj}X&!U*F&4u(|PL7n|CHFjTH0=?MmDQsYs#X~iO4gqqOO1Z|x zhf2c;JCix#tl@kQr_1h!S8|0PkM&^MD)D-cp_Nq^-#}Qp*Yi$1{R_hS=U?%m1BSMG zT#eo#)Dv!h?p9M@5Mg3vYf{Sr?=qd>BAR;?+nsAJ@(pAZa0tsBH7U|@4r#P&rcZRQd7M=2X(A`hl~6LftuSTA=3rZi;k5@dgyb^N!pHYg6Sg7*YRg$RQoH5R87{+>XbwD|UK0E)gQZeL$SJKmy@ zKil}U!$vp;D&3I&YSVBCL6jWd{NZmBgNRRJq)~-qOaQ|`><4&I^-)9-yGQRh8`HInWJPwB%$`H1QP%=L*=~`-v6WOJ* z7`pGyp2|C6&V5Mz4|oq7b?Y`E&sFGjyRBRCapxgbwAYN-NP}eH?G2{oGB zk9>#GpN?&Ix_Lc6U!|0&O}a<+YkDN~XPu+}D0l>25Cx7XMlYWX`L}S}FUad(7W??2 zN=SdTA`MLdOUN0POK0|FX8b)!r8Hk^%T#HVp7=C;_^~>lc77#|f96sCqiHi1OGDar zwoCkvuFFvi;1c>c=-oZ;mhM09D7a5YhUE$KvtviM z=d$1Mj5)_2kBV|(X2SW%zZo18rOVY(1|ernFz?gUinEOFbgV?bn(r$ zI8?LqzR%)y1;xV2Gxvu`8{SwU?s&>%?lnFmb_Rd-YB1YnLOQJm z=KM7qRyW3b)H<*$wHv4wbzzn!#w;dpr39wfJ%zonY75_!ArHJx{XG3ljxCc+-#m5m z?RI8Z<6PgUTtbu1`!uOAdcm99GVfi5(mj7Q$Bnb%! z{aLT-uhj6o;M!F~|Av^sjpK^0G4!3!(f!y)>vK-gBvzTD&x)?tT!nG@_*gK^LE|Z0 z_KB%oB>k*2@4mhv*ZlXbEfYHiZrXoYEC1u)w~A{c-*z-JL7IQYUy+c{b92D1YM(!?U~qU4>Zp~dX+ zlgN#)5^l6sQ4;191Q%0^TZoNh&z+?~*GKPXRE}4ye!e}zkbq2!yr9?zN~i!#X3KpBdJBE<-s-IZ^Z@$8A;H2G8t4rFV!L^uZ$FoKLkA#A`8AAA;M7l|@`)#94Pl>HDw z>gq^${$R9BeQ@%79EGuCP{tB7G+8RKjv&ykwBUa7?YYq`IxRy$Iqe>ZxsWfEUt~c! z%RJup_!v7)DD=!W+CD)Nv@9=Oe^5l>*Rs-W0WT2=MDB#U-NThKKm(L zcq%5ZsK)J7gJT>(2Qfq9f&xP*AS|$q;LHXw?KEI$HM25z!~@+5Az(Z5C5kWYE~-G2 zlmdcDO1%RT#o!FBK8V=n$MQMs%x*t&b^756MKL0{d>6|hDK4qYkPxcd(v>~9oRa*= zRjp^ZQA{WIrA!*cQO$PDf!rqu3$69{kE6H})+($wtjFxe6?+;)q>f|P^;dU4d^pAx zBL;7o4*Ly2_%Q;4(As%q_eb*V`!tw(zBLh|=pf8UiJy_F_Grp5F3M-gBV^O2SU$cdGI_8i{Jw(%br+9v;t;N$^0QOb`0Ce!vlY#pbx!6 zNZW0J;7C6-ELu$7D*m&3kcoVY%OY`_-StB$Pk#tx)3xD=Tk3Zox8uvB4+(7+Gnw1J zvgxmavoA$gIr=xZA~Y6>^ouNj0^BIbP?S#P%X4J|h^a%Et*&^DUgQ`wxw6$jK@>U^ zYtq{>bO*R5%b*EpcEo;T(E_yRK@aR#afOTso&+qo1&6?kBlr~W0hAKW zjo5G%*MtECjP>7D^-t~g2b2;iMpE^vGN2bq;ZG$|6jGH}M9Yft5>j%W|ebVcZBBon2xgzsD7vov#c^@QwE{;@ zsfG3&{)0IuoJ57DffG>Qm|$k5j1YqOKbXkBJpfb@_x7lRVd>|$>WL}(uy8~%ys@1T z%%j;KU&(z2qeaPIC95!YS~Rvv!;V#|W}ZJpg^vXvPcOuC|Z^K`rXw zYT-H2dZdq{nCjQcJdgb{#hC6IGMnaDS*EC=(u2ZR+t$N&v|Du`w(=w{&rpR1jqyUCyy$T`h-@-jG{ zgbs5|*0JY6Sd1wn=ad`Z&?_{6N}o`#(DPQL*p%`t%xIYsTZOTrF@~U&? zn~;UP=x3F(+>DtLafVE6vK*X~3{Dy()&vjl7g&M;?zS8>FjB-u6JFJU10iI~5e9Fl zgQsu)@J8-* zq!BD1YSRo^Rhw(!;CU+1)_wA+9(fS#)d9spJRDBV7P?q*{QjH4uRG0cZjK%_6e$v+ z(TjLZYl;LTSB{=R!jR?KFaO4nqx_FeLRiK#dzD(YQ6v zw<)|2s7XMagFK`cT37(oAIDy(QiS^zptB~Fo1jE+Sc|)D@*GMv7oJiGbl-;4#+C@$ zY68#}RxSrbuj#{!ZG5L7fD5yGp9NvDq|Ak&MuP+G6+@036IhN!U|Jo{84UCL3iZ|z z!WqeaVaL?VFu!3OVqPq?9T6K?g?ZLWJOdg63BCla*I`|OglAk3$Lu=sW7ro!@A5-d z2;D49L(&2-O-$j&F&K)d1`USV*Bm-vFnj;C@C)h9-cT3fPfH`1ln) z3+)w9R(Q?c6d}e{9Lfm#POEP$C^Lv%birwg|73x1`UGg*gK!P(7$QE00cxzE%%vh+ zWKnHZ7k*<7+u@C-Pe-qIV%nN8ZAoZT0@TStcnT9-&J|L+bcB2{bKoz9dKuyet(iXS z>%(6hz?&^7Z%#qL<881b1(N$-z_1)py8x&u0V-%WVMh)Lq$)Q;m0-`oEf(qlOA&$5 zY6KAC;d1_f8ZS^S27VnLG6K3Y;VBL86W?EwL-}1^C~8>??8AFpyu1xJigyfQ zEbqpEA!g*bWL&DTLw^*eNIJ7JK^)(!HidHEOH zXGsEJ@_AvYJ+JeH(Vn3L@wjv=@shZ*j%*>qeAm;S`<%I2*yDm-!;-Ep?q+SS-X<>Q~sr&-6P``5#6`BrlI)w&_MlkR-9nJ!HrAvyBpmq>Zh`xEPJY9w0;@# zV6V5V^D1TI?!nJzxU6+Anmd0I=x>efH65Qsu_Vd=JACCg#__Mf8&rtmF9UB)EH)$% z@YW`I86{jhM#`U`k?c2l{%Xx6DQgeQIQH^q{u#CTQos4zz4;yz<%%cH_}so*_tvlP zEy_dkZs||EYxl3I&H4Fc?M{n!wuih`MA_a3_@~2pdD)-%(ZFRq{hK+HjX#(z+GoiK zZFAd+t2e1CfC^8MI!h9GUWYA3fE#U}9LQgbwr+P<%3Cuv4EffO>u>&U+g%FHymshu z>|WAGm1J4*Luyehc1Ys*B}zc&#Za@gy3J!s{g@YZaY}BTI4|@C8Uyl zxiz01h$@VF5Dl&;SbkC*U|YI)|1Krl)vKpJKHu+kTy~(~J^4k@K}z#PP?4)^-mHb(v2!;9ipRvMKyw2TD;zJ~8;zQqvcA~rML@x}CtN}aF}c=1Di#fgR_lH(ek za|$ZBK@!HRX5R%&sW^hb+@(Mobo&EYf`Q8n1ju( z*pIcDGC_TRqMJA4bN5M)0}Ks+RA2XdGKusP<&_V!_GagpHS{Ekd~ciDQQcUp249rZ z_U5>Ask5mE5Z0&8t~j5ry}Y`IZy9hn&AYs^FOoLwuRDE%Cuf_Uhq~YL;X~7>V*?rQ zf>N@t>QrnP=a-I_f7s6%GTwP;?>~RjBV5g0-nYu?`|B_+=1@>s zbi!JZUBngV`Jif zT(C`fhR1PtSXZ5_r_;u$u{^QOyE;?oli;)WybHV+-5!Y5?8DBmZ=}xJ?N>+BGb(Kt zJ(oKr?^|os9xDAw>-v!Wb$Rx2R{5_<(ieH6I#xuUe~FAuek^n9_TnL*X`QA3rt6bX zB9ns7iU}(dRSPx8xX%{yQGFF3RLPPyZ2aTNtfMoZ=58JWk8Wba9;8l@-}97xb1WL7 zu{9McWtF3qB%_XcZ+&3}y)2@hmf_kvR(zh>i5l)jUH?r{g7oTxB+?d~++-W$$4H{a zJLdJ=JCW|M854I?YY_WXg%O@@@7gP<+T`e;*<6bqyBYR$^rQCsdC??lQC6O^$o3nP z2T>>4X$0w|{C_Y{wImK)Qx#}?_WjcP$%M!zDMR@lE{8dm^3w7rl5A{baxX|*zYpAW zS!{piaK2^eR-#{z>ogkJi63WOeW%98Q;KXi6y`81-nv8WQA%?B-C}Tvtx3Ax={v~{ zhPB!XF$PIY$6&xU%Q?i$2vkw?9BF~YEgkdIU6zH`gvWB$yWEoBLSDp+T+=cZb|+!5 z-#YP*G*!^@`%5L^XRAD{*A!gbI0Opw{WwIBn*I9=O<^WC-JUbR7ek73ay}|jeVKo2 z&59VW)%_8A*Xj8n8>#-s^PI z3M}1`d4}9GL9DENy4X{!#b1>k=qD2CP9L8#{q+9s!BE!Ml|!0S-Q8c$ z#G+-%%x8_gg0b(GGJ}_t-t=qNt&?suzZC1Zj@OXf-Oy-cac$~2?V*vg^uFUN?W@xg z3o>JFWqf1JhY73!8frG5xLw& zqnVD2*TYj~PsdNUvfQLme}-_qx5iYa(=nf5e8UtoOK{zMW%pxcLoP{pL1Fe+)igvIw^6v++%k{Vey~+pvAZqN?t}6sEYY z_S=)MLJ~Qmk1% zWy*Y&?F5LS$aK$Hu#hlSz8JY8tvPR4>Dg}~{CroY^jqg%TxYGPNM?KRQ?e4`NpU*i zHo8i&T=WX=3w z{BYT`vX)UJ5JiV&$@18;wVkOeH5utj+xOmBNiSLMA(UOw3CO>GWB+xT?BloBR+3%A z94c0$M}oheJumXMNVH;^Ur27*ef(o2c)-65n@WEuM*MQ>!;Cb|YkAG9x8i$E6ltje1d&)mqwwVhA5F_W(3vx1r5Y*?WeL-VOYZK(jxjRyQ|*D*R38= zOgo2X8My=y(Q5^3h^;q>%2krH*T$%*Di2Zfx`6s~ zrB2lINlGws=c{ppB~R33`lq*C0tY$;C3-YqZB zwUA$IdTcw6b{Mx6y{&_mS$eWMZFTz&C0k1_j54u&N&4bc(!$5d%u{3-og%uXkl39o z7Yg3Yk)O&oO-#7+a;KerhbJ@c+|8?27ap=!ffpEYSm~h|{*EPmPu&8O4*^+{eFx!D zk1j1O{G5qAjvS+sr$gjdQeVj~tno4Wc*#tWL}XPS+?DpyQh^D%-y<7QaXVNT@H4&R z|6T{)nQ2@fRnRu`i1tit_wsw6Q@uus`WxqTPnltxf*7Ma@>R)8I5_sMKA<`8cicA5 z;r1ZdHfAezjicsH5BewaGs99fQj^?dIs20_eK7e5HEArmhyYWw;yp9M6Oj{ zslmb-UP)PwUuFmzU)YN*Z4TabwP5eFm+KTcx{EZuUv%V(D=hkaC(8WVnJT{ZQ^EOPlF#`odH8k`S=7+7XZ8@;SPTLLl58iV|wN$GddtNNGd;B8T$Cq|=V=lfO z^tqzZ79ndw3C_rdmoNkjW4^oZZ&XU4c* zOyQy``Oe2Rx1qO1;GSFQ1GXZz{x42(t$7Ngr-dpDbQ+(V*dGtR=CWZgZ7w5|GxgAb za<3)CrxfU3<kP-)xTKU`DfDK*_cNM` z=;noI-zw^SiU#(GU3*$WZo91G7N(w@9l0qPtDx(sqGzQZ^PI&8(Huj%d(KB}K?tu` zQX9Q!aJq5&O$&>$f5raF&)C4CuT^zb_nyp7QCz;H?2S{DOlVCzCNnK}OZ#{6wZK0X z)%#m~Eg~ZDS7#d4TBFuO&_UYUdRZg`M%yh$*B^&g<;Tsv*3uc$ za>8?PZZgN)Excwov>bFc@D#ZmHqrnnxb9AHcXVIfEBSFNNhfupdvlSq{oD16s?S2Y z?MwIaj|fu@W<`LWqod`q%CnbzXU8sEN$MUO(xj$WDkCFbJePX&K@+7cbhPJA7n~?N zD~&YJEHoFp9gj75?F{bd30^rk!^uBRP6v>O3RY_vVzvuMKtUx3ic09jKsVO4GlaDZkzl;D6u#)eqx% za+!pjDw(5nV?Wry`b*Sh{j?Co+UQM@&o)4jx7yhn4IFj4{@YHL*ou>1(zla0zZOZq z9=F%%(e4;=S+Dxy?l82%qv?wv>PdUf84J8Sk$6h{rTjNYca!( zs1lBoxALlQy-SUWcl9qsE^ZAa3YzTQeQp&KzR#R|e7#fXz`~z%f3K*%P$Hfrntda& z{aYWYTy1o7HWgU|qvR;M6JWw2j3(20=7%$!&eP09c?GJ=idjE9XnQmEO1Gdo@#$yg z*c5NQIyk0y^z-|VKPXkHiW*udXf|&!=L7f1B%eG1fu49CZplxzlH;PO_FpI^IyckY z#$=hL8xk@FY1sJpyDz`yIkI!dIr7UJJ*GL6U#N0UJP-5_{@|?!yqxi>N%-bUDUr%0 zPZnslNgTJnvdkwV-zC<1-HhnB>74=PQQl~{dqyGmE-o4y>rKME@)~yfses2A{pX_c zZ=S}ed1lfSN|-lBf$+!_O!obBAM z#819Df9}XLT&k=4Sq-_>2|i~cPNqsPVwl|Py;*{|0~j~^Vk*1RQeTQk3`~nZh%uSF zZ=0IsOIpT9m1-KJ5=Ye)aC_R=r5K;%=NV$!d?B;$)#{0d`A92LS&HVBA;w5sL7lBx z9H%dAfu-Z^GIvqUsOVSv7hjQTSZZWhR))b7=dqwuAqty(Etu0I^fvPL(RK}y>z)R^@KK{Iuv%*3E9?6ZZJ&d}-nk+Jp zGS@yyQt5B_W9505xR*9n^7ZE`hng_Ib4yL~D}lO7ab4eivaNsw=CE?bo0esBh|%v?Ptx$Eii)Mw?}?SyxmC*5@FYJEFgs>r>aFtir*yz3N~Uwlc?(SNLe zo|7a({oE_6dLPl~XV(orCfvL5fU>>*dAzRx-nl3<^0X#-$=mR-jWhaJjal3uUMJ=S zzxO`J8ip#$zT|^3i+^~L?Iaza;9GiT(lgx3)z3-pWLP{e#WAcN2-T=)oyAHRnGq}B zc{SY0epc`Mvi$L#xz|sZ7+P;q@kH^fdtK0YcUsD(_1ce@JsiT%Jc88NRw||qddE}UVbP`++(fnQO|thw14Si^8*GMHH%iA zlGE{ILJE|TSD%z^&-#tnUV3}I?Lti$xBU3~B&^nHpMa1H7Q=MdI-(B81CD2jhm}Fx zmqN?s&+Zw__+Q((eealRLx%LFw$eum3V-dd6q6W(OMLD$k(zAIE3ec%&&!8i(Gk8war@j^s?aJQzF~nDa4$fc zlu~`+)zRULr)NDL?D#3^p{@zu(2)6JwsY>sBNmDM)NStBQ)D17{i{I>_L48iKt&-5(@i`~b$2m*>JxpRET61Mbn^oD zUE8;5_-dDUAKAAqtp2UNjROnH8>MK^DYYlsag1_@{M$tm(a< zW+r)SH7%Hr3t)RsA#Y+Xy%mrcD2p$|4{cHP)%>!-l$SUKtQPqQUpPzh2D#TAVr!8 zp(ga+Yv_t1N|hQq0)q6=K}aaliGb8d@4XX>fdDVM_dVaa`*`l%-+g1;Hy#5r!uprB z=3MPJ=bA-q%;Xk8^CJ9tb5v+QbpoPX+}drgt{r@_Z*7MFVFY+iKUk1R`Hng{cwH%wJR$~A(wc#_)#l?+%i$oUF)Jtu#QR=-wjqWnns;EF?U601oVjCm=h?FY0_ohQ zdr$9k_P?X}3Jx=?b#(O(R`^IQud^h>Z&dqGk39MQ?3?%5J&xTZv@C21cgh9tZ|LSiAqqGxUbCw>IAE*og~+ z16X}a8)Lk?*;_b)A48XleDF*Cw&73n#k)hm=d`zMQq;Fxj96nIvMIR*pWwR}EoM9^ z4HBI50pt_%t0ae09^dADN2vZlj)a-ZfMc1mg&z2#zWmYE&zIdoYy_33SubYZ$F5%40G!_K19 z{P0D}`>5=et%5vUbNdN?fQ=Sk z9Kv4jRiq4h4o>hug6{iJ!X$IR`|Lc&BJb?@uNK-NrlmT z7UfRx)%Q^Rk5&4H*Cxg(Tt;Z33$G$n%xM@lFxSTheLGBsFwQP9ERBqprt_k&!Vj0? z)y_meeK|g*(Zp-c@F8lfrAV_cp|%rA#}+><{1z9)T=gt|?>SLO;A*qAidMg@bH(i~ zs#E4QD}TOs&!w)<-FQxXk6~`Ajyg*sF$Au|R(_tF8L9G`phcZgy)f^LzO6rv{m2t=3}wUWXfN5NtF@@w*E2ajZC4A@&{074WtAEt+Rumc3Vj zzOW=ezhd3!n0l-k{`5inyF^wll3og92s`z5LFQ{6?D*8+s0@Mrd0}YuH;MVjMSO$( zdh2uJ!_SPcOvN!Bno{8^wk2v!cEWZJ3uAP`<}zZxV~Bv@KQZKQnPmQoi2eUII<|-y z&LmUtXWjplwNg+(Ncisx5I?h4PSn#h?+rD$d9NWAM^zgJ96xU`PTAD9@H8xF;xDz& zTnP56%PGy%Dc4Y|_FYIx32|H(K2fM}P?hmf|Bu|QI8)|dzmG|>dSeRw78%d2RF@v_ zdhLE81;GE!2kb3k`~1lAXlHDZ;#2g@UfAe{@4nTt=h*^ki4HX2D~(;50}H~Cwqt|K zGvU3;H;l?7`I!(=$S3HXD%jg&@ZlCItPAlyh<(xX7!B+Qy&W_cLI;BU4?ld4da!SH zPz-gq``f=Yu2Ss%7flDi?c^Wts&uOEs`RPi$b(&{c`!(L4M(cmE+sq`_cwSziK83o zuE3k0*0M|gjjNN3A7IS~zp{tBhE7RXxcUayRB?!SF#F(M(E}++}vj z+3Kg?ynOUb7-jus)M^8GBHTQ$sCL>XU)rm1FdAy0*tr+F)Q|Mfk#hhzAeg8k63^yl zKB1f&<7HJU0=f-O-Rw+=Yhdk)V+!h2M}skbJjY;EC}v9IQ?Ab7mVNGRqOz|)dR!wZ zRxpb9IxF|)gnIYh5TFZ#?nE_7K}r(LDH3nV5DA)57Qujy;<_fuYr8jL_Yw3ORss>YjRO!`DNs|2F-kK9*F@@ zrxbhO)>h+SL*#p zUDWe+y!HiN=$Ak`ib`WeI7KS+ zb<20{wy3sf^A|=Y5hGA@A18qG)9{`o7U%@5 z`)zHm6s7G4eN3yfzvzHFN z%2eh4lI4DcPrrTsq1W(aiJ}M)Z-{t|c`|=Aqj#~C-Xl8h&1wxJW!wr~TC0KefevYX zLis(c#v?uIcMZj5*Jjxl(00vNS8}6338WK=Q|(EJaoVTKun@M9%NY&OnzW-YNEnpf zs%B+7VP~G5?a&VNf=!xErW6~H`L32L=zCC#`9zfD4FK^{E(4L~cyCdw3u@Mqe}_d5 zzzgJ<1^d9e|78S7_@fLhKc+8|QY-}fXTzB)VCBEF&&YL`ZQ1G(?Oda1&xok5=?3-++?Dj)`;B(~i_<6Q#}mhI%7*xm8lrn#HB3kR`pCd~47st4yfQ}rtR z#wkh+@1?{7(Y+Ol>z0jZ5EkSadQJ*x=W>Hpzq%-gD7P+U@k0u}ryHxbsTP4Gcwwcs z#Z*7%`)=$Bl5A^7fCjMDMe_+-16dP;d!&IxkkZnAI{T5lLy=zlvAO`p9H`TfJ=@Hx zTO|}aL{;GXZfW1U3B-U=gr-_H4)6xpxNko_O-tG9#2Sxska#LHjo%T$JuZr^n+>vD`ayGRPb0E zFf*8XrtkO~n=jnV7F-#F-P#DMieB9dYb(0h58c|p1RdLgZ6icwXkm5Ur~5C?fcF8W z{z+lbym@v=$eu*)tv>IAXK;VNsYZ$Vajp)uILH=)kVy`jOL&bsYsi`3SSEtw`M&Es z13vO(-~UoAE5!b|8L4^0kdZlsN^h2?16jB7g^LqrwlXrJx=a5d0I}(TwcMwKn~evJ z`8)v0`&6$?2rVr*^or68?9YX-y;m1*_SYJRdh(O$luu!@A!Lgqu8sK*9^C_tf`s+K zcwgz-k(ogD3EdaCi$435Phe3ft_cJIK-<`&Hw_>6PM(B;fsLcwazGo`2dG#rEOkfW zsNRPd1cImaI~#rOpJ}#dfjMoF0sx@-^%2jD5NIm23PLD06Ap|vgZp|KgFj=3XRFm< zormPx8&E?y0YDPqdgN=Q;W0m@x>tc76dR`l-GSH7FkgaAg9clQ&?nUJH2BAE)!H?7 zh(7hg(1r^6KHA!N`#1!s0a%;$luWf7vPIy7aC5GAc6hIz;ctQIIS7CJs4;*)KTeqt(G3W>+ zkPn?7X;(GJk$z2D>oq70Xs$*KTEq1(Zj&Yk)kv0Q;Rd@kjy(XlKOI#T^j8fd>$$Cd zKi_HzX?~rqJAbH=lH$P|P zeYi)i*RD!4vh+3%DR8+0TDWL_z5WaCjlU|MTbjZFjP{?c{0XACjB&TH1N?a5SB`L< z^CvukeFKHzkVMGw&!Qsn%;BL|m7hFvAPTG1G0q;J-vQJ9SX_EBia_(Z1geR!NdFv- zRFN%JTkxxDm`~%t5(g+|r->G$n0^F?kO|2c^@^_Z$+U5mEC4*`zSu*4HR>nGp+*d6 zkiu@l-}t@yzXMrYVZan0b~Yr~_ROsca3cSY#{XMv_+MV&vW0aLJcC+ZX|pZI#A}k^ zmNQ+dDrxNL25vBp1HXcbh_45U_wLBRGjAtGZ|z4?PDCXTDj1m7 zc+1^e%SWF!*l{Rxfp#J>n3kSoXqm3VhuX^eNfRlM1{#gzsQL9CG^K7h*?|`)lB{F2v5z*hZ3&At z+4|vIA9AFl@5EPnhHCw^;BOunm~^&S{Cf8kw(j=oMkGbJsCUh6O|@5vmG|$g{wNFl zeEAQAe#3bZ>mtWKtMKFP1S%-r`U!5u$KY00+@eKp1p>0pWYOlJkSlb+nKSg1R$%4D zMo|uB(U1CofY{6(ElGY~#(16$tOCftcYNY&(R;S+15QgBrpe8V9b?Rg%`3?BvejFe zk7x^k1C84=XSNsF4)*BtGXLdu4B-fsX;7VLoH&%O@?}S_b97ixPv$I*_36o3@d{Vk zKAu`2y5Uvq7TZEA4%!od{^2aywz_aZ`w7qScVgEs(dVz0TDZc$EB^3Al#>xWx{6;A z9&Gc9Rw+L?EsaCzEns|;ezzP}%TgvqpG`(Zyq6Jkcc$;@7Y7Hn3$@ofs&dXo!o#XL zk=?Y$BE44vYh3`*M-rqV9Y?!iYHT5t$w>q~)qyq&I-ZOkHldf<-8!`)Uu$8&1`83#bMdT{6jJZNlAjNA#`(6E)O zf@er~*bAtIYB>b&Y{~#s$0u8@WXF4&2Y>XUCY}>Yo@jX-h{E1+ymv_D`Gz%m?gJL8 zW~l&=WN5Zb9{X6=22GzGjYtX)y`*_-!TCI;uR4FNSv;35;jLKjvUON#;20HBz}Z=k z*if`H)sP;>ZjH;rlGTY65IvJc|{*%<`SEpC)XA2I<- zQ@y`V9n9P|CCujh{Ox*0n_TNrVSXyf1F;9m0tAC2PZ5%!)v%`VF+2e%Y+cxxPgG4` ztq%S5Q@t&hfJ4{Ad##9i@`?oxpM{=nxxV%(gmk|;8- z`M86Xdk3AQXQZIpZ;eaEf9wQegwQ=UU*kAb6mHx0#Ld)M?7sjClZ9oRCnnh#mpaIJ z3F(;Xif_B3d3uesl-K8i&5j0bcN^>t#zZ@7D(Rk9PS=X;;7pX-*{(a=PA-k4t2T*w zl`pCDAw`@WF-qpZn%#qRm`=0vN|>=dc>nUL_)O&$W;w=y8YzmkAT7>_sH{t_a|Qd+ zhG1IO!ViR>`oHMlL4%D3KaCNkIgCeFW(VUkPf;>@&;V_^mLCpth08wDPot#1#fs#p zcqE|qE%*L}vyqDTH6Cj$X<^J)!jIXqnNjUbqW&K<`z{F`4ejr689{xtm<%UAb3!k7=&xVOkYCvKXt_z4Yj*lk z8$YC6soAXWVNDe;X~tknzHTy+{N28CscUIPF~oQ7?%3}l+ooKqR-J8O%C^f)#G(kx zR*@$iJg;fYx^`}qXue>Et3HL?uOFRW+ zsW#@`*yJ)C-X-AQAciake!a18vgC2C_ifHB<@7!c>$&u+I#z4yt_Bh_V*y$jLh-qv z2&cp&34GUs2W_tL%ubGK8;AB?=NAIoT5|@sPUcLbF}0@s(ILLOn-szRR=bZr`|jND zTh4K(J&~0=$$X@A$5V`VNafNKH+S5;DH3HZKu!@hM+> zY%tEB*7uc8=ze(0r^37qTcA5)c7!g0y-pwq@lA!~zM4e=Y&V`5Lw${J<{sKVx{}e^ z-pI>&74-qq7WRtCM&W)tVXpi4=d{c*te53~q@UWBWf|w3Xx=X4O!`J3=}$3dYX?iu z2c-nqqG?yZaP-&oICqKM7uqky} zg*pZEhBQc#XkO}EcezT^L02U*3AgY^zxTgG?L}0!s}#p*BzyIO)S#Z#Am_xi}lyuB~o_ICR?K`uiUjX-KE=*nlj{v^PrKr!OtlNAh*UOkSU#N`o z5^usEbU<$_bkBd5iRe1yQDDad+29}eGgSL2eIIQ!mMkyXs3U(&&U<}!hVZ6N!Hx3$ zIZsts7?r~hCh*KEd#DyzK+az7O(Qy)#gFFRHcyz_yBynK2E$i+Nh|=H>?{Y%>*KBu zZ+X1jnSmsn(pEY=xpu_!(9yv-BDAx^le6qiFNxaoSyj6bib^`+X1S{@Bw@_2x=9U7 z@t5~}`&qf=BW>&7nAvKYKgRRyEZC;+96?Di5^r%I4VxRs@?_A=WQqC%d!ho=W0I9>i0X+{dh!2_K4Um|vltak`@>q=5Qd!|xgxd@Fd)SLv(ujo@)5_?c;s z#u1^oRP1ekz8|9@sn(aNzWq3KysY@*_2Vr&f`Y4o-a?!77#>#|r&t!Rd8U8Dac;*N1 zNlz|g(k|>+!ctnKw5e zaQD5lurs&*;F9>((3N-1FnuqMlqclG_o>cn(RwGs0nZ8g9I7IgOmaf@YLu7wYdv>m`HTl$T^%%yYoc@xgxQlX2<%& zvTL;Au$FhH^#0{yvL(yiyAJCbn)g1RwA04fa^F$_t>|dcVt%{@4S0DJsW;msD-FdL zAB{?`RGB;!Ws9UIRlErk`x1%GkllY{gOA}Hc;NIlYkhX&va}F$N0d@%oH4zk+m{}? zC<{g90*gtu!wnj*WX-*YN+plwAM2PIKEWVIeP=nFLS@4+x;Eb6#7ks4FQ3_ z@9m%y-JssacU{3Rs6F1xv{Nt;_V^VGrO?%G#_Wfrq%k54Oz(aKq>04@f7<;qog-;Z zU`%p{BY_f+<2fU=Xw_<# zSccRh)%%(-?#oxmFf)Q>wE=zyWYKyqJLTDmAG&q$ZRTj_br8?rjeSh>Y~PxcGIR{SADfz=1>Glm4b zgg3!q{JKM*$1H1;&n`P~Xzl-wV}e5e-yHUc;|_ap(Ki0mVUK{Y@ZYu;+nORZoamyt z^iA17i7vLPPCqfR$}^RE{H~*^e67?qa)Psqrsv%vfu>`zj$>@6lhM~L-rgIXvny3( z(}?}g{xjL+t_jQ~Mf)*d_ndWy)d>KhbN~<>vNzuz{?{)w<*3$~Mz3=jXYBdrnjp*u zu}!}t0rHleK9GedY#jc`>a}kn$OVl zB1I4dv-5ChU(;`E`PX{?Qj)kiV&KGmiW!~ki%pFIM}_{rwSRdEvNx^^yN(AtfrRi~ zTA)DyugaUi&5lV-Z$8loar<<~Fzaar{YjI~$cN(2oxQmR};E!}( zgvVOkG9cu9If+QR=+#|YnDRH%@?Ap2TId@Qw#hvx-5c7MqeQ=`-oK!;NB0D@cMn!A zv!)*Tmst1fe~`?;FHX2<(t`reWmLnC?`M4vi4w5LdJ9Y*n6ZeuW*_n$t}=Z2vm~UH zv`MflZ)be8Yls@)jSQ~d4+@k9#f8nk11Y>2UY2%FdMpyq*%#pZjM8CTze@t{rY_Nm zkb6{u@wO<@M zYV2^^k_N=yoOeFCj>t5uuc-+>5A*Wej`s#SI?C!^?yQ=DFFJ9vO~A)b)l zs4Q9F-uH4y0)6CtzZRPT*Cl;5-tng~P~n9qx|=%wl#jW`axJC)M-(J-+X zA4VDJ2xff;INSvhMiXs(k14Xt`&Ru@X*rx0nNHTkT+>zkQVgER?_6*W8tP*txfWX5 zS7g&)g*i@L8ExkGp=mOy+5`FEb1J-n0he^(+Y@2O?n!};CR?DaVZuh9-IT|)k-*Q| zvpl{{@#4G`I@>zFZDtMAxNxdVSz(Cfnq{X7+V5fNCYmoUyFG+$&$GZa(*7LiVa7xg0_Hj?XIztO=oRIS3omC5=s}F=)!kE z)S3qr7_iS384f;cE}}g{oo$?ve^&*q%Y)!G(;KH6rYlgX+Ch3(;*y_U5x~w6*h*0k^zOdYQljcjr?f6F0MkXT@#5Z6j%}!TH8UfL3YB^ zTCK)nBJ?=2b<=eN*m3I>A=eJQZN?PhG)4MuQ+7XWqjhm2v))Aww08_PuKTJIr~qb` z@rQShqPL2{V>X4NX+gi+asIo}glpZn3}7Qtv0^$O>n4Tc$zS)Zzf)434(cCe*aC?5 zaLK{S4;4>S;qwRdhR#(@wuk{Sa|gsgOCB~%vZ?;pfK{0OmqGbIskD8g=Okei7XSD6 zufNPB9PI|VKX)Y57|GjS5BIVU)HwC|8#)_5>~Ze`R1iZ`iQ2_`E1Stq!8|QWUI{A^ zrMJGu!{-Z?`?Fu{-h*Y2$wR!=K*Fm{Ofu6RC5ym=LsP56Ex=T-qBdd<`+Ih#E>$nu zIOs|;1mRS`Kf%JU^jXhu;gsFXvZC~|(-B|$VIR{$q$@O{f^%N4BHx~4I~o~%rwe8SXlv%s@I0ZSaJD^x7IQ^M4?Oq0~5{I&}c5( zPZM8UDrpdD*0^$m!1Uuls#9Dsa-m_FWPf99>q_cKkaVDoa>w!ta7e?;Aiwd#H?X1Q z%1G#1xa1Dq^InySwQ5b<05Oag{;xb=H=L`2epuYwr9{3xtqr;gh??+&b|2>~92H$B zmjKn?fq%rQ&NtGL6VMAO3H5QmCb)QC*fsFO^LSn~(U$@gyZw>ehpRi-#qM9*cE8(P$Bo2Y5;EyDslMzwuk78CH;O^(`Kv>Zr_CPn)3)n76f@%&81M-RTp=SP zBB@CnPZ7?k7V;nP9wk4s+^(}pd6QMYAorlxi|R>d>DtC}0C;uneT5ma{1V04;dQr2ABn?$EZH^=XRU#Xf}-4>bWs_k4Gu5-S>GGx2Kxx)G~vQ4>YQ_ZEH(4e!d z?@Lf8@r&e$4Y9$u!keE!(^Qx`O^$!g^GXIbCUKx|hxzwZr72uraa& zXM!9C!EXlb;*8mjWNR6f$T@0$X zf%GtI;jaReJsVtZc7zpneY+W=Gm*|0rQ<_pIIjawe$T+Qn_?FRoU@WM2WgwY!$hd) z4^m$2aejYlGi*P3-r=`|SR#b$24Snq=G1mMu^duA8abk%@;c?!dz`VnpN)FcX7WjD zI6#D_wUa^)%l^-W%T&Z~r33Fvu`C^?kfqBOu^%R7WM?)HmqK<#Z#2v z`(QhPPRr3DwSJ@Jl}AN;7>3G*S}g8n+w=~^a++qQBUQ>? zJ@@s`Au|U~+gEy7ms=nrN3G0f?lG9-@ZgG|oe_y&eE1pvbC6?nM~YCZw0p?@Qan*fnuFJ)riR(wU1D&G{=Y z+jYe)1RGzX;@}ws-;l{GVkG& zJbD1y^NFOhk(|*ek8+Y_MB-3k2PsR2b@vFu=QOV}Q#uoeJ=)f@^HEcHIr>3ARBJCz z+>he1VKPmM4El6qOEechMrJge0%iJ&e?pKs7sNo{zz3PlLebm*LtI#5hZwLLy&Rd9 z&MwNir0^N>lnDcdcK|?+JutecrU)>c{s<=sa&JVY0LDvw zTm1)WM$pUQHM~t<=^WJ%rj7Z&893T<&qs2eh3|5M&9Pg`Ky=#^RdM ziG;69F$>qtv|#cRBfvSQ?}kJ7!=Zbs0r@xyrbq-oj67u=*Jid7 zo^0&z+jkme2& z45(nAm~LF#wC*=-66j0f)*IXT=rbC0Chmt5J2Js5r`^z|YjgGRdFG}!F3AblutpiU zk9SPTilkms#gLA(*!jPXg1=Q#?E@NRh_pm#VRB{vr=myuU3II5Ps~dq`BU(_UDSA2KeT6x)VO3(YeoM1zRJ$9we>q^s8esal(FI zl7UH8rraSos4pWH%+mz}KQT6KLKYD&;5O0ltY6*owYU-MSU0xzZQ=-=! zo~JXknyP}`3ln6cta&Rd_g0=PU@+Cyf(~<_O5v@(gRLtY3&P2nnYrH)xbiAN2h;ko z6t9&5TzMk+UQScn9u_;kUpHg?C{Vh7$^_8?n}~kW{R)FQjEH+hyUyIiG~-mJcxYPh zes3YT(9m8oht{ry0_`mNN9NBVSI5E6UV1A=WHpDsJ(gNA5cV#5iO( z%^(535)UU(3i}g?iy)n^}>5VggU^ogL1D4kG!KH5;g-+ln50y#Y7b*=3 zN;;2|ZsqF3o-BM=QB9Af=D}`L_}!ZDJ2Y)ufX=f28EwgIYvDo)9Zv8&%=ruV{fj8_ z|9cZ2VbOm&BlwSe#Ns%AjbA_iJ)~{d(~q9)A`8UjBM!Z*YyAmXC{K5tNRMIqE-Bl0 zlamKE_%;T@f?w8?2Kb)@V~^h}sPK5&J9OesQl~u2lxV}TO6yzOn}-@Iz6DBbG(HVn z8tp0{%lEQ{u4~!Bjkg01F3K_R(_2?_Zq12YS9~+A`%WqM@zDd#JKA}IJ&&8~=^}(4 zRed3Q;`^PrO$EOD(B@PIDtaUt7M zc8Ht7YeiqJ;Nyh5)z(G5#-^mYp7Dlcln?WB{4bJpYIE6r#jS6@6&fja=AKhzkyl{7 zHfu$bhIOqRk=WmeRKc2p$BWxTzBMrx8jjcZgbglN(GNZYKd;G2uv)b(`J8<*e6_>| z(Eolz`<{rQIpu{$yL2DVfaT-3H%+;i#sXQw@x_{H!*R~EX9&j~(L3Lu#pPQW| zX}bU=(-c-f2@CVwHYSRqW+1j2jdh~F1&`4tU~)YIUC%04s?G~NTd-m%{1)`FN!DR) z29wPevU)4vh4Lx;1L;omqNT2Jup_NC6ZjY`HXtIO9z^oRB=CUfWMbh&H)H7Om#2ff z#|Pd(QO6LW*+f@J#ne>w*-Nzj?Pz!v8)#rX2m6%z<<(#aecqZc-(95M)r1?}+|MmnQ^LXQ<+J;Vr2`kiS`W3Lt z#OkoI{Oel-A$*)Yn-x{OHdPm1Vw6T@RjCmQ2Vdu6vFRL%dI|Wi<@=81*x%s%y=Aym z-@e|pkGX#3(uyf|NvSM)tP!Wt$d6f;WSvd?K?1=gc8cDllRWVgM0*{?&%9VIf_)B; zNVt9P*_vKy5`L6CiHh0ia(4bMo_zA5LV#NBU6x>E$s{vyd#@8AQzG%!z*W83OH(#X zd5~zL$?ay*S)T#@Iy ze>c89^d{(wwe`-``Ct~QOlGbqY&{R(*|03p=sjcH(LD?M=cmgVku{p4 z4G5l0^0II`GfUVT-x)faLRa*qu?sg!>W77~s{HJAoiWM=*T9SO+a7@_zSC*EuIetz zCen8jmY(%I-92hTibCEG_R6V0c6+n3gxzIz`yrFi+clg6*UrI4ohwGtD*Kx?82cpz zMNd2_W6P>dT6&z%O@rOwf3R(&bgRdL->L!Sm3C9N0nT*6U3{>{8}T0JaqN#Twy(u37TDOieV)#SOZ3;)WkFbR^Jke9)9 ztB)zm`@z?A5~?rhN|0ZES9ZzGnoaPSk^L@H+DeIr$`vp#;+ZDj+shIkMqWl8%vy0q zo>4HLtGp`h*UV?ojtpVrjWgnq=Te!8x!dRS{=pMl;0}#9utyT#;5noH?g)% zIzHvRKzCAqYG#>qXN)4p*kV@)I0or5P_HKV=fJQ5qr&j z)rLG7c=RTXjk%ele&>;*%~AJhOV!{?gg1JufG-mbIabYYovuq{`7|C|>EN zqR+sgqpSD_+xqGWM|~9Gk^fNjkyo){{>1P(gxK17G^r6xyq^1_BZh6Uc?53M*4S~ZrfO!7{Pa}#-if(jlk!$9^)-=zCrkD;T`=-`-$jFNfTb1HR_LSRkpiXKf9am z1)&=I#z$?HEU%bT7vXhLxPOT=9Cnj=CG^`@SbRqbz0i{n5Azx&p88rdLVK{tuy0_o*+#T$p#a^|E z(A35EBKsleKe*tbztCLwEKCL6RyX>F2`?^j+MU5Hs6Z^j$L=KKkdH^Tj)rflVqdaN zQjlj0rn(n-hZeLb7E&F%ZWwBrTD)4FnY_9BZg03>`gGs<#rRv@QNYHQz7zKn+h@E{~OgKc%trBv+ER@6VZVezw^{;2xZNH zn<~AR{5xXQ^|OnM;ky$z@ygl66CK%`R^qNm0=}#w*}5y9#k2l3r8Mj7_vAMk;Cd1? zTY5F9FJ}pn|CKwd=Xd)+arMTi?1eeBG${QrMd98%)TT$u%g+pUx04+n&vWjUegod2js}1>xyQV;{AB3rO zeKbK1K!Fr11SbWmeG+e1GDb?Io(DbEolL%i|i!;x*B9}Q)F z4h-<=VYl`Q=*+g1i@U@w%98U`XoegnMJOT>>2tB%V^RJ}*VNGPVT~Gy+@ySjuQ6MH z^NY@?XekTn%*G)fvi_O5WT~A+IVq~XJR*Nuz!zqNYT3D zI-I-3jIQ*fOQ=NnRvhx&C<{-Y4lzv;p03?IPu5e=Gs)4gxzQ%vE3r|HHE&RB)j!*8 zU#U_jYQ#)_q|zrs=w@~lU+9~Caxf|#Ds)#Qbas)Us}s7pWFY&$GfM{aO= z%@qovET7Oup$tO5zq8j#PkauWFu2$qWI04Y3uV8pXh0kAZo~rm-qT0lBhMJk znaOc~Le4)4tkbIe_y6)G7Y$d45Wc} z&yU2SiYN}}Mv{OCsO{eV=F*%156;V5r!NE;int4U2_c5d?eEt+5CY&{6Ub8mHaJcZ z3!aR|jcG?0n_ljP)*=p=FzVu`3)6b&=yZiDoYPMkj<-h%30Yj)zY&OYM?$WBGcbM= z1>MN1Zn(w>fY*!{Ek%>fE_vBF202{9{r?@g@6H7A>{k8h^}_@w05u{M2ll@Cr$3fB z!3AW7ZL{y~uLw(~uN&4QegxG2keC!}+TC9jUMs>ZiA|l0JV*CuW(!t2R`hDwLi7XGa_YryLW($V^t8Tm1!ln*=Ys#@7G- zFKS1+APe_Oq0GEYM~9F`w?JYzJyZXu+DKf})go79 zL8}uB{Sd;zAlpKsJLEKfJX_7jT%84ND{46>g}vY3SY5)}Q#0Zcp+AWk>+x2_4VIEa zaOF@z<62}pYHnky8dtvvoFhQE>Gnn>W<_IQ4C(YSXI1(glr<$BdYB#e(qNMQ!&tZ8 z&SC>QhP?}x*U13bP(Z&{q3=8&ER}bE3tid6qJ0P8ttp3U6=MfrNOc@|&3C}!p;%K& z%AP5BEP`Us&L3Xh?W^??sEO7px=6vtaXz8nx!94E6xTsKi>L(+xlQ14EJS#QN#?mh z(@Q!`2(jBLOCZA;@W>}S)ehJC5BsDr%@bVL zX)d(RH%*4)M%a?JR`#duz`ACxq<*U0@Kwg2wM0_%5<#aff@)%ArJ|wzM|}FT4NksH zu;xyNzm5j3Gz=O1%Z1;;zi}-@sr$B1e=#2jBs~U-U=;k`{<|&f-+Au+cVQt=CpQ&3 z<|yGka8tUoC=T?q(~=_v9Yyv08Dqi5M=1v*ST`(OFe&JqIz72pDU?-$?q_dV zfh5|k?wO;nV`v@*M)nER$bhDBE|uUb$uyV|%-$TgGl z@&Do~7V^Kqm{uSIh8L86kdm%`$TQRBy|stp%@p#!5xTO=FJL1w;?2`NOz=lMdVAem^+*41w4BCbRtrzqZlR{*kjF5*I z;hsQQ>i|HvRJqKEi}|24UD6)wJ-&pgKNqM&EJu*FZB@gmHfhNAj6NW zp?ca8`VY5F+!wX>i_3@)Z`po__kUsTe^LAX{Rl!l5m`4B{^7aF_xxq&5W$sz^+|h{ z%v8*+Yu$a6j!8;yk0b74&c5FL+99$(^SEaG?m@ebJVIzZ$w7-swX%gAXMG~iHrdKW z{~-R$yQfd_mJGI+w6t`L=@`f7$RhT>+PU%Jr@1>}2pe|{k#4H9v}Gxw_>%KPHw3+# zeoNGEtxx_wQEKni?Autbzu7q_iT3^0`-6t(U@$u5XjslEm#9`bv6n3v6dujkoz0RL$& z;Naxyn*FNxaYhCJQshamTQzB5Xf?nqe=wdU3uk}c@;k=ignqojPOuFfxIQ3ys6Cq4zszzTFZx-(F!A z!}l$HIdPmucRMzZ_5b_~zV-*Pt|0xryrAL9%9g6C7!8c&wV-y`O zUQ#ibY@eQ)>@XDcuZQHc>Rc2s4oI~Xx~hDKncWL^%Ip&XDmg{JEoY`L5P{_e4fQdL zP1C~-_Wj=#G}6kh6LkbGqIcU)3VZL3V|E+O)tLQR*M0jEC3fpK9;m>4RKNZ|+`V;F z9R0Q?oCJ3Z8VG^l*0==;t_e<~jRjAz;2I#f6I>G<8fdK1ZrmkUaF@p2o#8!a?wvDt z-dXpo`R7~f>wl`M%YIe6YS)%$KO5Fly&UMg84$CiRhP05RmQq(+i5&2$L#wAEyx_3u(+rNOc&)PsSrm;Y zTU_rdHB0<#KO3@kUnECzqI43zY9elNWIolMybPUxxp9A;=*hkWJJHLWdh3Y<0-u!R zAU4hoKrvsMJjz;kF(7g7qD@moi=Hdz^B|*SlE11FGSNI_cA!VuaqeY{yN`8bFxoTE zpET>lP+$#S`qUJNOZ+p)t!>qJ8fF2@@ta<%n)3Kk?6~Fmm$f0HmGWRs2Mjt)^(cw(nDsOZ}wHNQl;qi6Buz?%ZSShXLkGQOUv}bpAS8U zi*&a$uG(LUrIRHXGc#U-a-Z?d?pq2JLs0)r6*{p{LOuA5@1Nw2#d=Qc0bKR#>rfEDT}i!!6A zZZ7TMB;4tqVjri2{EH7R73_YKQaO+HDWLMs z?uTSB zv}_~0$<45{kew&zLCLS^2}SS(k%*Xd^}NY^Jly?iCZ?j9Aw3MHxi`6JGk+AFq-yZKjcb~dDi_bhLi`wRRVndVl>TpQ4Q+E5Sz~LVk zzuoODfT2L&h%LOmo+rMLa!>-Xm>>o0l0r&uI*vzG{kG2k^GN^O45$?7qmU+iC2%8P z@DH6t|MbC8*S8oR&-Rv&hwo*S>KB@6@w*FshD2AQ&4dl__v4Dp&WjE^ii^cfrf)m8J#OIE=@aF?M^4i$w;kz~k9Ya?Tppu{zpi2|&*0(e zAxl4|4pLW+TTA-Dyh3jzewr*^qc^ubvFlf>mrlB3baW%sM1Nm3aWf}g8*vK|_}yMd zNwSoFuX*+Ru=wd4FaYk@8<=SGjMW`xkDmH{0!WuIZ>oQ+n_BT)i&^!m&c509@oYT6 zFc|VC%eq$;(S4pzuEJ0K+TM~5;Z4S`A~5d2T&rZR8#U#~4_&ypxis0(ny)il^lLb7 znF^Vj>teg1s?r(^(Z3FGE^L-GNj&F+KN`KVM7ZiElG<8kwOL;FM<5PEct8m5B#05> ze%pFWx_|$OzIqiJe|=Pw2=(UyiEn;gn2=bcCb`_SK)6O1i@zlJ=g*`34(07ng};SV z{KKO8|C?^YJd-g_yZ|0>7n}c0iZ4e+mJ!Iw8_a2_m|Xml7?j9mMc? zq^YDSn@`p%)H9zSZ+<%jSJ8R7eZ`ZZcBNNecM$XAk!qPSfO}`}Bogy$ZQE2EA@q9L z49&So9OKLQnK+1iJF`CQx3OCB8pR+~>0)IL;*-aMw&{zM$X;=Xa#^K@Xr(V`h?)Fz6+9q+Fv0Z25~%RpGmg2Sk5^Lnno2 zDcy&AV*FMnaEnDNKHaySmqtmc)tdAvl=+?m@19!uB!v-3QRyFUct&OGE@+J(+b+Rm z!`fzT)d|81VH?`&ot6IfKU)$%z$ewIBIqi?Lj+0pVZHq8Z@C*SKhn>s(Ds};P4NpH z6qNQaM#%mc%F3#fSxP=<$EjI~VHbU3ZGiT_=cRa!S!cGo))4=PZItSR$J9{qyMi@E zhRQGfH+(ywHdLa=Kg3uTOTMt*Kdepz5-L=`^~=QDs=r@2Ym+SfVYt%xu+lVxV#flpP|?Z#2hzkVXhxoeEfSw6eQ{-@91p%~ z6#W@kg)tzUfsi;E6Y{qa+oc()vpqL88K5_#*`gzEbI! zJdITe2=#m^`qL6)$4}6|kM%`^Zh(%IX(+vX0!BKW3L`oJxM5$pAgB*#<^TyZo!mkE z6RT3Lm+23=EXJ#O5-Y<{+i!zZKC?OF+FC0}=-uo5RAr3AXRjn24VQ3h1}l`>JvEJ@ zcaRqRtH8KEw8O1@>AyC33WQM4%!avL= zEqW{`B7uJmN54+cwzHA}qBi=@y70zwi3Zki&z|7|-sh@S_MSxY5}e2%yRTlSU-nn+rrpeoG0(|`1bmUqrRqcN zWI;Ex@)M9j1PD-v32v=AS}FG=SS0rnY|^Uy@lfRX9&gS22l3kRm*|)~P;xI>0Eq$F z=)L)OL4{`&v!c&4ze}K5sQ7-79DVjuKc@VWYUmc3579b`POj_ZuVvM%Tnu11z|*9r z4$uc6PaX~O)FRrNQpwG*7fns@p<}V(&wH7QX2;aBzR8^AYwAdkSQ$(=0VI$%G}*YX zNe!?Dm{mFL!)j?0F*q${Ov!u1dsq;n8VGyt`jWZ6_4oeD0l79rPR+r;wmvWt%;TTE4 zcmOK-FQU=yob^>pnu zRCbAYd58(fk+2u*mb^L#9e?6KS50x#Un)H1%9TtZsjl6uEw!y^*6F5OA4gV9?yS=A zJwm4nH*jyTjewZ*IB>W2f3m_dvhHTA@ppNKHuv0flnSSgU6ar3g@cF?5N$9LvB>RI7%7_hxq_3w;pt_2Q(vD6i2+&4Ocpbt#-&DLWZWFBh7z{0UmLxu0dH z8D{rXo$TxL%OCam0=t{4rP-vI6S<>;)6v zT)0u$GS^Hn9t?%urq+aUg8RmRq~1hCDXCU!l7Q0Fu)`|~mIMV2H?AM+d4Wye<;8=O z;9Gc6m(zv9)!$g|!B~Qvs=e}R zaSY9yiV}*<$c8TDrOhSV$semj8>Ktn;}1LnrTycwV~V{^d0>VL8dWol*r@45*sECX zWSy-@I83kVZ<54Krvchnw69?vrynN*ED&$UJRsk_aC8Th8Og zSwO<(E4U|`sG2y4GoNT^4M8XvE}onrEN)BL=r}9RywB)ruyh`K`7>(ars+d&J#UZV zkU&50k*i5t3Vxh58q4p~*m|EtISaO{!CNVR0R!Hjjjz925TG0KE91XZhjWf74(0g_ za&Mz2#iBo_wMZ{0>8mfSGT__r?(otL16bWGA-yui)OE+On7%5fmB~B^aKY!rQ1GF# zNy9Y%^8&}h$(%M$m`%rY4sm;ey-G$Ie=yKMtDl!TgNlUnb0|5MSO3LPp)4%bJht$; zWcbVVL}B7o<&OeLDyX@Mk`jceUpyH|XdF9ZDJ}IkJU;4jJj;}$+SanpwXRcsLlnrW zQ5XRB(Z=F8gX;9}D{F7%xqEkk;Z$`dav zY0)j7A9RssN3oc^@SNs*@$DqbCG^xzIxy^QKeC2sa@XxIMC>@X@G1(=9c$fUi-`sB zto&@`8Ir)g!uybSE(UxMB4bnl*PRS{=WuXILBwv6(nU>0T`hr2!{G4x=l2hujx1d( z+tJpvzXD|WHwdqg8mk~XN)@dFnt!6Z1$1)6RcZFEY=HB%GHtutB>l2pDXX^!BaLSM zi{r>O*{}zBIEFX4AMo%Zyl!Ou%q_<~fR<}XJ}Aq$I~emBa!7yrM_+}=e#XSddUzF;JAWzPI3&4Gh5UjE`2Mq?MM%&6)4 z2>&bO8o_wT_Gh~dXDRGB0<}#XEmJEDOaZle!ku!=+@4bkm zjSNKhb%>39(?$?(MfiDVxzJ%sTdRrI`ico?tC)KPDND%ldkYk|_r2So>L5Kb?#A!< zBR82YMH;q7mixW%;a5(A4{#CJXNEsGsHaMtOu4h;Qzd@TWhs0WgX-IknsS4RmRJYNTQb~hOFQC)ag7A4NFCv(>KKCjVFDy1h3moD zn?Pc<3lF?YkR9V#os?F})ZB!gEfNzlQ3pa7vxEUP@hZXob|(*4HK?5Vq@2_K!`y^N z(0jNJvoNkeFdNQ*EWwV(wUZEBS?Iv6x}gD4fU_5-m}%!9Wk7m z+#JX9&s0koX;|lRWqNI%7X0}Wajb_P+GZFT&ZL8ywd7DSwcfJ-`_JFy6lqSKZQ9y} z?fchM5pbQx_ng?OooSp$?VWFKE-|C)0^_XG1uVGach?s}e3|+$qO_)w;cDF@-!^JF z7Ah+2%k1{P&Q12fsz+kO9P0CU;-!rPueyW-pP7$5_PpVEb|@X?q#L6Upp(PKAwljV z8-qV@d@FHJfUkFnZ;}p2h@503(wGN{ZT=a~_--k-S-e8I^)e^U+z7MBiBd(kd)Z2j z*X-I38m&O-TgR{R+2;jQDUI`|ZxVq%p1rKwfr^fhv322QuEPV}Fbh1l+$sB^wU20b zC^hHXI#75FY3cj)Z<3(BKYfpB&O{O;9S_&-;(F^o(XNIdESIbG@n?M zQaNrMr}nMJP^IHkm{*yzt-ErKRaCmxEg@&`-C}9xKE{0ymlWl7oAR65KAv@QfH!10 z6ww^m9Z&)ffaCfA$GXeJ>Lpe~TXvmOQcDz&b6>=d}AR8J;P+sEkjZ0LXhDN!aZog~GEc#o-hTjpb-DXu63(BXx*p@}Een zNagS*o04_^guQVqD7R`4<|Sa~;VHSB*kEh6$HJ4{}@0)kHaM1ULn`wsONd%e3cy+`zuM{us zFaq^r)jRxG-*QR|1D+MxTW&Vgd{LB0p|&)V zr+~;267y=Z^xLVfjk}APEo_sSxt8+&`(G_(sE+rs8xs;s})+dv8=F5b3{^0>%! zyJ$oSo!yg*P7!BUi--pKV`>}jpXoeeGh<(NnQl<)rLBF_>(c&0o#uq4AKUY>0Jk@v z(#B*U!UiH5Oh7cRfV#HsC2Lnxo>-+D_w4}Gdm|Cyk7QpwRxkz z=3Cra{+xr0V*h(315yHM9ITxIw*W06p4kpBTalRl?i%<)UoinN@!l6dsseS_rJ)?V zN;X&8-rM%IEkhv>KQ4oCev0E8w-VduR<)Bb!&CjbK;3{s32O{lydd*Rt8{ICz#>c@ z=TS8tVa~`lbfIk{K#~*OiZb29 z?6E$kTZ@^^YDaDATzMRV!n~WphPMfCD2AtROJOSW3GQ=~?O(U(5JQK_b~pvLQN&y`YSb>C-d0l;dXsnd?p) z0pfQbhT8Zy5Srm7?qQ6_fTKJ&&yh*phl3hqhJ=>`Ql;rLk?OAp%J&M{$GZsf!aa3w z6phB>9_rT-c;?{#1ap+QP@48>`%pf=@=vxwiyw#+-NSQ046b@k;mpb{te;K`+};i} zS!bH?pQGE(dMPf{E3K$-)wNq&xRNSOVqorATIfW;u6@>yVZG`+`pkB+hpHZGTHll6 zxR6=mE-$2QTS~egDU-ey&O63BW(5pXQX;o1BDlY@^DvgksOkA5nWK^sF|pDdebSbW zUhw07zgqTy?zRNZL3EN2pTw%RUi`qVdu?CZHP@>8(Gx zL(ZhtmC1<}oRozU8A1V>fmA4hu!J8%toB?bt!3fu@p-8Qq_Nm^W_q#6bSDb(!!?n- z$OVPOLn^IVlh&Lvt&Mcg^?nq*IIOAUO{-FVCzC{6>dwTftn5|DiL5|_gQX-a^zng$ zfKmaIpdIM`>W#rbBBZ*)DajXRk`Xt3M&h+d-d9!JAa^p)jx z-smydP)uT4^gW})=o_!-=3yP5FEi8DJsCu`fGQYC%hQ-*j|@@ZeB&~+6(L7GEzLK^ z3b7RPU&SV4*wTd5$EGQQ7k>*ov*y?Tzwq>#M^bT+`3tGPbth#4pvk=r2*`EN4~hH@shMAHiR}gZ{jJFx1YseKJjL#SCjl0-vF|Zo%cJg_1`{2BX^j_jH(cNkB6=}@kfnmm@$H~E- z;bD)-q3&b$ZaRE4+zr3JMBFDrh$wCI zw*=F3%+_8~*%oY7y{_#R*)NNdX0b^Mvr0eUPYtU8)Q#8Y!9A&rf07l%tJ$FO#|F_= zMPDZdj@UYiSZ96b2>jAQbf0r6#49fu`pL$9=R2&VsC2bPU*<sfjd@(exPKB`M<(vYir|bsoYc@DoK$-WGkjulTU@@7A1U984^=-x$T)u zS|f3`;47Cw`V%uZE<=)<{lxsQ&2?P3?7@i3*N_Vht^Si|o7&~-muy=o^tNfuF7mS} zJ{BKWxLvT4D|xQc8iT;4@?OEuT6e4MkF||)0pnr$Gi-nl8G-3WNOh{Db*f@7!VJ`@ zB$1TZAA=p0LRws_>vhE=(xF3+)#VI!#4=}B(M*iVtQ0y4wZS=KarzAobLZkod5Yo1 z$W3YX{mcb_Z25mveH`u{>)WWz;LYSS)z#Jv_>DcuYT*3IUcjJ!J_C7WQXaqDE`W(u z@$=6uymIJ>g5ShvL7y8T^M$5~5$3vOUqbj9+EAJCludu41iVnZ47~wArIw58GQFrC z&k5S>9oc=ivlCvp0sC=!P?2p2jX2o;K^+5%_VR*shArNIetm&zbebKL=OZ{?k=^V* zexKquk{OuHC^lbHC+@P?4d?rax8%H(ztKG6nvNW)RDv0-9E*I zVA4_S_xD>T)f%j;@!i^CH?$=uz?HRnfdm!NB(y~qKsYPJJB;2t*+=&R+cKG6hBZIOd zXDFq{WCztRx5ClUl2bu_z2_t<{K&`mlxKNEQEi}xIBKJ&4-ZCnxBqu4)#N`VYpaL73hvSD2^$aQY!Nbr7nBMZWgYBM{~soER3~1 zkt8&9g3@+Py9Xo0rKR^IuN}xLrbHBZIffxAs42rk!yV;E7rJiq+Rl_h~wIA zGlbYD|CvM=!^wBl`6!kRr{RE3+QbOyF1A3WO0jQQ0(=z;#4s1U(1`I%2jq{uywM?~ z0iV!0c>^=!-U#;VbT9U}iEJ!>(Fr9FLE8uwG0CtrQ>$a&IvgnCoeF0+Wa)$d6GB~3kj&jnmmke+?nJataOZ7sxyvTH1bitO|ADgo(3FXc zwBwB;3Xokcu*ql5?gpNmab9cjlZ;!+ZBxMeAYDT#K|foy;v4eWaBgjW+vg)Xj}iQ2 z@l>zEf;K$EuVb$Ry5ti_%mpL}gt) zA*S#122}z-0Kt`+wy;Y9SyoFfV`h?|9}zCmxycQg*K`=eSUiCWFh-#jb14dleGJd0 zFsuiMSr@R*{qac?dj5IicK~u+m4QOS(GXGHpK)Q09|Zf0TdQ1Up}O%{rlE0d-W6rC z4Ak!ony-KJeE(Lj#=U|+pxU`CX33mk{^HL8786!C2|+{NkZQ#zNxGdwzOEruk~a+#ambC1Ad%k4LdhE-;&F0*oYIp#QX@< zTUVzEC)kiBn>{_BO8hkMs;ciz6JTmKP;dY}ziPDJnf=VhU9{X><5i;VGXWkVcR9U_ zXCsJ78Kh$v7{)26-Yi8=(4_qA3>ET6N$d`wz+*`yliJd)m8X%ZvMO!;y><}9OvHvD z0vEWDvA>U17e(#Jad_hMXWNPTr3e3#j-*BROoSA9^_MX8O9`d~;gX7+AGgBMSY-3k zJFHMQ;9}-d26>@t$zu9DS$cV#(1d<^By71lnDUa8$l2sYRAh_*@+&oScaPx8Z(!A- z^H&+gg49bRe$UkCOc?P6nlX>J9XE+$*r#gFQN)m2uH%v#FTT7I*YWuTNol-jM{QOrJv#W z+45!D#n1hoE9ejK1I2hwk8j-r1@&%S3+%?IDN}jj^%=dNs>-TZT>-aS!HK`)J|=VA zZ;wkaVQH`uBj0SgeKEuF*PnhWV$kn`uRRuW_!f$*UcR)_%m)o zCEM$W@lp+Q{Br<3KE?hdUYK2Nm~l9Z=omN@O-I{=tJIx9+wV$`SJ1@uJaK=UpSXsp zFaX&BK;+3dRUca_^6-*$gri|QkJHlU;|F1;(ktR}8;{lkw8aXsju!)5Y`AtCmq@`x zJ9Mq}l{{;+!@8)tZ;X=?cvBm8rOJ957Mv_zJCctLDzjsmHbNf$Tr~9Jy&^}fQV&P- z2VjudI>3Grd25jRY6nFVYM&F5kZe+A;NAGW309YGPf%odl!)X*ur+<_p=&Kh!z=84 zWdC*7!psAQ#yVV@l;2!FILC5ZRyzB2OS8o=ArXofig-bNFBq8#GaCL+50&eQGW%b! z#s5)<%k}Sb9e#MC1bhNz!%tZ^7%2N)udLeAWv8}=zNa6zHfpBVD18}S0_fYQZyYTi zEsCIi7bXp;oYzCH>1C(5BI^^70Bz<*nDqeF6e{^2Qvi6wxmGJQ@lyu%C_sHDXhgHy zxfY@rH8MFknJ0hWATIO{!cr>4$cd6CtPsl3$6 zkHbxKfTak#$%DC`M)r++-(tRdlTdWm$$H33fWX``_jgWW&x^jV7YCq8Di^a873qN} zF7r`Ykj%+eFZBSx1K>sAAzxY!ArHL&c&l8I2-{eZ!Ux@O1*+q_a zByauEw2Zx|Mm14T23-McWvhO~!k5p;RQsz{OF7~|648!AOC~RGR^QU_?lx&* zZf0R-#shwr6{Dz`mUTLGI=BbhH~#ki4L&;Z3;%zewD&sD1=ZMpo<$2Fg)O51+}p4~ zM}N)I3waK0I9JaECKH3Qv>})EQ$J`?0izN%-hij+INZBzWPm6zc?LuST&JhR1_j)f zn?PQ}urI>vO#!1Ar~u~e#G{?I0Ubz+?VOTBz+cBr%lMuw1Ns4K9UuW1VO^W|T{&>e zb423pyFY|0(qy%KKC)3hNbkH0W(VN_E=(RqNC53Y*B<(?7wl-eM%wOtrW~aJ{pG|F z%&rQ*4g&}cdoeXSh*pUP7I~Rt(>9iAc<~qAO#$Y=7CEx<=*ypM9+rYsj2Qlgg6Ga_C zx1pL|6Jz1DpcHoUT21@e?1jzBwNiaD6MN5gP5W=w z6C0Oh`e~(%Bp{v-#J3&mT(uM^^@fz(z9yj3WNV_`U8*!$zRXzUaZ>-j7yPsgDoWNV zYqc$25W2qu|&hAM29ZZ4=U9sZSu*~7Zh+VvCb<6OlFB{ z0N4rGN()0T2#qi=uZ}keCt2y;!zM8+E(CuQULb{D5|$F12}SRkwnSM0D)Tx0W$G)k zKh!*({3XfpNl)lBfaSN+&q+@0SP<@oS;Q$J5VKl`D>?8KHvlGWJz9}tqvQ920>oWY z^*A4t{2m~Y&xvQ7f(1HxPe}IlwF9}sKl`5(Xj#e*9%Cy20 z%Yu`<7R10q;Bpcy5_)*)g(r9J0N8^0Q%8>D%-(s6`2Eyn+Og!9b9%k`$2I&2L|tPb zLHMC~Y>mS%;Fu)&x>SV29UAvcW3JB(Bs^hb1p!%&MKex{qTBoU$Ep~|8GHeUtD+LLe5;ce}H(2O^F?f&u4|J zr!XOZ&c&r|<9)=x_>>0N6wTYtV?x%?#hXcXaX--L$ZQAQv0Q}L=)jOoh@E~tK9MNS zRd8T~D=198b@cFbe;%=8txK=v~WbD}F*n_oAhkgPd#5Qcyunm=iX)8R9 z@=Y7zl$83I5~?;va^op()sj84$pfd0=qC*fZAU&$kE~vPK;gC#9=QSI_%`o^F>yii z*s14*2UIU?*Dp=6BD^6XD+F5kny_-~CcEH8X}e9)6|BJ=wBnvS;rXtyw3v|^%mmkL zS!&~k?72VCZ|ks4sLuE>YeI%PNcY@9S!rpW){5NPCVZL?z9CJQ8Y^C9G5g$qSox^= zh^5Elej}=U-Vk59O54341TJQ&IEn(7F@mkVs-kl916;P zxRc4GFYQAC=nnnhDclu)((SBnsEg1}Te&je1|<2pOE+|MTjsrG$ z!U7T`X+;6}4{6j69zN~U19sL3#sYkW_1BG{uZ-^q^DVB3c6&A&!#)Ot7Q&^ z?zSpHfcc_F2UhO7w|82{odJRz?vA`Fh*SFVYBTtqn&L$ltNO#MttlLBaqRWkO?no7 zyrJ?KJdz&bb<%rbmFQf3+J?^qfWu$xS5l3n=ZyyCfv)A%U*WZj>E{pA?x04}ePNqs zv$%tg7ZmQAy9qs?6%6`z7lrUD!`IvI;SD%Ed*`8-y_Kx$1C6(prypzN6(J+Lqq%a0 zLQn3o5`gtE`%)Ny0@}VLx`v<%FN_c6-`@|Mnwbp`Y7aqxYnzb5WJ`<1YuRw%$-U8N zM8=h#Ja2N4FuF4`6c6ny<9`s(ZsbKO<{ow&oI<;XttR$D&(wU7w~pgb(Gu zBAknE79l_n33@&;*u_Ds6<{S0r|j_bd>A-cKeDooIJM~xfr9wuogXHy=V~`bQjN?`NmoMp)HIF%I5Xxl=C=N`H)W!Kk#x~^4ynqXUu+3Da}ZtxTvVs08M{VV$>H~}&SA)A1dxx67dy_Mk1)Y#k6Ar2(gD%TSSr z7J_9w4NJ1)bPD&FFqDtfOo-wl;s71OvaMB=Sfv=XMZDir>j1B6Bz;GUMdBk5Ifiv? zc}z^@=%X0(XEAcmM6kn8DmO4I0>lQJPd!n+xEv8Si8@`Tf_SWh5w9ZpO1cA{!`#Or z%K3mOSfKW_!tW8`S`&(pDfD^2UOgOZJxXnsTZk-6``zqK#}L6Wi#!06Bqnk7m}X)Z zQk|=gXJ}C-B*sRR)iT*5m;czW01;u!QZa443$8yK+mqR(7%MT!_kD=nAkN?;$AEB` zHhAh-3=Z4heUk1zojFsqnW$APAf0U4@xuA4R&6E_T+|nW9#3IkT(X9?*p3!p_62!zqa110 zi<4VO2*~?He}^S}Pcr4A@rnGqs|NIL zflbn^l+-tULFX@LW4?~DUeWR)(ec9jp#jKPHZ&L2YCR!kzB|_gx3>k993*V<#&R9y zSeHVoerY_QzHCdNwaxXYC`B|c??Dg$6qq!ytlL0zWxJ~!;oM|6zo|NJn3n{I$Ds?2 zpc>d5^UDeyluPv#*9TO4j}qzO%n?Hy+d|acLBk(p1jX$S?%}BjthjA01=bCxCwnK~ z-J_4se(xFqJIE0O(wA0{FwAO=l|0iI^!(Y6IJN zI@%IDwMC4ADS+ul1I%Lqq}9o^Ws)58b%Q%vH1Hi`;z`N<%hcN*s?Ln6goAuqXn4b) zasVWsR^$=78&IT?6b~mG2N!6YuxrBn-tI!r+a_;Gr88OL1Bl&7KO68>rqtIQ#GJsR zuUb;z4~Tx1vUn31!Ql^X;TFP$u5|nUDV7cHwPwXtbDPJ+EsmrlMT$Z(?HP?^@=L9o zppn62ld3y5;z!{4wJjVzdHo3Ncz@hx$ zp@%RycI{;a=HQebQ7A&To;E)c3(3lm)B$0=g-o&n<)vSB?w>aDpEmjR3TG28ZC1ec zmNsLpvuc^xNjOgYCCjd!1-oprX2VXSE<@IDDPM5Bm`o^`FqgkdHp0+g6Gf5b*U`YNIrlssF1{rJ>$)>qUF{zlS0IFnvv7N*Ef32 zqy!Ojaz!Lq$q)q2H*V16gjCo)uyEq?FHe!Jp87o#iPRJyfmJ^0do@3sB_CYEr^h|k z7BRs2aD7UeJwXP@5_Z|?z!N{GZRW(iw5Nk<`|_-xM*yH+nKuccGNjz?tbB`t4e=!* zZqYt>!L5oqvo2jh4pw3|x9uA4`8}@~i9l`#l(#o)NycRyO(ctM8e)rE*&^ib3&%Cl zKBCUlamwY`Nzvyo2X<&0226rNul3x558uWEu^6rNC?;9iPcaYPo^(c2y{nCHh>@oN zW+mO+KTGl;UXr6LCfP8>0sLf{wl_-$ZivyU;Ux6%sTCoL7Kbg8fiX_hYAP1&Y!gl| z4@seWcfe|ar?dT}wh|^q(mvA2vlFj12RVTX-#}wl1JjLNeJAkbJkH;PBzkWI*!X1_2la&^JWn7}ge$JiifU^NW++z3+2YtP&*& z>=SV2Wz%cJP9n1H-?SPUBM~~BTVmz za6!cB$&0S_@GdQ1YaMt2c>`@!Z&;=;@t|dIFgPL{7mqMLJRc&8;##+D1N)wK zzLg|4I4YSDIX;4ezhi-1(io2`@s|UW;bGYE;9dj3dvni&%dMH}fv}5{a_cnUt8nSoR)W0q%I=GpW z3CXizMg#z9fID~G=H^O5q%bF0iVo1G)oyG*bXh9{6gSoI6`-LV#plol=7hh1d5G+s^C{DXuDTV~r4=)m!-+CI4u4vxE=FCJl3IG=nWpHFl zMvZGYQlN$f`aEpZcQ)3K0_eVu%6bmP2PJOceOL|vCSH5U)setX+L(5YZW@-gMnZ5F z{HFQ7VS!Lams%bzZ!a&Y0_&H%(u`!S+9IHj+)p<~d95S2z@aSyalJcPmK*#uXX3jy zS*Gsr)VmkVh=x?7_fY!RvvW8NS@b*!@E }x<|8qq6THQzm z2OwlRZeiV~q;dM!?*9>H8~B~l_K%ob&^OZm8Im&^?H9uDZiKJe` z)jj)s-nTqHRlw+5 zc5hiD1O6o*85;J8lo3dQ{kQS%zhR!{<@)z|C)uNim0n3uc5UMI`Hkr3(We@-v2Z+~ z3l-^>Q9pECuhlV9IMxxkW_`53hn7TD=$C9sZx!eM~r2Bk98L+e}UG22xDU zarG@s8w6@Z&PETgW{wGNZ&M8otH56QBUaY^&W)zg9aIcd2kMbqtoZabO42A=txTK@ zsb_AN?YBHn99>(SW)2}=NuvvS3azE^>Cj1@ozoRtYLIHIk)Gzo4e<5{$zFp$MSn6- z1l}Bdy}z@uU2TkQF8*$KQ2(s%(>z+ekAVK>qbBIy$b9%Zb8|yT zS>QdsgAfyDGxA8Dr!HWN+x_`;;|k;7$@)pG|BE-}KXMy_JpYq%=D%kdpP(@OF;{zh z?~6h`ivbX8>qL_nDW_hI%yeW-wO-q;$)w9%QAp^Nr8SQ~nYlhUG$y5EQrnF9;aad( ziSaz*ei5Z&H@Oa|V)T^?!8Ro+SGUV#;_;gPC`2wblsl_J+G2EH?1skION6qVw7v5K z5Cq8k_TOTGL@cv8VLRHMhwy5P80ZHAAGve2VJ6 z{9T>qMN&7MUM;C+Mx|lj)pVS{M=E43{2S(7jSSr8Vgf)vx;0laEqdv#d+pnJUp*dl z+`p{M`g(i5q7z-S_kL1M-BF)oU3NjeD$3LfdYb~N$)l* z7HWl;!_SEf-qjbiC=a`PlFJf~*ryx59}mXuy)i`kAFRCxR8&o~C~8(fz)wU3L<9u| zl#wh*_!Kb^L`ee(YzadSlGDs!KoAg6(g+9$N)7`^4jTahVF&`kkaHfw48t&qaJS!o z&b#NGb>6z?to!!bYfsqS)z#JARn^ti)q1r{o!`sPcO<(v==wQaEIXskDm|H_`gXEO zeC5*K(AjhA69M8DhH9dE2`=v`24mdJ*U_Ijq}_bg$*TNk)i`tCo+lp}F8y>#cdyV& zfw#$NEE=mbzdV2NmJQ$@<=1gC=(MVyv+) zf8Pn(L3^D-O>O+okyEJmOkaWdWXlFE+rzJ_tG;t@9S&jn`-R?3T))AyW6w(;w!a&^ zI*Ql7SNdC8SW!A7Q|ItvqK&28r~7Q_AAj=-Gr~`VIIv1yF_CP%!7(CGurvmrS-bl8kNJ~wU1vD$yFTz$yv-*e&pJM8L#fKTxk9+H~hy<(4jp3nXwZ+-Y(;aOpw>2p)v+2xPz;;vZWdUua% zRp0&@)Lm`ebYxLx?%eCG2b^P`?H>x!AOc zp_R3|SJMd#uS7I*4oh7ZHPLxtpo|dv*cmu$=6F~^D>LdWZ|uXfSEpA7)&^=NfG=GuWWn+J{ zY&=(>uAckT&D^ZYt14)&!2Ra?8;T-s=Uq2SKF6`VnoOAP+aEodz0c%*q$~`1jI}@K zQMr4WWv({&<=T9hNtlgtg^XWx1(5yL$K;^M!B6 zkH681x3`{G<>M5u**swlTu=K+nG-U)?EFCOb!_xE%=FkCkE=J2;v?=^M)0gIFW`9a zZl_m&-dGvp;kK`aFs;baabVS4EG$}V?$jB9Wb=muLJn!M+l9Z4o*7(#_2k{vbjngR z4C^SuZfk68OwGRWXsS4(_9lhxgN$|OmQCutJbBicBu{@u(=}`9*Q*DxN7F_*WmvmE zzGS!<=Q!Bp^gFfw>n8_;Wvbq?rP za9DYX*3}96I8gRLCujM}l>QyFzO%=@+gIVagg)|e3FCBhFJ>US{`Nn{o;EqY^=n5!0h2W=K0ZKw);dgTnq8=`%8UZY6!@jwarg$?l@P3>|Xz&hT77Jt*3#^p{o1=&rd}y8pYGaJ~@XY zhnt1R`s*H6*cG1M@Fw=_i^`H{>L4o=L$EVDt@jhE**S)e2!NVEy4BqVEIdRlOu# zA?4%2&JHhmb{1l|Clt|B547EE6G9H8UL~oGi2TSFhhs4@>lwcNcG5vsO=EP+4)sl zIx`~wa^hIsWBImxo_pTe;*YF8uJ9`#3%M2-qY+d3HtF?)46c&5nNd2o z1@9&OPW>4zPEqJcrk@Z>FgY&~gYV0dN_3z3Tf`c5@^ia~L+mwE*T4JI)*ctW610{Y ztF`<@>KimnfgCpEWKzNoBD1~+hH)2VMN(hPKF{z`kytj3n`|z!I;s+xdgi6=FN2^~ z4Uw@m=Y#py&eDyHJ(9VER9&udly5<;PbK?|zO$kg^Ux4;Xvj5K_Q>6V)=OUSTch|-(Depy|X{j^?l3xV;Jz9OT z*i`hW-QV8#OLB>i_#;XW%_GR4)L0Hh$jU15F_;znhsKRN24A^Du(c;2cu`IsBK-y_GkZ;1x~9%jPiA+vXvSMn=O#2uspBW> ze%k6CoBFyw+C~}_i{rld`m@`(v?XsAr{7{x?hQ-da(?CO?nzI&)^=~GJo{1Kd_J9_ zC1HjM^|{Ey?_Tr*D=_@*X#L^O8KX&Zs{ySD4Hn}H%A(|Na)g4Gy~aH!1B<&Ora2$e z`>VjPFQ?^+UHnrZ>eT4nk!Oo~_k+|ir)xgMK2O(@HF)#R31vR+Q^b5%9@51=Fn3(M zy(;)=?9E3$SL|NiS|1-ISc@oKr}w1B?-e;J3ykUC3s$5bwR_NZ+hvrxIC9#i8NTsg zq3!5lv)hg6;OCE?VkLUuNL}?)>B|kW*V5gsmXE_X>X6UhO<&=YWLciqx0Q39?FFPP z)k_FR(Vkl5m|Of~Rf-zm(Wwg>uo6_Q&AG|1?|ZgrpRzxv1>a(q7ai93SX8bbK3K@O zRHj&8b}mWYJAiTS$+6&Lj`Gjk%~Yy_P8Me<1rK zn6nz3bm4eb)o>Rv#+BC}as&5OdO2P2@@ZEiF#GRnlA83syMnZ$$?IOXdlhOEVaqQI zB`!9v$KT2K7V=I|uYCXFb@kH+u^+nk53`LGsQ;G&;D3BB$xV4#R@eiCgPpzCMOhhX zR@fZ}FVDva4|O+}=WebDSFejVSz+pK&Tbz1&n>MH7oWl&y4oQ85H=TKcM(1g)(E}( zcP`$Pf_x&S|5IdIDQQI+#s9WXBzxLi#RtJ>N>Pt~DwcFB5l$EA>Un(NQVb09@IZ!e zJ*};-x@;w4GDF>$b^s4t{rp|DHWcwJvKn&8asky;m7tO}0)dkqSr~b6e>!>}Gy!n5G?wm`+%H zX%piZMBP94z@pY)e?Hf?OpT5J#H0V*kQmkY^R>L+$>g86f6~Ie!83n8@sF^GH@oD2 zPnrgML;ipA{rQ3W-*5gC5O_N{0#V|uso1V~jc;${cz_i3a+1={gzvu5MYIr_o2%iB z5c+?F%s+K7T-0hu=jcM{mnXx1>%N51MwZG8PW+*O4@7}sPDv4pb|3HmA&LUjO$NJOz% zlBS^>uh{G4lA1de}NbUE8eaX+)@ zH$MMnL!=Nbzzi_u`Ukas-NoAYB`x!OEF23{CsK02vaB|uq5IxdT7WSSQ1SFAn8ApL z+NNc^HY*7!7VrQFnIZ_G9CYNE;VW>XPy4693;mdG?|z~;F}m#(QspzOFjpE7U+2MN zgJCrBN3lt_s73cHDHp-2$ACfR^W>bDjMTd2=yfsV>)RwqE$H9Hn?dprJj zsG-5~-h?^|MOhgj?5$zZ8xu`<$=xmO%2#AT@tPT^?_IBSQ(NBKOjejg??mqFXTr3I z<(1g4ss$OERKcXz?jEGr?<(cn&N=1CGo!CBXS!VMEE8xjuj}G*Z6NN)1Y|9t18isHy{|f`brM zKf!c37qtvcClCIMOi|mJNYv2CHnj;qOYf5<AIZO-6y%Q!$3K+56_5yI*A;HL+;bseLWY^*SEew zRH;jtH|dD`Gx`rwiweMNj)&~{9n{c$VkUMQ(%oQ`tKj1XkLolL+Q@wQH9>S3M4fM(Rn7p(;}}xq1w_n+ z2(kypDTx)8uDa@>rsd7-+d2Urf><6_eZyQo1+)=%ev<>C-Q)7`CPQ7 zEnw?Nl*9M5?mD|uEaeI*hQ_G+vEDeNO+Mt4NbqE9XeC3ucn|*ymXelbI4{8gh%hxF zF>rk=Vx=*`IPF2d5>QjNLtd`tTGV0zbkyY-(jSAym-2rw#3WGhm`U3~INy|>1(9+S z6n-MIkcMabGhAknxw&aE;7U6zY4a790y=-7m*DB*L?j|$-DVJe7|@-wyjJahE81$AJix4sBl*B!f(v=?Hy>7szIMiQ!|p@Ybgg_hS6$9 zdRkxq*hD&*&gTChi^+g}!~tEfMmExo8{%S4nXsqCJ-P@G1+NEl9s$QFlmw;;#7W~9 zwE7D{FChq8SN7q@!pILl?*KViZJKy9{FdB)_nljLX##tl)LXj~6g@SS-Psxk^u}5k zwJ65bk%0+U>kID5-F_Pw)INVQXtHxleP?sm`Ai+_J0Ku8jT4aWGDsvz@Lq_@))Nl{ zvf`)NW}BVB)jJIkoApsN@7U4meTS8A&W~X%{OR6}`LaNg5yXpM!vsMqz#tc9eF8j- zda?%Ip^2~U;Kcz~udAE(Jj=LK$nn0OgmtZch?8q>WnLv{iKXs8*JOE|{;w>(U+ChC zNW_Xnf$MW%Jw+VJ`wIJ%VIn?;u~^Nf&las9<;LK2=3*ku+P)C$+m(s>jiS1#C94wE z)B-txZ4xrx6K(Qj5Hmy9I1C8|-lY7E=i_+Y2Ae9$p#Ao5G)FdSro$VU4uv7SkiSC~ z>%>&|;yKgMzqptg4yv^2ep}9G^Yl{ADXsfB((>(+7-aSFpPg|15?IXup2-8a1u^ES zg6OTwB!aZkwJ}Uq07|TbIYSRjwijsd14OrEDw!*@c%EJD)z^ze#Ig`l14lYKUuiO( zABGgmYlmk6X32ND8yGaoL|U5d-T=%sZcPhD?Ki-{S~h#`*NGDmchoMSC?9f7Ox5bN z>yxN^JL5g6&-MmFv8AYG?%f$==+XITJqLxeVAjr%jv+l_;Gis6bPH4tQ@Fq47 z_0ZxCO;2`Kt7MWcZo2*arzWMvuUr~=ompB(0nxM`wVf;ffVfYN4~8W*!GjC*^9}*` zXc^2IIQ26qY@x-WOUlGka76J-`ym-3fCJ5AH@u7U`~A?ulfZErpPdtY(OHgOdQGOS zky0&&N84MO%H3u?IE}IArj~yZ{m zNJ?k{@Q3w6mLd#YNX&%TpXernb`=ue6of9X;0k586b!j8;BW!FghKc|5k=CyPib(| zGlt-vIWUDgb<%$;kHSOXV?7M_xra=JK_)BVqS92K{sj~9VN8Y+D)A>IcB0+_z!rhu zZ8AuqK|Rk4kPr8SI(!_X&$N~doxWj(Y?;Q<^$cEOFC(jWLVKzH`#bWkK^tma*nXaM zT)1xsg?qsv;G`M&)H4efznoRmIhbPIr$JF5Tzs>I8AcokdnK*~PSgkSZlb*Up(8}Uq4t@a zYAY+--ZOiV@t~Kn-S76aVw8QkHpO0;%zW z&r)DL5tm^~{g@7KePuTtE|ImucAgec&!XQ`+Eaapk4$DNRh zf=-m!%r)qLH2XW>7)W{rYtO$63=hN2!K<6U@vkUZSp&@79(WXquqJ}urH$fICnOy9 zood=KQS#ji@Xg?qHQWGx_8WFpG2sxZn%H6$_X%k}u`bvU z!fx0cm4{_i4=!pH3Xa%wTZ`_8%zF3jO$>lu=i4%l1B-OMYS(6BF&PJgPG&gv*56|M z%&5!m}hD>!LNfgUeAZ_(mC6feBNU#wfI#ZZO&(#r_8l%3z{~Ft(*h`bQ*3V zb#{|P)pR79FiA-BCLVHT4lA%rwrY=}GoQ&sfnDLCIlL6NSM6L}G zD?#=1U&3S^{yf(?Uk(1AlVAttnMudcycUf-qm0@mEs}gB=f>6NdlhwbxN&?M1u-6# zvxy=d{b7_!Q_4k$XAr<7KkqMIfUDspi*b8lKvc=kA5X{fSM~%^znz}t3CvOU-NT7; ze%2*s!nd@V@kb8qzmb5LOd{T#RBB*ovI z2xfUVe!>=~R2~Cf;xp5KMNfo{KQPGedRtRe)%xDo;G`9xcPFuQzeKlCu?Exu^PIw2wf!uRHT`W+we_~d?OxS3<7b+QNI3uB; zq}+n(@UlMMAU&PW0f3X=CE~=waVAYXCBKK{Rg<)5h^0&JFJSd!Fdoo!IQt9`UIcON zKVWB3k#!FWhWP_LfG`1xW#eSUGd?srwiYRu3)4Z+JA$!!$ZE8H%~ecJS6K}HFN!4_ zHD1v-h8^&_gUUbvNmS1-&3BMu!`E2>0saf(;=eHM6dB9O{+p-Ld^Y(i)K=POG4m0; zi8buIS4`V+Yh-&(GIuDpEG^3lNeaJ<>el}Y4Btd7HQ+Cj4od%#!KPs*D}@j6@z&DFoYTy_AT{ZJx!nZ^_{OBPkS%^9|tsJZ58s z(sJPxj8U3fq#ZfXk3uwIHh79YbE>7qynl-mSEz*4gI6#vz3ylVE0+a7Js{9`7fVKw zYS?8K_XNH}(ic91E! zW)yF4&t)fmIwPK+p0`EE7kdmr9f^Ucq(FtYgS5F~?=*ZnW5=D&X=wE|e%i_OSoOMA zN{Oo5-G(?!&i=rDA<7k)4T*$G{S{4elp%zO(EE^3%ngn&tZ$*E)`ll)ZLI6#|WW5Yz=hs(1*zra)s2)SQ(rPUx75++v5zrG-U2l zWqgFzN)k+NpKrV$E}Ej#A<=$=6;MsIl<=pckzPZ#3G-~5^Rf55W8jl8l9$&Wd;BJ( z-|~fgSm~HQ!##~FDaL z*tc+7ub(T)q3c%|R^IhvzFv(OQ*`8oIb79Ji}ypw1}QNCzuLV!db!Jz-_KufS?Q-M zwFtY##?L{y85Q@X*{A>~Sla*(BhF`igLd=H!?Kr@fkYxA1Sq#0zoDSaM6xt)%GMr&D?e~zDQ;}x}5{Lm#Op@I1>}pZ_D5XR!nb>4P%xs zV^7t!cEoRbKv--d#1?TOn(^zm99P8Qlsv!S?Qo&2pqW=ftBg=2fb>LHwC`x_)**+WslB_9rOvt zo&rn1s6$$=%9@@_$jB2qXXH6*eAe+b*tcnDk#`7046asZtOr%Pk;wyIo2;#$r<#|b^Y)XiCVwxJoB8FuYNz-n z^N7PFm3X8{IkP^wD%k;6H=>%`Ps`nz?~DYtr)QwalOKfby`FE^fSaGHCk`?dY%iSP z%;(;L&Cca(?Kf*-ja)fZSLxEXPVqLuV|fL#n4E#u9e9)W)##C1`*nYj#rw6H&<6fO zDFgJ5QXgDz%+l{lXTLHwkzk_Kg;`)qE5in1$`4&-<~unzGlfrCdJTzwX-;STW)?1} zmVli3xQM&=jwv~f8`w^&{Uo(D2cEVls8WAxB``ZbYqlP6e|00};kilO9Me6Fg98{L z3iLD5vkoCiMDDdw!~U=NxL{DedHouo?=U~?xxaJjBkmN#8O&%i_u->z@J|;~+@nb1 z5GFm5auVOhN}^j3Rz1Ybw@j-q&GIvHd>u@s^@*qL5z)-n-t@#<80G7MH{d!k8ModJu0jkgS((j*LB#O+6KVn_N6CQRW%VceQAIv=#2NMb`WxKO-`SiXCqGK)3Z zu1`oT&~yqHuEIMwpleyT+DUQtO@i%>`H5R~d#u|FsAvsl)Af2^1)LL*1~fDK%a$6h zx8To!rS9$T7A0Dq5g(#RL`1*$Z70;H2)~qhEjwn(Vyn`EoC|VBw^lZOD&h@nQl~ne zO$6rait%^8wzT>!loj&5d zZ#jsGl5TIS@w(z4|Az%E>xTlD~XsIa+rO zSL$coY#{ObN-5m`iIWS`BxgQonRSK|*=5;ng1-Y&j120th8DC}aN=f20LcMR+TWSjm&Qcc1NJoQh@6sx+e z`?IK)!f>sg%E)Sl*fo*AFQqzB(hy`>!e@bXty@;sVR72yLgR?mghfWuX?}4aJ0{CB zRvd$Y_vV7{mgIx6ITf1&2A_X!_#z?$TiaodD40NPy zLK0eT0O+=O1Qj<|rTdNpK4^#*2jn@Ftq61~o~SqA0n|yh z!aMW1rX|{)mSEXAHvlUy`;(3E2@Lz9L?ZTPBKFqd^L901NCDaX)8L&*&2by*nq$|Z z1A=Zft%V4A**|{7Y5e*hUyz>11;Z+$VAwV;@6~0k%ScwmVrH+U z?PSp)^1Oc@*2Zh7gqWE@p*KMt+Vv)N9|=0&+;16_gr*58&7EtL#dvW2FK}$>#%x6N zB^*gkJg80{%3;8pZ=Ml3t$`VKumi|K^kRXvh+r z=@rtTb9&20U2DY~R4LDw8-!qw0Jhie-i=d*7Hn-)#!FUah};^mnLqrdS_$cc-6eZ? zEp>~?SQM;3cEwLSKvtmLrT6hZA4!=@HbA>);Rm>{+eG;lGNPG|{Rk&`%Y7rfn%jJD z`%_>hB1Ox!4@Gi!I5QqUhnvTrn$AB2L<$NdVZbAT!#u#)k{_*9Cm>Pg*H$mFaL0?kMR5{vbbB0Z7W?f;avxY@E<)H1ALF2oEuUcMDD zyQf~Awr32yf;a3xUTpo4N!$F=EIRL9*s1v!*lB!3z7`LSNPi2&F#Qa_E|_s=yT2MI zDRZ*SDC%Ub&Bv-(?Bn;!3nE_E@we5!!=o_aR{*c_&JIOT`xE}T>F7_=E9@^AdaDIV zHwO#e*JJG`9%?H0V2=-w^h)^|9Yj9-QT{GgBS)0Fz}74ZT1JXx0p2jB`{xKp{ngXj zd2qF@KB~hGV+xy_EtDC8|K~!pfm1kIEv?UTJS7xX=$6xu{0#dGOxmk+>>y|+#(N@L zn5-(WS6HXrm=yjAp3%FL9Cb6{ICDp_x!#4E?WCY(GgQV)jO~1agG&m|6Deh(sy6xw zrh%aIDRd?C%+I5Q0eKy^@1h2~O+wM?A7K22ZeMGbP60{F3$3n8Ptx3+f0X`iad|E= zu3Gxk75WTwCcB*d+wke_FO1W5e!H+XTzEWfdB-E>UE%!EBs*|jo?0YEh=}AsOgyeo z%R2$tF+LWfqP;iOMvK6*&geDy zt}At?O^!mc@5#as4+$^s8JwE2D9v;>LKkdeq%UezFPbJC!Fr))PXV7W`~r-mdDDiu za8!1H}dr4 zOSR++!QHaWXC^)u7yvA=4yN8y@SeivJk_SQMtJkC}?R|MZxPFX!p%wLT zYX!CZwaUX1%z7OwiL7?ckamyMOK!LaJpb?>Zmkqs`fxlwo5)QTZpWEg0I-0fmMj;a zdHi@IX7Cb>wN9RXdt(ttX639u4LFA4m(GHB6@CPZQvZ@t|a3S_V{Q>`pe@V>Y~VtMN$QbUSWA3ccH4l|`e}XL1wG^?>P#S&u-?1$xOVqVKrQffv<(K_ben2XJww$uFxrtuZX(_d zucU^53~Hy9=VIgFZQbIv=|FOcNcgS0gp0SLI8(4TE??4Kcqe+?ZK3m>J$4i}q5H%H zJeh`kXu%12WYFo@ha}CQMLKCYOaNwu?U?2`W>U#iAfB z;|^+mkVCK;zfA~z!hL-IQ+1leo5pl*y!F$YHQJ|NKnAjL+6|>o%5lcq*6M}19FWIN zdRj&k(%s^B(tUC{jtsk{vafjo6&aWja`2)cJNWA0S=eu*!cO0pMM4s%|-Lx}2}y;!s->K(b40 zDq4Vl%DN6M(bvIlt)*2u-+n%Wa5;@}zOghohDVRPkb69VOE90Nd}pAj6k59Ya6#M1 zAP4K<^5qS7J~<&4CRoRpiv6`{9dXG}-mPm4KQ6zIyty_IU}5v0d~r$g9WzIU}`y&J!X0oNBOT_)>EQdg{hK>ks#UQ0djyt1^6KBP$9loLIx zd8v5*OS4$JrDW*N>K>f{;1?J!Q(Cc4FaIK;BV$lf7JEC8z3oXls4bIu`VH0_G_&yM z=_nY5r714Q<2Z|<_|ot9A8&>W`g5=2ur6hZHuBYE+#=JX4q2*BGp%t*IP$stnx9-d zGR(yBP0>8I_WoY;6pl1PIbNTMl|n{3@ryT0c%rKh0HUU&FR=Wm>f`1KE=y~{k{H2@ zsUNTZ_N*wYhgd=W(La3si~9WLH?h1H3@H{yGX)(|FFAHnfB2rS@oabC)znI?ng0eT zsDxH$!g%VX>sYrlcVe-Y`<9}W_p)*`#F`Eq zg&_}dDMy)F_6Ukx6vud&rV-EslU#@yj+>6KOA5ulNv`NsD@=p-w_7!|RH?!^eY)`0 z!(t6Nf&%i_blt8)%1Q5o$L!Yw>%XDc-;lL`uz^N&kNqV1^C7VoalI{@9ax&}S^{8j z++ugAxbPWi z)Hw>k-9|jDrgJeVMt}7$HfF37`V^dcb`B(wm6ndz$791Vkd4t~^6D=*Kr@7RF9F>7 zAu%nI1~>6}>u(NLs*tIy78Kfwf3=NsY<_$gR+l1o1`WBOIPGO!vQWK^?V4lHygSZ7 zD&C@7)ye0K78e0o&xceXzmo>=Ngb@64YYMjN}+nxnc|Lxkd$VVJlj1|+;ssdEGp>N zmqJN7pERr(Z!&8HnVZJvFE{gD8ri$5c)uYL|H!VsAU@2$R=a4Tg>0knqdIUMrQOha zEHrTMbN}ow?I8ik+I2&|pc((&2_G164dMo~e@k)PHIi(@D(Y5zfsH@3VD5b8>5!dV z&kt&X?AK3E9uki+SRW@uK{=mU&r-Y6&2+QS43lamf zsZl>dvgsamN;Ff9T}T1Ib3Qddr!imT)@UCgQ&6n~b763v|GKKOO!f%z`0gE`zwh2c zqQ^A&F2Cd?AVKalzEEc-?u_f_FI6;DB6svOaBRvo=lAqvy~)qmcGIq_v2ivAVLebU z9`~ARNtjoM!o>ZWm@F7wH+K@ysn0+QTR=+i^-jCZBJMTZ$-34nkcVamX7u?G4dLJ5 zYU$RQ?_P`z!>_A-??TZREvoCBM3^g_fs{Iq>g-UWT#*?~`9ODnXEF9FAQ@n|m>-m? zEbw4i+U`jQW!^7RE0kC+teZqpbk5pIhl?6&&{v|LM7?hAK#H|WU@EiBfZ3s6*v@9W zN2)9#G$gWro=1A*6^cDsE~^?^&&M@bHWX z+n$e)11`@S03*ru#EtAoQ=c(GjT%No@6G_f>KU4^kg7LtT^-Ovd0$)YH5xsOAjv7+BWkg4zr z0!IcgzYODHnZo9BbUv>Siu6=RB6nz!mw|BwEX5=i+X}t@fb9%nr$HWEA8;LYb#It4 zJ+V%tmLC6s1&K-lR&W8;(i3@oU?#u)D@S1q<~IaRze&YVO7*h5-lvrYqd6@e(&4%u z`~5*+2%Xw7u0v0)Q0R`B)5m|K!MTSgBo&VYpp(r?hDK$zYM8BF*(Y9{o1B@`;(i4c0#l0 zQ{(e#Ma;eCh3vZ8qg#EZAnLVW+HYh-16|DbV7 zz0`lTspg-JEpAqhtjfy&`&6)-|8_vg#s72I67F|GQAdSN-DTcA|2eu*Xrxfsgu+#H zAVc?dX;lS(Od7JE@AU;a*WWd!IdS>$X#3DG;SY}={CHy}BE>+3zQ41Aw6#6G(1g$k zFR3}7^ycnKvB#`ZnG8lcGDESl;8LZ@$BycUVzowR9lx%CECRl`!rY3m)H~Rc+T`~v z{GBh-{c|{6tUKLj$*=HBD%Ea{wx|l_mxg|c1?A%vF9X6Ak9_vr<=hdAFs$1tr#d z^9>GSp4@Vnf8^+ySF631^O>Yv^pJASUA*XQFKZ=9Cp>m)V?gHU#v9Uwpu7GDyBFG& z6e8wIg!e54&42$_WR&`kVgCPh5}h%06dl6aixsBtZRO?v9B~o$(8baY!3u*r*m&9h zYjf8B?M5wm`9B-A{+#9eAA;A*NXyFq_k-87JsthY8!`s))S(FLE} z^`e%S@5t|iyZeFz9}urOjrks4a<~^Tfy2za7XDjdvUBOdd9FJrgo01~`{M`Gesrj3 zI2|$dhR4Eh`v2>3bXKJ27)-bR?x$CO($LMIq?k+CMjv7m8?27lBw-9|lsJPuXaNSW zSWNq~Lw{pK>W^R_*B$etFN^rSS(N+!Pr^SB$963uYm3thOti%-xSz#gC_fbP?O*x# z{Qo`wAFyoL19g(3v1A*@>>{O1!lF+D3k9yIjK)-MfU2rBQ*uiMm@ra}QeaY&uc&Cr z;s+$QDz^ILrxy4ksW}_x#rFUOt<~0mGx^_A7)`kU#n@X0McOp&g1Ec8GdPXA%iuaN zcq4LB|FLYt?vrU-;py%Xi$NSv_ncltRlN`T9ZhyX^@z9Z9^eIX>pi07KMN~i|!>^9+FvJ7VVyg)N8j1D4r#@o{}*g`L2MTvFL?nQ9!{cSD4@pMtkoK zLFi_|=TeJ+s_UP9&dIf1s+Ky#;$(o_&#B!_5G^jNA)v4?-55OY@w{zzXzE}{@aB(@hoB6j(~Z&2-uBCE zdfw7s#{HYN1@`vxj{b^;H&~CWvGCzBz?JSUkB0{zr;^b(O{ovQG-}`jX$X9(hgZs^ zP;VJimHJgmEoHHkSEH(ar`HiohrGreNFy)*!}DqHe7^OA4G42$;cebpd3#g%QkC7&-=Oh*{P8l? zjgW*C0r)pnXyKZ({jFgNI|HPu4nwqbqS#;eM)-EE>H)0%&C;H42)OFE@>6iXz~($H zkn@u105V}~P6HE83XZQ43IuPbyEzx#bW3{f(zx6PS5~YXCFxq=4BH zj~_+}9fI!DW5>DgLXT!d}Q-CeJ-GiTBoS>hFpmy@nYu!&O)tV`M}B6!%coa z!kxbRnb(MG-naAx$v+yI*o36_Q*wJABj-;?E*?pX?-&;ywVtXaeO90`;i|(p(BpOi z;X_1YhxbyK!!!fJpj=4=u_g*GkjblrvFb7P*TCiVTG`}I(ippIZBthyD)%XWHBLu9 zP8HjW8G^B)1DAUEbuBYXC~!Md0~y-1j^>XaKyO*Dq!coy574#*~8{xs=Ihw{#|qUDnQv_R+5bC(TxRB~RCS>7@#<9!JX>N$e(LKu$!*Mxi z^lEuq+fyOafqJ2w^gwqOBLeie)iBXOWY=tkgllJZL}?AsTa-J#Uf%i~tYFl({oGH8 zGxpT#RTi0>k-yrtBhjl$k9lE5Kjip}RF|&D61JvP$6$j;5((-3njBjUAohxWkNVEN zBnaqrz_3u?#qA`bU-@!9D93%SesHz7ca`{br7(z6nXosn1bF%Nz(7=`+y2J%LqdN< z&hu1VJfOg>FSYCSCi>+$!t-2x>Mm*I@2bPT1(AEw_4SFTX1inicWub_z)Ni>YT)cH ziT;^qcmwEm=2Np^#&(h#^w4yM86xO$HMu*un*ZFR0m>lhnLKP9nK}yDqOL{()voQ0 z&EI#TpU^&$(75fd!UyMR66oh%Ltg^{DM`}`i7Y#C=d#_k<#qABc4zb_yp9lfNbsD5U3^Emti5yet{SB{SaEm#Sod#tt^-QiqJ(Hr0*y}grrx}+ zYAmIoW zJzvjTfj*t6CmmQftwKZ$RaWUlx2?>eMbE*5_PH31u@(Kfg@EVfgU|oO=f5PS1|7LT zPY<{DKN9IBc6EuhjfGt|toR9hE>HqG5LF6CG0EpJmXosr0 zY<4%WDc(cruN<38^v1P08!qVMb_cHB~Thb(_Zh;_N!RImR;y?;9yW8M$~ z4qgrH4o#iV0foHQ6n$Hx>^6LSyf^pi7Wzn0e>3lt)#Z=>De-VJsui{ptvanrKf?Zn zRMUSV@I;gOc`ruHN6W)zLE{^6yPEe?6IR~GN~JGLH}_n<%^Ut)U7NoAJ5U+wCpOap z2>|UGHq1No{-9=tPprTD^AB>vf386Qpf@OY>3B}9*wL@$v8CSFZN0$v4;RxI`jp3I zJICj<`e%zzBwRS5cK5iTt{g?IYVF_vlDPjvmMq#|Ke1ix>Yoljv0Tej#^aMJPy_V3 z<%p7SfeoJ+N$~$_;9?L6(g!a0%I*Bq{%%k%!&!gjbB@G7z6Ckl8w~EfV2!tb=KIO( z+yM@(tKDVrRR3nZ@SnJzVJTO9jF^>sdvWq zj%9Zif6~8yC?E$g=U-O%|FiiJ_w*i)*#Db=HtZE-I&AIjc>aH>%0I-=)=SH*m+k-B zONxK_Y}?cQRqZwgiSKXub)SmnxX6*+_GeTlm%lERD5kCmtAh)j2C>hCS00%a(^k)Zij--~VjlI15|1^LO#QBpfK zJEUK|_AHv&ou?tL^L|wmVnkRLsO|sq>A;aUSbW>gPA|cgXK$f)uO!`Ff#ug>fi|z3 z;G_jHS07g(Ve%e-@nQ~i28w-Y9sD9lURM6kVj zeA+&XnqQFhVd@lCcyIY&iLe8{TPPQb-|UX9m8oQ8=L1a(+ivro9G?F=_b*#j2y(EL z@p*QXXJ9M+kt?{^`0#Gc)tizHT1;C|lZOPwkR{~&^kCSN8MQ9!l$WEGfsD{dKie8H zpKxyOOn|w*Yn|;aQ>daIcPoFycRDmdsj2o0`rnp?^wTS;l7X+rf3my&wSE3iTb|>; zx8;eW$Wg>a;Usu*VSDGlzx`w|;DMH*ZX~z3wyQS1(>UoYO|Q+AN9#)`w`7e+V|UP} zl=1X)FCvC_T3IZ7B`1$p#2kur9#a?Z;0R$|`Z~a9)3^l$GH@7?1*`IWbJV85q+Pf! zCTOrH!OMY68Duw^wB2c~RPX%CxY>DeaWlg(!TD2N#O@Sk+V2(2LfZxKpBBmfPbQB4 zW07ne?EiHds5LDeyVXH#--RlJ0D>)wW%z?$9c5hURXCY=kT%%G#QH7?1Ofc@q zsJNe3XN$wgyfknKS@~9US!;4DW6b-YbwfYH+iCEIjK-rkVRM^aS}X2Pn1lc_^XR7D zW2s&RFQQy{Y*eU*O%VSI^Ou(;=xByo9tpAMFnPb0#LofSqdJXEB!vG0)paRw` zZ)pB~Ns>sKUbneq-?JgW5ofej{m)vuS=VS~Llw$kf$4o1v`h-Cl(21+2(xqxMs9VP z%|c9*C7qaMmM(6A)?Dqy#IbJyQ*d`~xsm?dWm72UK9mbyf5QQMU4{36!`Ke4DJq;V zse{U>X%KFHAS3I-7+KpOWSv8VQvx-Yd%H!(9XPLq)PDJfS9MK{G?7V98hkG%p~prg z8c(EQ9#)I%w#N4f{zpx#-MUQ?<7CsRG+!M~WHfOk=ZU8qk>^4>6SH+Z8m%V`x?~-G z*RSTH2mYq`hB7Dv3e8uCNf?Shq*%yZVsks{M;p%+?>Dq(C-Ld(*kEqHQ01&jq9kkn z2_~n*nlO~c;Gg;cQKy^n_EUUK#XR-Fk|b$N46tG0KcdUZj+}*_@_Tky?UfMPIpUYA zcnpx#BA}vD@+hY~sLGng81eavSUHXc0UfFf9ur0rqbvcuO2ng2 ztXu7!>rD+$jkDjwO87FDH7F3#t%re8(311`$b)DP1w^pJ2&4duJWvJS9l0@1MaFSd zGIN0WftMUDRg72WZMYUoVn9}55Wq5Z>f(3Cimk5dV#9IOF@3 zw=;b#U*f@xRZ6u5^`ED!AYv*_9DZ*fiR2jIjGK4*Qfw<#`^fTN@z;B~i}AmE)!%q< z%dX28r1LCokl*nYcJ`(E)eW&m!ApGUw^wVTU4E)#W!&dW0R%3J)gDz-Q92|mE%9H| zB1-qYTVX%d?NRr~323nD62=r9)8AwO0OaZDTNv&zZRBbuSHC=>$)9%vMS1F%{KweS z_FD;bVNP03XH{skC~-9jVvRLLGaF=nIW?2~Jb|(iZdW~3*l$x_r005|OCHqhM-q$w zy4li#d0H?eo)}ig$9O*grsh{-&oOn73#mW;`)YtDqEiyw^b4fbV2aZk_4xFkEpQ_* zEK>Md+R<2W{1?VJ54k{TuLBjbAAv%UwMWHMu&G^N!c;syNc~TH5Nm1qa7?VjYlraH zKkJx?7gC|`Yhx-Zqz3bKI#G z6E$e_E;nTdCr`NSz0d5eU$B8qNMCPX9LePh9kkTAdp&P(TFJ@nw`9cqyen;M*RIA+ zdv~{ycXZ!Vs}VfXT-Vz+sD_GIZ$smhKT5K2vp~AH9)VjOe zj(=%P#_VQ%oWEKG(DAJ_+dqvYWuS0v?XP28^eJ)BZvr=?96Z$duiCyz_o*0rzKZLqo7MPF)u&6dH6W0FwQ{$(N>{mU*;(9Nk-#HXqRC8)eZtMtGu zmeFB$L!WFC29aubrwtsbEK6fzL@U@b$#HQ1(MgAVS5#Z5-96`~&VMg-rK>3MnY6EU z_Zs1`)1%Ckx9?_$BRTg}^rXXG$Cq(Vy{9p7;%5L=gIEquxozk=TxDAR)Ws?PUMo{Bo8!=#5uzub$N7;mdgkf~`*sF;+pU9f zCBd1|5VV-T@5oW_Q}XH(UOtF8Rs6V6WCHc9pLYs=gE>r|9fQYyfd{}HxZhZ9lj;!Y zj7xzIc5F4AQB9qf>VpWiC@=x70`ob{Uz0EJ?-V=nWuE5R^dAY=!{D$shq&5J=27)r zcu~|>7y;&|d-LXii4Ys~`?irBPhyHUUS;=m(f=*6oD>6mWv&}T@*e{TCw zsM67u?!Pc!N$|?2-6O0&Bz*X_Daiq)h#ew-zWz=jj}oia!IPtFNpoX|`7(bQ81b%> zO}EB*FK(CVqT;u-DFl53Z~F(K!j!eNHqF;scNPkRay5;|_+^xx^UkiKuj$gMt-U1e zkHSvXDH&|dwQ~fl?X!L}^dIJ0X9wPZwI?w91(Ae5Bl9+UKNfhRh*7i=#pBGjdv#0l zlrmXY^0=+wt1eRZAzZ39-}{7raS zWZpmY2;|fyR?^GCBvA&ia7u2Tztbett{w!U95fX2{TOyiEG3nUyP#aLp(hvIf08&+Bk2Fj z{jK|G-mJxD_Z%f@Tr zZq012GQF{uy5a0-jtj)M*2zG>f~(+D(#XNQ-sK%hLZ+0UM0Qm+ z#<@~mp6vumZnQ#6YbaqFv% zBj&DQD{CrzV#sWX!LkwvAMRZ1Xuztmpl5{3BOA}%pYBVj8_|GLTs6MVnF zNa-(|$Y+ta(8kX+(}gVsOo41ni@g1HtBRV4tatU^HR?mlPr;#8oWrrKMwN{jaP6az$SYTmK5tb~> z_=ouGE1m5Cm;$&e&WC4&p6lZ>M!XgVbBK~&JBE{@yI;*j6KS`f$J{E{u$w1JdrCOg zbdPw#klIkOb%E4|%^Y3122M}QxqQo>14uJ)C4#D>&Vz(J(Hx_%$#kdq)YEz54w^Y! z0&JPl?STO;0YRv$m@~tEGu5;i$Z|nw0tU2l!CRxM)%fk9V`t2r?VavUb)m5vfku+qT5i7P4JsaJbjTBqH#-BufUG)2@w^;;}FUXD+1`SY(bbcaQiLr?TEG!NYru^ zQHa|%Rtgz?l$(FpJuj$_P2`D~xCd72V)+D~e4v)$(yHjJA_<*(*II$Wkmz*j=a2=9 zN1Jp%O=d@-8SdcSL%<^d(iHZ*uws}7dFlYbV3l2MjSHfobgSB^%-G+Toa;MM$Be_9 zdWidWEWS0$I`2XTc=k==GF5I;8dal&v`EIehYFMCM+XdtZ2JjzSF5 z|0$^}Q|4b*C`kuJcdy#~8E|7^uz)H(D%i1x{T45@;XfOUDpI9`s0;7jhb7>TI}nDc zxyP>R!IOx2r!na&enKx5I&4N`IFamaj#kz~ibJ!A`^GQvMDtZnI9@(arf|BwFOzY0 z4&pOY)~hUS%y7F2+ft|(nTgkwJiTUQOy6C<0M}4vGa?ymXt?JE&tOj01?~x&1pIJ@ zz?Ga;#9~?{G$j}XUg@lajbHs+*i&&NV_~pAhxW2L*Eupt*B)eZk}}t1CT5$Plk*Xs z-r`wHxGEsSXQE~u)8u8t0yIWR3_SzBOiL5Pu+ z1x1Y7FYEMKlCHnX_fg2PHkp(l!#FR%e&E|&4tbC}7Mm^c#%M(om> zE57F7zc!}8_BDxg`{aL6|46GZlvWyh#g%tmQjTUKA%7J-harjTmhY2Ylnbx7Avc5I zt8-_-UhV}+h1&z6n7Uk*Ig_1S{>Ahg=11!>A1P+vhB<=JpSB3nhd4n zak}`6$#*b+h%my3D+AJsE{Bm_<3c+vi@j!!4eGsRRmKA+iX*#)NW>B#VoTg51v93X z2^u~ec#ZAxN0`=3Odk*!bU3`LzC~8MJ=HP%>I@+TG zANYgd02LjGjM@sA;c%~H*^)yJ4`YB=Wn>6eF%VO%)(@6}qx0KX)|=woiy9k9B|YLVlc;ru#n91V zM3pJ8tFh;6`NlD&B=bWEU6y2{8;JXSshPj8dzNNTe8FB2W}a4WBxau$)2(f;7WB_~@c5$?JqtKH84C~dqNw&E zg^S;`ht;NcP=E?FtOZ|yTfYb=ez*IA*{_xCN012^oEW^R7x%L!8=%)6${E-gVGDEE1Uv~$%s1>i zhpE1;B;7OFXOJ(har+@7!d$x@A4uIEoTJ_(fo%`xhxXUU@qH(Df8+JzI+g{u=jP4wt_#8>q`Qsn~0oB~+Zha%1$IlSjhgUG8Aa~GgOvha=c6m+Zhd$4j zZMDSD07W!QB5(S$NpxytSsqz`!j|n`2=fg=Omu-b@C?6veCmj|9oZKzIv2OKlw=%! znkMx4huhCn=vjE?7#@{mf#{2h*OsFSFbam5VyAb{D524#B&UGFji;T<}(qRcpzA%-%h*CV6;WI7l>8^B`)>!vWGudTYt{2t!yvDje z8CXxf!Qht(Hs40l94CAvu_;RS=?rC5c0gqca&dCWjf&pxL!`L5pa=( z4vjg(n44k%8}CsY(}-O_=I0d67(NS1OpcecHxU}{{v{nS=po4~V`>zDi%m6$7Lq*z zU!MB?DzK!ml+>BHdQvc)7QOGz$~3luH_zI3A<~!{;YznATX0PwMgD!^N|@LN=08Sj zas3nO^FKVnpQglrnU^^ee<@~X@XDu4%&yQsb6v=tBYIKvFYGDUq_``Xu<=cIyf#K% zB5mDBs@|u?&{|5DKsmt}1s>v~?y@Fz@_FsZO}NZkJ%VlF>Dad^A-A3Z-p^8fKR0op zR9RK}Tlslx*$352CGn(@52%oR7-VUN-5GuZsU0O_uu6&m3dFI zJ$IozNAQx#)_+^&QRXeCZk{Dz*FWwVkhS}-cqHeBPTbC>(dRIY9L_N(CQsVz+im*{ z#%?nlh-1{(GNx_bR0tp!KfFGs7>tJH81m<-LSV(#CP_hln6To52NOyIpx=2g2Q0B7iqf(~Oi9WbvScNO=C`wve`D`9D zZhU;82;O7xZgw9Upt0`=^B+ZMKC2pQ+7W--Sus64xyRbg#u2M-QBJb~uMbuc}n zZmY&ff@J7;S)GN8MU;3xH$$fVX81=pZ)EL=gRG1Fd*X8Wd6WMu2w;}aXaW8U^%WPN z4ANK8ON@u9_4_}W>pc+HI=^2jfZi%)@2|6G3-g;HVS4g>nuPWu}q?2=M7(6C2o4kqrI(UKT zf(l@YC#!Sn*>VlcSx4l5X|?C6%P^|Xpnx;y7ZX6m z#$!;7^3t^dROh8)k4UA;OkaMgLdS`Py+q2E5u9ko3K~KDC~=;!f#&flHubwe8gi#%I z;cQYEDHD^K{0Dm7%P-AiP>-F*u=}T|2F~VE7?kfC@OZDvoD-l!v_FbJDG&`2Ttf?u zeg_)7Bxm{@SFkG5YawVQI&JFXk?DzIF-1v&Ato1TxXI?5$WAgyrhIxrs_XL9UO znUsgM%|j?$QLg?10!J9r%xZSd6~sdGTU<38B+n10^sK*@C_i&ZLY^vnF}Qgs;n$3O zbrAsU6bz0Dl9CBJ&z7T?`)QHsWsY!zhNB+y&~$S+x^-{2EqqyoZj>=ezl_B5GR274 z<#&ZIzlTd?SxtXi@yQ8h8g@7&Y0eIYE=F24Bgox)dl}N7p`eYZ=U?V=T%wffiw}># zb%`Y>QpAXmUw1eS9E)G~`Id+alzjXes&MCld=vvIjID|Gg?TV0O;lhCrM&Z3T3VbldH>EsQ30S`n<(#vU{l$G4PsN(AEjOW_$A%W*M4>L|McUX0$?{8)G z{gqiWXEKPa`YC5g9%p(~!32C|4p(IcyiFyny4K?OTZd>tPGE`N=&w9PS1_p_19LhS zZ4YfuyfBVvEteuPFs0CqnFD(Zj_?m@v0_O@o#JkDFGN!n!IPZ02O>;Z*gN8uw1nMQ zP-LD30ow!?0A|S<>(S^(Hi8O)pas&RORT6D*H$CL?#D5QR=7u@=l?Y37Hp9=5Y!=?zvm}1jIWw|Z`^~($Cv1uDSds$Q z#QVgNqMNe63u?}Oj)Oz0&`YA7qWI#ax{amfroavcE29;)Feg^gqv42+Sl6TxJ@e&2 z6?Sz6qK^6yCz$bA*{WsH|85FhHqMpjwHGAjX4))PBhcC|`(*C;DFfmAC#7OR1g(XPB>E9a=3|aH5)K36*$b z38h_kCWAtn);+Z~nM>?r2(D%OXkkaP<|XfOca2TLoI^Z)!(4Xl<{@;(X$AIzk8PDn zSX)^>ahmNWqKc2 zct4V=*fpn9TD*#}=-no8#*h=U)HNinvu%s972UjZi@qXQ4Tq%HrvQpeidYYKn^&Xp znpx&&0FU@^VR%}t$1NoiQw;mBuV>i!`J-X&;|C&u2IXntFo@nIq|KKFTUIo+YipD_ zS8oN6v$Q!l`1v8EZSB?+JtySDwgvMs)f`zt*j%+T{n1K4RoH6pct{B}@c1QvIF~Rb zD_-N%fUGG47~5|hbBLvC5#Hg)-wFOy73RSE(APoS++tld0aPH)@OO8XJVsS4n5zzo zM@&A8z?e84>S!@QSXlf7t~Q@6Q-(9CIr7-g)rqvyvc1bt78EvQyD?@kwWEVxbB-r> zeqO714_+f-$G>sz<*bXzTstCGRSHrtQI{CqnB*i*yM{bcHza+f&76qYplvDy(Zc6w zP4e$~uGOr4-BS7L>9OH>EQ%H@E+S47i!&XPbUS7nUP$`{X@e(blA_FE#7c4>ys?b+T5A+7Z(XO^ytv;loy z_DSu8I{a>my2h-NiaZKy=ihe1Q!YzbY+btad_b#J9g538KI%K?#fo?99KYF&-Bma$ zQf(xzIWYZd)&H>n7)bA0>w0?1V9!zgW#%*~e_rKOn9MWgm>QvrVfK3`mu0%Vmcwm% zN8~qs)%PZhLqY;-_qmje5mmcWOO9g7Bmd(H5Cl=?YJ) z-(IS}3-SKFgHUQx9pwPQ5L(!teJFl)Qy<(8GH#a=1!naid#ff`yFIeVECq_!E4-9R zI*)xuN$7wi>Pe|`-^kFJ_WGB=z*h!S;&Udk`zP|VR|qSV9scEAR_hnwSaGg^$1AP2j2;af-vfQH&s#NINRFODcOJGx0h4vR_D=N$BnWUw9GOX?M!==^YY{J?5q%=*1jxJWi% zWUyXrRrH;!spLEF(T`c8*6V{G9*c=<$WQ$*MG-WAPWYHj6BHb>CvE0!V)TY6OS6Xl zm_bhE!t|uCA$DLM5PauvY#-o~)NrM1B$2%*N5-bQamRGsODz{#V-V|y?Y zk&nXsDMJ?@?GC}e`FN{zmih1|-z=ysa-n03m)WLosZy0%Nlc&@ESu)gksy@z5n%%# zI(2l?vDEbh=2rDOAu9s?t`0-8Xqt&>dFG-cTdzR(*|1?Xf!MLX7IxvM->()s&aJ2v+%bf)zY7R`(#4Tl<2jk`HuBb3g;U z*f<=!b3YQM!I&h~5-t_CKp8R_9;D0)6*2uAYn0sT24~2nvScS)%z6Jz=!zx4>r*}?BRe~0ei=|xoUza?M?A7rE5eF0rbyb)#niT*F-eu zzQ!YnnO}L-LP{OV`HDj4+K9k+C!^ymh){a&X4t^(h_&(l)$y!AA(Th%$$R9tu`=(! zBoKjrZCShuHDJpgt%hFV)5+I57yB1%3lP}zxj@Ckp=p1A^7IwP(rwFxR+whxDQWlWn}M+$Ur_nwqx@7L3FX4irfzY#o6;SGoNU@>he4tror-a4?BTx3)Ya7)0~g&Y{yhZ-1%_6Zpx zT0?O&D>%+w2Yh|myR(X%k+NK7&I4On@dSgZV7AJa*g43Vj-xPTm*k(_lc|7v-Yk|k@<7``4cU^aw9HQfH%kwdg$q0nwF6p0RIL_2-5Or zzI4QSYGFHY!b=9u)LEGO6seFS+6tzXfs1DRp2S{opW~Xa7RtQ5vWj$PiBD%T(vKf4 zRTDj8{}zHxE}|gF6*V)(We&}X3Fnp7Y)ovB%^DF!hULunjNGTdRp)!04aBqV z-7JzlS(|Z|+#f5ESjKdPm)+LOWwVk|Q`1y0LmUWnp>9N0AH#ucyt(I}B>KNw+;J#& z7^f`_!RwvNn?ZlL1p{=PP@1}7$e9y$<2G0O)e<4CYL4ufcZ249z`I;9erK34L(%M! zD_wdUT$(KT{bYy25in_%&PF5KrSl3ADuvF$ENfe1{V7J?QrlOPG6#lWaBL4PN^KCA z<#>7DYp9d~L1;I~PMsJSJMZ5TYHyD*X%)4JiAg@B1%VVX*rVz+s;eo!40GaBsmL`btH5-J zl@2bd@&%u$&Qc#Cb_Yn>Ly<6?WKAR4@kQ%S^e=aT7(HXJ zn*2R`b#oSXI&UztJ&aF?gS@2FspW9f@Jl_b)l1{m=MBj9-g&$?v5cg4WNdWTckqJA zuPE7S#{!j|iG(F{CP9-%6dXALOoEgS@Ua>4?2yuh22ONB{M$P&>AT65VyE?ACx2xf zJ50C~7E)l5gch#Nx&$aRsu$BPA8#$r`oKD?(l){~Tb1w%J#dWR(Eo)Py6GNjT>4fP zYkk(&a~V)Y(P?8F0=dtC4vRpIDmgDPtX-}W9LFK;nAOh!-gzseeC7EvvRcsMT%%^Ovm_(ma`+P$+3KTP>g<7)IO^PFa`N|L$9Oh=hxziqv1COQni#VY*&3rS z9zt-100dWu1jaGX6Mc0H>j*Ue0Mt1_{a)UUfrUIh^@7L-7dx3nUsb>(w31O*j#WDY zb~*GVY$c7epH~9s{&%sG&@PyixIwk@VkO(5xbDD4ZaPzOezz*`-o%(J>nI~r3D+4?mKkew9=+SbX~ zSj3Y-WAaKT(ZT!I`Fo(-%fuxm!tYZ$a6kOFX)wwU0egfbF;OX^Kb+mC_4yu$qsdqU zEhBi2QAKh|!2gqN#KZDG zZ6h`|?*G~WKA|P&v^t31d84tJfS_5BIc~6CIt*KopxK$PYpkwX>kZ~6%1m_{2vc2{ zq<@ihM1qDG*p_HgZnlC$0X2F>FL+EzBVe+(v(V?Qju8Z#hxFKJZaJ8_S$M~V##`$dy=#lA?A}MGGh0=)ncn$P!P*)~ zFL*XY#a1{`d=2%EKseOZdfjf+Hc0=aaU}^^@$Co<0}+Jk^6ie%HZJIf=|(l-kD*7x ztoog@%bs$FEaESi{Hv?|Mw_@-e^P$k19o2pY*mkW7aJ*lJjwgL^UjnOBF{tN9c2w+D@X8Dm!*pHyj{?_+5Ap!ZtqUa z5_grAM$FUlyZv4l=#OCLt?Hq! z3BpAT9~A}V8&(H1r7)6RyRn>=u2=ow&`Kd7#q;6nI-!>iL=g^Qf+-pUKT8}O3}yXt z)FH@0_Cl%)m*v|=c_QUoyA!!RvVCri)J*p%5v z0x-yfB*@0k^3-Si=+z3lb1h#mML)zfP^l^fr!z1}-R4QqihJmcY0NtnBGGX|;E|in zQQdjAAvtVm5MVVAhh_xR42>Z7zoi^?q3C{L({^rJnR%VYO3Vl{O$zIvc+#|1sk5=nIV2V!+a2adbur_?SPBb~}J zGScR#mZ7(M{tUG1?TazCY30GPW|){cruu$8!UAt+WLR}?P73WE`{SoB7#SFY{kSn} z81wK((F<+%w7uI`?P;I6(k##%vOlDRH>h;;F8{0QnXeRfY%TYk_b{*#3OymTJ96o& zQ5zI06viS#-3UsVy%ZXPHUAV%eP9hdsr)7aOXX>e2L!)#T;M6ri|yCUtvpz+S*WiN zG0`b0R4N}SZBl2ckhY~qfDl7UamCV5MFv4F@%&}NJT{qQEv6+-UbG6+SC4{g zq8iMdqb|$*Tezjc7FUF>VK;^eDe9N~Vdx%1k-jSJ(;hKfWBOe(urTlpl-wcmp*^Zl z@<%wGQKkAtrpDMvh=}rZaiz%+baL1Gbm>rega*p!Ki39s%O!EYDr5&Ts8;*byZLTt z5w%G5mj9m6GwMESy1YPg?ZhQ`D#M!>pCEVk_o0Sg)2_iB)eLT}9_k)OMO`>T8^Zzl zz*;pHeKF5S6D?&-BZ*P(LXyjbq~!jjQGeYa$pa&Muw9i?whwVC1ykG{-c#r(>l<+z zFa8OZ7=p(A@SqpTV2!WNokcE<9Uo@Kf9;zx}{;5g1?nB-JS1d_YUEz@4 zds8T8fsoa4(N&%AyiumYQ%bFY@GwJh6YD{6;ISxzDP!*g2Bnns9zN^$brhJRr?L43 z?l?1#akq3y0y3)$?jtVI49Qf-1~UsYuHDe&jNll8NB;v5KsQxChj^+=#L8@S5W2K< zU(eE&luM$8fU@XY%0^7^EZ&tb(BE~ioly`$ZTZ}JiyB3T;2O4q@S0VT|Ssk$tfX>*555U7!0E5P%Apoh)tV&5D^)sEApi$K`nxL`B(k-3 z)$n>7=;JwcQ(LEreeA=R_OmRffGY3?I0q;{$?#IyKK?wN4KMqxq??oD*I%#2_h>c? z-m9D$-24#PL>xm}_Uvz%?tAId>G>J{Az>zY2eXUzqPj z(iI$Zlb{Yj4BeKLw6EEG@!*NVe%Segj9IrYc?BWxR`cpjgctx^rt1db0374*-cR(5 z3zdGNkt!_4p#m zLX8%c61|4d|HarlHfa*D%erm*Y1_7K+qP}np0=%No70%KZQHiFXYGB?i4}3)h`s6u zRDHd(GV@Bc9zE<#)%tA3F+Z;{x%;_q35RV9VgE{Y$uxyAX5z7XS~fA!d%xj9Eb)x# z`J5)pwb}i=w)B-e;LJxrCvQ41|oc(kU? zp`|*+B?vhcNhky*c%Sa4>l^p2+e;ySZXJfPCxmKsC)v(3p5iNM}v-ubvD2XrjuOpw4dZ2E}K+ zHwc;rC%IWKY#CxM?|!g*%#O@wR6c?uZqT#N%K@D{s8o>Xh8~}ZMgbPmkmzLy?C6OZ zISf#N}|Lk z?-AZ*wTCR2ns8a>9Ltkqt7U$pq@4KkTI#k>Xp36oU+2HI2qIgXbfe0rs>cve+}+)I zKG3GoOQ!RH*69Ud^n8oUQ;fm2&O+RC|MHb1gs}6A&eGP_F8XY?cD#-bIp9;@0=O*Q zYb(|H6n{O9Ur(#RQM({AsIeX;A5-`^&`t+$XO*2DzZa~i7`22dgmpr>!M=r1tOizlUSlX>kAs_$X zOG-gZr(xmZZ-+N?K34dUYEP!+|KO1~Gm-Ut;<~6g=;y*%C z!{Wf=ZJ7@u4CsNqHfJBh7mXC`gZRwxIBr*HZ#7DYVTFz3FcxqblJwTL5^HH}8m{iH z{mtpwpxF{wQX&Sf=ZMpfM1WlT`sDv@Seez2OTq~|w9Odh!wF|2-9;b2$zfW`9(r{_ zTrtSJ^I`6y#qIH8jisHr)T-o-WqjZAD&R6}PMh<3$&7<|q>%G) zrv80E(I;IhbLP9mw=xE3Qg~6Snn5v9stNfMV-WxmPNlI5;tm+6`EsVAF1tqtpS%g9 z0p)lJy%lj^j<pG zK|7|_oJyRcxv$+rdAFol8_|y+fzdk;idQd4f)Lj>oL3~URNg7CJ>#Y6fVJ+C1^pr(HuW>1ky9P%ZB$89VhiY@dd0+vAKe!&IP3VtT zh&+gp=RIA8VUAjU+=hWy=X`7K`Sg^7%kj(cwcs+p$m<^tC=|9o|SZM)EIM0S7J5{RCkW z4?4jyL;en=z8POHO?tf7mIh~g3c9SU^OdB;Kn|w@GCFjrGffn+s`v`kNUN06>@L8e zS7!098<)Smq_P8lBSoa2X%xc=AxTcIFUSBWdmDgV+}=7HMcB|0Wl?c5fm;cC67w1; z4znrzntv|L=*#@kf>PPVz%{&m<3#IEaAcwP?`nT_WKur~gFRC2Q!NK9f5hzkjH=_j zTGj}nnm_H9UibY{DW%(AtNAAqFZNsPDhA8J=7nn~rG>eS9 z9Q!jz;S~3Ju`mmfXPl0I=PJSB$`~dBCo5OXgx(nF z=F1`fg21yVzz&>7!1?^3N&`s(6d(W?-n}76y2=VJ>5RwGzYlXz_Eu{$TB%cRW_rt? z6@2m(6}fshr!2s&!<;Eh&QOeXHb@Kb$omkx;hA5H4yZV(gqX5sn!O<8r6Zw=GHNoh z6cfbB&cVSLhp@mI8G2E#`^|)sASJ2s4g9Oes(K435=SzcF}Et4f-4RO3|BzC|Mc(3wv+EvptwXbhDDq3XtBd~ z7}mN_vzk&1cL%hw;q}u59-1Pf=ZHWIwragVQLQQfv8@IX(e-5qQe0x*g2m+!n=e$- z;2N*DbQ)zE_JlVdsPCQkKQjb|MEoAyyAlheM7j)}qqki^&IdE5DzxVa%eDY4aDeR# zMmKW~?@6w(AKFMFshDjPqkeKZH3QWfo5!Fx88@c`cL5EpxMq%ui)h8_`~wtOBGY-h znUkchb>TaJS)vn%3G){F%Ieso3csTfv-hT-v^LYt7ey{-91${Wsg%f;TgrK(p!J%h z6ol_}aBNm4WRegJ6iIU3%CpzinYw48q>Ct;qG`fju<~A)=E_NscU_q`Qwrm zI{9qco$eQW?;MhhCx`GNG3wq-eGTfQZ9=N-HuKC!`-Nq>D1EX9$!bMN1!kyC{tUa5 zf#-xYrfxZkd(SFoiR_^y-bQfjIaCwEt0e@e;V4631xkPOK9dq;f}~-IF;!-TXh2J0 zaA~SoT~4Xyf^YOb>iDjCOFBpnY(cctQLL{tJj+#E5EvM(H{&nE&Og2a&wvxtDq1=I zp-arbkNU+{1U({_lOk$)^Na(_#&?I1YdGL=ZnsI9zM88qN#LRYlj$+d&v-En+QS_u z?DyQNMSqO0ivljl3Gg{seTU;s=`1|QmXvYLcf(!+TNGhFaFMOd|I0#&(T_JsBx#f^ z#em_iLa%8?p{B9&gO+25!>*m=$_nD!rQE`V%ja?+H$KmCSdUyvFygQu4{Ad-{U;tt!T6r37Za1_gA z`4i?mAo$N7_sV&}kUxn39iD9sldk5LC?KavGs7k}tQ~$Dp2KH-@=E)o9%wGs?45ixPZzSR8@eV6gfvS}>munw%L1JQ^hC7iG)V;rT9$ zRxmeAd8Jqy^EVl-)F_nVo>-`^?2O=tzE(F%fiX*?R&71Y#jWqD|9T=`m$@LXp}5-jA{!hhTkT#af2~0Z60Gf63JMuxArvt}Fb62ra;cnwuVZHela= z{T(F)Gv(-kQOO2>N4Dmz5jebe*RLSA8fZ=Cv$J%;&!VWSa`SqGciQ2qfSlh1!&AUb zJECU$xBrBP)`eCU$`o3^MPIq)LEdTYF8}=PaM`X3^{-Fa_HQ@(hf3(k0crdn2aomo zrv!R}ps3z~565pw6JD*4dV1NdFmxc%1UavVK9%*3Z;auZ8G~(ph$FQOEtT0St!c)7 zC9bIT!Pl4aqBg$MOvi&|PC^UIRVaX%P+0vN`qp*Bvy|gVkzSuVC@K(UUncuhdF;xD zw8UJae{-*{0V7ka4feV2`BByxnZ!H~gJlff3Enk8F5MY3LG4*xLHKw)N@~h);q1qm z7;yTZln&Q_M=<`sBuhpv&i`p_=IYYC1yLPH`w_TNGLV%^y86{;uIF3poDS*~eJ_CK5Eb-S>ug`ABJvStx zejM0>|A|zaALG@BitR0xj}3gS|9jHHV4Wk}Yl=<3VFrv4GahBY-?|Iqt<(?fSn^oy z&D5>w_+`v(VTdKC((3eJI>oQw=({<~O-9}c>amIl`=Az+jAAUrC6ekZc^a^3wt<*~ zgyJ^srZMl-P&`E_l8l+i5Go|xb^XAZmd3wEZw19Y!@~*ZpWUH4HFY^l$qu`*Rl0Kk zLqMAKV?p^;EAh8XX^bp$yD?n+X~N|hO}^pUHo_H=R({iDInw=4qEQ=U95F>!h*z=Ck1Hny$v2psK-LtNj9VsOi32V^(HpHIK&>$B=H9SGVLp%5jvzTEP@iu$T0dK0 z%Iwpn>5vk1etbF+$GqA?CTUkj4Noktow@`<{pz#QgXjgRjgfiG{x)YtPI@M0?{ffU zM5475B~@*6#FMu0I

?H;owrHra%s7@_sSgh$0-_qm&TS7)Lu6{%s3H}r|STD3?^ zfRronFN;`UNOrcpu}soGSaS-B#0oeR<#1U*)DM%zjq|Wwr-}-$W}>t* zq-qqUe&unP?~3p~kv-ji&_hArf5R0iJNC4$gCp;f)9`Naq?qk=*~K@uJE&~#g9OUn zY26MycM)WaN}4~Z;2P!aLVjz!8d?;b3)PVe87zV- zpnjL*R?>i%hu#AGud%X z*P8!w%|h{$gQ5xoH$vMe20j`pI>dcvkI^4l;|J_lQo*3mG@btG)uy9+2RSeO%TQFC z^|f~_azQ0ks0|V6Jo=u`_edcabwQ$p`E1=cYm`e9I!Ei_-*}aUEA5?g21J_UbZ-tg zP3gug2d&ZDeoJJry^ESL7D&~QOLVyYif=JNriKpP5Uj7}2F08P~R+D@(@ z%NpPbumuJ_Lgt*;+3gfLaNvWJ-93j{V{|FstbrKla`G@?|3eB~$0!^qmk|O~QRUNf zUiY`F8-~WLpou5TNwg~D8%%nw<>!pUBF0U>1`$p55y!)#X{gE(TMqe>6%BaP$*+XD zT_r4txfg84;o+i@`eB#@cGl#NG|>@^cYWX#gRls4pB%JZlrdsZaxVC^#>?q|?VeBF z&IjE44|8&@Z|?&JgF40yefSaaIGA-%XQQc~-`xTO>sl|dD~ISRZM%z39q<)AYFFg5 z+WqnVM@(J48bL^c5z-Sea=$P%e&dfEq}Bx}hN#yZ>l|#nl2FFrJ1d}&_fM>4HjE@K zWn;a>JW!mU!{Q4~FK{xY%Qhrtrky^m`gpt%kx#Q0Q!PBEZS>M$G-Y9)UvWLlgfeBt!^eXwZC>JHhhA2(Gb|zIfxl9)V%K zV$(%+cU2}!sFA)>p(iG4(?BW(2-=g-34TG&Z;H_DlyYiy{ys7DL&%5~mET!ZIr$pI zzaD)QWgZo*0VTne1TuijHqLx8>ObDiRiBW8Y*ZnFl2m&%d-E`Pz5uHvHlxKj?rfYM zSm}#G2<&J#DB>l^m(Iw~Wu&PF16BGEi$e@SMGL@NL^v91AUZvkt!<4lsDr$ha1g>; z>^rsjiRW|bXTs4e3WYKd9ecT!L=JrvpIU|!5FoHpaWi`NCP1v5Q3#b13WFTOVhMm- zRZU?%G8pSFA&q>5S{XVLKmPmr%06eU_!gw0FvS9bqlK)}JRM6VH)!$97z0g~O&dl+VD zN6dxX3#`k*?x{b15GalaJgLd{2&17s`X{Ke)k~Nk5$y<}v4Mu6`u0e?8K02`6{O;m zL=H3HqoUZgzTUaw}|{aw+vSvv5V=QKL;b3rS6 zJbq}yPN11&1AKE0fyjYN3|@1EP~{`6V&X4HEQ+3Lb$h<{UJZ6bE6qjANwjwbelZCq{ZH@q;$(NXHU1*&W9*x<~K>=US=6TR>1!4Fo{S%T?H-O#C_+ijIsijDviabVq8;I*BdH><*8Mt|7!! zH$o6Cn>Z_e^wMX;E3hjQv%>ZbebogBzfF_D2RnXD%cttP7BQsP4bNZ<8UBm;z!#ec zH)D8f4b!Fi>m~IV3od-oi!3~p$%~9!aJUP_MCd40b9C=sS~gg;&jcZ+44C;&hg|y; z8bWlJgYHm)xA0k7lQ#Fp%ZN@>FtMKh$VX@uDlCWrri_v3I*2*$2~3XX#8XZlT4Iny zq?jJ!*hSn^U4W9N%dExfB=p4#RrV~z2%b}u@~sP03`~O(XS&ke2+RypV{Rp zDeKKC6jmpt2j@pEl?K+84l2y9~Udr;E1iJb%W2E1A%M!eh8&Ihko>ay$c_zhcfQA z@6B%C*vho_9WII+eK=^G=-XF|aSzLBJpD#)A1C*b4^QsQRWFAEyyfsA0CXy)T>Uv1)&ee}kL6O%EvGXftzkqXn=kVu;7Z0K5 zr;O76y?9Tul{;+B{r!F9dS*-Lr%m63(YLYl>+j1=osWgjR=z{$%=C|sOQ)aL!1(E zW&e~9S*wVPAnou9#`BaC;d?>RH0XJTIDOiijg=aQSY`iKFMDV3gCAX--4KvG1%D0< zYYW_$Yby%+!dz7io{H#UOVSk(Sge{ba^!D(f}b)Yb@;bbQ$;Dj!WiL-+4|%p#DW({ z1p_7z?3oI}9Og7PFKfp06ntq#X9l)dl=cW8<5@A(D~Em-c28*-DsG4W)xr zR%$iLkxWT`FO6m!zS4c@s$EPd7-7yE46u*BEy-$K^?>kW`+{fpwYp>fH={tmEW+OT z{oL*2kBg}eq>h0|7vGxa%DUXf)P(n96>|e8;10L%Y)*#*1#!M*ox-5|@o44x)^#yy z%YP2S5F%Ho@N}W?))dOC<0^UkcqkoVckUXs?CW7#?dQIl^Kaw5tLOGdgdIJ6e_At8 zx{A_0M4X2-If5Oxl(rG+#WP23oO}L`C)!0;d8}z=yX%x#k_S+I2tC&&@`V~c)otJ7n83>(g({hYz#3u+Y&t+R)%$gtE z92ar?I;iCgf8KYmjZ?bNw4+D%X3N9(Ylbx!b+Pf34hol9H2s9VIiS$G#0+E8oZL#I z+jzWxR67l!-vGqQf$wEJD)XL=W!zYlan#u(VQ|bwi2SF?(l($ZzH%oT$XC z{&l)f{|KCSt#5yAGRW=zAnKbTPif^S^LlFS=Dv~t*Z1!HeGb*8p3D0!!3r-~sFn;D zDaCE!#p97=UN~6InTxh$#~I9Bk;I3%W0yxWtYz&0v}ffIB$cK&7mU=d@#}QvtaHW8 zHAYw=(v*pMlac3o z*G!WPDrHXu{%~u8qYeWaC7kg>?&1Q z@7lp)<*isiwI(O4%rgZv81);A$8%mA!;8!B9Ud5q;{m%}V6fa{89lbNUR61@Ha>6& zoXR_JO!FjL;aY(5#WwY^#W(fU8Vjs@C$SImKB1r>RgQ(vmK-x{=NoCZ8;C0>geRp{Lb?*`~2P`*1u|!zeqn-EAVTRtFD3Liet_%@dcva!EzH{}< z-RtRH{Xz>XJ#*50Vv07sNte{A=zh(auQP_)V<+qR&7>%~NDuMU~9|P@@0+a`RBs@PLpXX5;1*WE62l|C)flzaHKYxGxxgim-sUfTTOyqi9!m}&;^+kNoPJ24oo(x#7w1&l2` z{pW6Pdx3)qdXm39kKwrF=Pekaiqb4hticZ8rK(^f`a6}*NL{Dq z`03KfEAcH~M~0REY)WeH91AN`ZHx2Nps)R0iP#d8y6M5fpAJ%aayy(v1fwzZv*iTbqe4{qOc|}3s?_MD{t=A=fMIwkJ8KM6D;<2GQOA%m9wAj0O&;Vls*x<3DS6Ws%1NsXw5?mk{38IdWby87XwzH&vS+YERe>KMa}OWJ4U;ah~s-X0OMe*wDIzCMk&K& zn;SoY<*zJ};@Fg;7(m#wPn^mEw(yXILpm^p(ZJrZ>TpZ8n3aD%0FBk&D~m=aK`Yef zSxEOn4@|F2f2a6HJI0^*w@0|3qm`;0C?N3n2H;YaMYra}R6b7Tn$FS#MnLud>gj(u zhed?%vn8Wk!x1Vm$AB9)uIp|2Nt@dc3bzh>WzOXBULX*o^!G{VviSr&?Nm853=tH1 zsX~D3oiMA`=gT9NlYre$lg&~BYtN{*|1tDcsNF!1L190}=O}***!NKuevNk7Ve;$V z^nhOC1L!{Apg5n0!`KDox~iwlnI^kZlx7yvQ8kU#zRK&TjLsV?3+p9OHZLcmoRLMr z;b%%g`$-9#Ny|x_rjvqw<_}W$kS0>nQP`nkkfEr_OZkQ-vEMCV+9CO1jIQlF7es;R zn=5?>NP(p5)Nkt!YEcV9=u?(wPvEsy>r6(4%ZN9E1eMpoW=sp|?bJ9IC6JKN-Y zj;5CL#fTY%+*k;n27+>wfZ%n_VeUWwby9cAfM{R8?#ZdNv0P~I2q_*Q^Y|2mv7L2t zpaHyWvr?nIaS&wPkHQXJ!5pZB-zjW+%sLro6rd)=Fyb=KXbuEkL39iw+~?L;-xi&V zyULTQ+cQvyQ2#a|bCwQ;(u@)fOJP`5XK;P?N3u zsMww46`hnz=db1MW;~(3-OfO-*)Bwcd>Br-><19c*QG$OuwY%TvuLPNpFrI;hfGl2 zDwemj`c&b*yW3W*_EAS4hsDW=YiEnM+#9j7x;;N75F$~s~k&(7(~tkuCTtb^6* z(%#kc%gq7w534M{@gPnvPtnUgF}_&cxMVT;Ap zA_3ZOiS;q(xU@TjmA;MImwIq|nSEgP$C*!FLdDHOd>q{^Ux<`QmvWb|qW!Lit8$2Z zqwgz57g%)en;_m2j8-3j-(x5>PneKtX#Lw zZ6e_~rPuFJd1QLr*{GPJ0>ZI{+x;KlK8j59|8#0F|EGdC8!O}g)Tye`)=J(OM)rTH zTTNH3S38Ds*;A}m9=k{WeMp^JGH52@)tfYn!X!BhRQ~i%)Wf&6zy|~Y>#Q*X%Y-{Q z_u{*i;jrWXbP9=PAW7wHvoy5vf>F^SX+~?DtKYSy`^oeE9p9*@p|_^@eR3s$|3#P_ zqk9-7?O3fUTzWPnja3N1XBavfA9i|t_%~cpNxQCJ+q`_MXyxkjN4Kx|tnI(*%fWf% z=T5g*h1wWyelm5@wLI73@Ly**-uitgTB^<#KD1WeEdjna3>?kZo zCg7LD_r_!@i8f-kT`J7_$DB*|#HB(GZL%Zy9(>C!n%cQ1)smz6jlN1RW}OW0=b+ar z9eSR6rP>Q{L;VEi&j2UK6<8F_wN^P9KD)VxPPYTjvyD>omQCB3?J>AW7|vPIFw};{ zTLjxvVg*tBGK{E3;t8Wwgz&$%FscRP#oKdWRSwbJ1+}%8Qc(4$qEKD`x>JRQcZPrU zdg#~1baIf21XOq|O-@`7UOj$ttwb>5ZDPJk*Ai<*^Qtn{OCWg0qRMMm@vJ2x$BIs- zt{q8XB|QTyFUO4)YV!5JUfG;#50eIMX`5M_g9@K^hN$BfY6MVc&QI4L_#cftflq3@EmMIqL1*gRoVxU zFo418TL#?*E=HjW@*1jh4{$+?;ymgYDe6*2z2XaL(uGZ%bA#QOsSRW+E-|e-4p*MK zBPmApuyVU2fiy^ogYr#vm;C6p6e?zmKQk<0!<7&qKbX2A(UgtB@X`^7&7 zChm+H>SZ#pB*kav!E2GVUP`Pq(IGi=Ui3#nw7@*s3~?naJTh+KCK}362;DCv{xM~e zusrx)K-!XgTGTq}&0oC{>&L|vGOpL~7}s--O!_*w7n^gcF63k?jz1eTMe(xWlp)@U z1mFid5}`v384*$z#kD9H0!r6RxhOa=^v4KJwg}0C9_mD!)C(WwR!hm#P;NxrlpQ*q zJPIgE1Hlj}mA^yKLnz)LGhPXWgVf@*$6 z${|rG-qj8lM6nr}qI9@q&?7X%KD$&KARfO91YUBDNAx-eS_m~vT3v2pd)@i7@Z^af z)?($Q4;(7}W&9)rA(`L5PHkNr$6KkN4U4*_Jm9CW^*|ts5PK)xeo`R5V(TZ@9P#Ze z`rxZEWmm}nEKHbWsyT_6(;H0+9v~+CRuk^^N4BlSabdwMn^&)JMlJUZ|FW}&;bpNV z)|caaZ1z=+>=k&#aii7K#nlI$# z43%q}TKZEwzpP{YHpl68V+VMe~%Sa zjfV`N{#K9xfL*x{fNmP;*-~zwCp6mxmus2Qx!F3Ambpv#OHjFtE7=eZF}&(EmB1ti z-Lh60gLss_LWpp~Zpq|3OUU)n9c?NpBrGt_PfUt-i$bce@`F9(v_!KoT8vSnG)B9Y zn@^3ClY@Vzmy(8ajDTt|b~3Aapm>2&zP?>6yL+yuIK0duSE9+jJTIPZ8oR|E8xfr#i7T5S?C3E~ZF5+=Y&BxlZS${?2>kwK#h7Oy)<(?i@C z%|rMb)#XV9fi?g(??B)o3r(LOyL~5gli0#O&dMv<4OogN(N4`c?kui)fm2U{2=&uL zON#GbOu(VQWP9j=onWHjci23f>#Y=(U%<#{Epwxp$5q@=lCgh_8f(4f;|Cr`vL5t* zH5DQtGOgPt*rbuFF-@Q!0!LWkm85&iviq*NS09N$Ec12WR7LQAXc-foEpRa=I(9rw z?g86rIV~rfkSycxjr<1oK%q*|e4+^B=W#UQD?TOuQ4t#D4iP7JC%MmSo0my2#>5V3 zw3kZRCfI_RF}jmj?#OlhJf^lZB!~}#z%!>29d%`vOskmmQGrAFmR36q)4KD?MbGs6 zd(fZq9;$Ddm_Vq*hI!8qH%3Kiy<*6U;Di{uZx1sgXS;T9QQM*8_fnoH&4l0Dg^T5{p+3G2K5`^4#;4O0I?*=M~XiFJ@xd!~!)BJY)kJd()o3F_3XGOdrVQykylFd4^7WW?0@fg9A0VfVI91XHHo{k0jzX^bXQkf8Yq%Y}78SC@w zHlPh2YZt8`$1YdPw(_k|+RE((=LDWv8TR82?H%QjNPn^-1Ho|wSDzwTRSprUX)?gA zc~Yh%UxIcKA+)7*w*}||lQVMgr((^ASgs_^ekfSKl$Zz?dX!JVGxOxgc8iymJ8o}X zyxJCGer~J+-RCx@)g5KsnS!H%C7-jZ{dG@uvM-pQ0Gr5s(+-FIf*Y4U->iq2tIC_n z1^q9oxk#?{^uH>sNpJfuqpZls@$dw8EDin@Fw>+pSUG96Ram{NQJuBn9y7C;8ra5V z#uWZ7?~a+2{KVl>2Y79mrANe5Z%z1~O=+HI&~yUpxGS>C({x?- zE3cyUv{SCa3$H1~=)toI%+IRjT^MuB@=KOb&)(|&hMMN>uz3aE+aGqN~-Dv%=y;%UQ=CV7gD!56K>u`t=-mJ z+ObWil%Z32CAhzIsOdfB1)>4s_SjL_Hcf*rR!kToYwIt~4&2sMRg`Xp&Bh^mc}(9# z(-p+rB~zS#sz8}5WIEj?#GYg(Y92RVu=dj-3kcu<{r(mV07fSd^Z_c^Oic!v1Pb^6 zXF0haDS$t|NN}fMW1w-IioyoZGnGV2eRFwOx2fqhtGQf$+8uPCUpPN_3N3ok$Exqa zQ0^4D0Z-kx;huV5O`iO*fZzQc56ae47)0Fa*<=zWYe&>Ro<@7-buhp7SF<%!_4HEZ z^GvRZrp&^BLG3%dL4vpw&ji%tRIeP9*|kFP8>A88+tv7POJk?!>-`+CdYh31IIP$GgoWZ6oSo6Jcoy(E2DX2 zl@c_~BZn7yTO}e6zx`HiJY8d?`k7-UMJ29P#2%+!Sp041)mFn6VXk_6!o2u=xmu!* z(pmKAJ<@bY%tg*M_z@7~-DYdWNLKf@3{5s@{HD_jx@lB5$X&tVTrq*TKHWGFZNfuF zko3>V>c?LN95VcEITDB)FPU|1^9}3|fpDDXhng~#rK{5qP+OdErvngBu(=ai+Y$(+ ztSETNdv|5*WV;0Zo8Rma=C&Q6Z0wRP4E@2B(ezrXM>IOk!lPZg~cO-CRa7Bh_w>6~6bbgl!77`^G>}WLVJ7%RHXG)F6BqA0;a?&aIiXe} z8PMgxCM@T%dd_w8D*f}h{1BE?1S_Pa5;P8n zah@GPLc!&X^fU>n>tflzKVaan4js>g|OhhfW_g`QDt~Iqq-4<)+7z1#?+MRNrYevM4B<(U+UgDzLpIpt|;Q@oZ4SA zd-nd_AR!;Oh|M|%A&2d`DfnPr>(JRsMiDEl9w_l{B*SO2$wC|a%+391nhD$`2^&5# z%4|*5{0`CWGb}k2ti!ktRb2T-)^h@Ft`<3d#tfuTC=Fg9hbhVrRgMn45#V$O6nQ}U8a<>te)ih|FK)ZL+N|b? za9ert=SvFiQMEltVVp*u@_#%iV^l&@YdtHkt^f+@9)|xZsQhUPPv|Rtf-w3tYi##~ zzNEkjH<$r~t+?-tcUls}-CZKifN%nl+&hlIsRMOXOWUoQrccSh*hIHOPLtMrUnTy1 zM=}CI(PU^O@VJdQ24}hR44iM@!MB8!QQ+5JBxpwFsx}q>82vbjCaFi5kcu>PXaV-V zKdwXz7KWWLZrkru5v}n&M&tNt@!{)bN1z5>H6HU((gyK|@~O)FU>X&B5n?#V9)k;> zbm%JLpxzN3{;tmXWc|E$zEi&-1b{AR&ukghtFCExpubl{NMQD?LA&`on+6uLT>me3 z#!itmV&J>EER==--t=W3i1%Ww9~>oqM|R>}k=CLBkf&knf!eUx!|#W)S{B7ptk?7E znh(TSK@))Gh6e21NG?c>swAkXBNDE)Hv2E0YgaFj?!!76kwAQIfAr?^mWs=14N!JX zvnC1FfL)tQx}}HLId_F_3?NzU$^faM&bollqfR7lh!CzBiO;kj`C`FI=iir&T9uc$ zRHx%m8ma*0Q@&OF03MSfEd*d<01|k34=7HzrotG(|K-{|i<>t4b-zF(dLpY}N`y0f)RL=J|-o#wW76Wmy{h z6sP(`qb@>1!5L{oWA3vycAP82(keCR-y`w4e3s;KdzsE z*a-N3*8<&S{iA+i-_GUz^8 zyjK&To7T*##(|u*y-IJ+K)}&u!i`xi1L3AI$|p$k>~tx3yT%$Db&(86<7BDd-Mad3 z@BBCSzpnql0E&X3gz46#j>Xk;8#rNkmssn$9&d=pkn@}9f zXpdDV3<|TrQ8TKnqpa_^4kj0jKTn<2`?IzwJ4njF1I3wt&W|P_jCi1+ghCf-ed;JL znEy)3QsJL(COxR9)O&3%sm81q$ve4dTHB7YRfaY!Y=mTS`MXiAOuP|hM8u^n%7{%m z;lq-qvsU_g$%F%Qpfe)Tn#O>^7?Sh`;naiWy#t}p_6ND*Ai*YrK4;juP~DW;Qb`{X z58WAyqZAD|8w|AUmT3X9hS(>iSOC1g_ZUgi($_$lu_Sn}vs!VuDU+<&&_;ioHf*b==8YpIOpXIVMdto<{5jYQmg$j!Q7u zdk!Mx7ni)a?v0mbJoJiG;9<(i3iPsunO=rtdY1{QppNuS;U+^b63dGcOXUyB-ws^; z)&I6zKPQ06jF;m+Z0Nvy1Uj?bqN-mqlgPAhjOHv91`-B$jy=S}OYVifkAVddJhK=; zj5W!{T6Rb?ETG3sF7Mk-Wj8z{YN}%dCi#ZS2HtRR8pQ*dQtEz4Qe)#UPn>+}<1Z+G z!Z_Z`QRZ~^TR}=tE_%S@knzEh%9%0nSoRKmS&SmexK;F=>kH)@j2p+iOh?r zmVk2n;euTJ2HUY@2j7G}gXD|vEvmuG_TZqHVE7EpY3uRg0H2uL@R|@^zxRGVcblgN zQs?0eNl8D{r{o>eS^3}22Or}4u!jb}-R{8Sz7LTu-2n3212l-sdsZG#!h$F^Mbk|A zL&RH*ftUI5eDN);iLvu)uOI=gZ->6$pS<@`;PI2YvptXrfOZIQF+wH0*X{bHt;S7JwB%da)B(Z)Lj zA%F-&PZ>pL#~&K3GOG`khghl6F4|I#K?o2;g%DJfRRvdxYoFND05h%W=T&Iava8BsV zh3el&;{q71cZgaKUj$kHn_GFhM>>GDgVo*&WBkS6;VKmE)1l^o=y+FpD@o2V!E*fy z7(?Zjjwa<4vX~gy%v6Q$&Y!~_0q2qI|D<5p{)5lO%>GZ;{O{-go_7BG4`lq$=lZ@j zC}a?$+!ejS*@58sa2y)&6ia&}#N?Y-7-t+mhI@8UoWFDeiAdjwPi{LdHBJ`xaJyQC%?hB#kasOjj>3R=X2t-Dm- zzlC;KjnUeAuf7e=Ir8HuvWoH{_G=mps5dT*X1#v%g_B8D!tm0*0R7T`*R?CF*G#_Y zLj8}_{?#Mxe9l!JwX4^UN0>E=WxsgKrV#ixE*2JzEPCVQQpGIlpu#Wl@AY@X>JN7_ zhgrMLjwV7|g*fUwQ`OiSE7U1uS}T7?k}@9zdsBubl?Cw>B=DyTYP`4APFjHqe>*f^ z&l#ZY6%bc3&9C6>e^KyNEL?=+Gp^dE&&eQI#86TSIOw# zC&mxvSVFIgDXxgwehtX*V7aB+Vzb~Or;PrzQt>4cU1+tsbd>Y5jr`9{Ppm*QXM)_^ z%&3nVA>+^kVciT&T}QE#6GNcwF)=;&^Lh-PL6-zN_|{4GI-R?692bSit&CfS1fdiu z(?XYJYLD_5KS%!hOe(kj>>;n(L5Vz)-&d`x$jS*t@KW9Rv!3HEL#_Ep7pbuZIR~3$ z(we(U`b*n2y~rscuP7giOAquQ1O02R`>!32PZtZw&~GQG>wYD`lZ#(lbuRRE`F8Ms znVWw+d||XS%5|YLM7SB=N%-aMVbM0Q%Q~DUos9T&=Lj-~d|w6SW!2!)?Bc)(T(^O6 z%=!-q%bA}_^T%_tXZ6Loq+L!Jy!2kE9B&K{pZ{3PeSwyzg6w|q{)CjpB4=Y9jFjGfEj&SmCaA*9vc$QQ=OLcK-u z76Mp&Mp}4OebQo?UnqGizl-dXlPrEyl+`EtIOT1w2wA2k))})uc)y+qwd9=NpJ#hZ zUmahCcGplc>WG>yK7)&$YEd*;jO*UYuG@Gc;nz1rlWD)f&EJii@ggA{!~Fexus~y# z49^%op3^Aq!9=A3^IyR)tTxRwfMrL1 ze$7(A_POQN-9p~5p&25fJ>8eRbShhCEfwv|)NC2mXXb9>qnGFNn25R}Qr}#k(@dJz zn1Daigc15r?noKFr{1;osBhf+^_7%GS88CbZ;d`V7aRx)L-o=0QzEN(7(91TKa_Er z_S^_Kh2KmD(q43ZeJ>wO)^OkTa^-Phm536xLEf!Ao*#>*<}>8Drq2p=uAU9iku5YU zbM_UeN_T9E*4%q5ent0qLP#taN3Qt$fhiLmEOx>zrS&b(nvd4kpW5-gq2+`l&gn@X z1=Fx-n*61g82NYze}}a}_5Jb*iPz#nvhU)o$oU2Hc+ob|#+=jCnTvM~PRBfT%ruoL z^cu>KUvXr$%L?bT>9S?#?^pETomfl`7t`8DlOPKiiXl+LiQJ1qS%-@ zulOYLnzt6jOA}IPe}Hd~NK%T|TgR@farIdWXsXUQ$Q_^_AN|lx@}MeqJR!ylRU+L; zFmlz$P2?QrE-?GiAO7x8{mF^ZKwyZCV+%oFb{7U0kNU^>W!;+(&c)o^;hqt0`H&u- zlPLMrq&8+6Pa?$%!Gx4#G(Xxx^{*2pReOdUzI>hq^K(<#JbBX(?q}aheM>z)-Ba!UDX~ubw%(u(6L!Q> zeJ=w?;o&>^d%??(uJ6#ac6%Sd2&j_y=|dFTaZvabo@0*_3wXHw z1D}+nhVL`c7fq|O_eJsbYMpUscA4TIx%GEur6g>fOE4`okAI*o6}g8e(|%px)Ws+< z{S$Y4TV;mrt5cQnp4FJ=+2rl&zb0K)(=k&Nas?cc*c4U}NrFF}gi z@umYO6ecX&jQ1PJ>EOK<=6x!(7uZL08C7=}&SR_dXY$YXCGzQg;TEZ*Cl1Gw+d;XV zKIab&N}s&=Q#v4w`!{l&9K{w1d30Q*2qq^r|C$BU3Rp?)~$Hh?hv?b~-DXR&q8BCR`%Cqnu61}<0%}TqlkvP%b=*F?^ zSu#^r>hF>L;#dWbDCJN0dXJjqEjYC?wkx6O3Ee1*P#o>8pU4;VYcpM|mCX0^Kd*h< zbp>fg*P0VqlH;Ek^Av!dXgI9}+T6F+<5bOw?YqnSKBl14v$hX`_vNdfBDkS7-j1nOwXJnoMEH@sRu`%13k$pf zPDDi>{=$ynt(7&~hg#%IVp)nqFC&@keq|*6mIn9{ zW@`FF3`R9ELb=iH&O(=yCq1!IlW_R$QYK^utwy-wJR#q81#95xkPkc?>(2xagHje8((Q7=rJ zxh!Q$zyQMfGCD4?etVT+<#~)J`S95jvM{>+#oq0x>sxX6o%j9?ri$?Yi=&-a?0*bX zM*)~R4cxb}11Z1MeZG;8<<#O!XLMftBcEpa7>9#)H3U)Ipp$K4qCe@84w<9w(**e= zJZPTYCcvp7lP3yqwaG}yzPK38eFoA+a>}P{!j3lF(IuMl<(!FRLuAF^hEZYzo-t|U z*^~-CM|V=_f9_nXH20@E#`;-Ccg>bm^>5T14v5G}dQ%#VE`{_Q&ty~ATQ=tUB+tat zkY6G?%8zI=EU(`y3gcYTZ#vCPe@Hrw8U9Rz>K)M8eB0P^o7*ycUlVg)C6e;S<)_d@ zH%=pM>B4h|GX)y|Ehtl~wvS-akxRRs#H!}C#NDULT4$XKbw|aBLs>sHVLt}l(%3lj zR8C*y(4hK&_!RXa`^mVfveUG&=FHp7lS{if+S0W)SCfY$sayR^g?e<*_fb_mm8I+F zJI$Ywi$}im?*lh8i=Rhr5#^>f{1|Op@!p<)HKjv0m>tO~?w(1QPfaI>d*C_EFIXBP zvlkutGh`OIz7$bLZhNmMv}E3Qk|1a8FSELcn?kzkU;Hwe^3zxTOBvN0xYJb}6`=Zw z#GO~f9cR)<-lx$Du%=_PSo5emrIeHBTh%fWqjZu-CjP-4+3Q4-21`Ap_v*1jNrB@$$Ni)7u>$D-ZES_=V-Ep(^3#NOp)hT+EAz!l`dyh_}FOmzRmGy2T zQr;A?Kxhxe*s-EnIBeMI35@vZKS+S~$@fGORG zw|;pZLn*Aavtn$GqHF@&gQWf914M#Pu$(XH*L~m2Lu|kuJD#n~o!!}kwBABG*xPK| z^B_OwxH~rW9r&I+(KvO}c(W6ewqykFo9D{4bo^Kr@e{io_MxS@k)o1V^Z`}uurNueYOkYL`G@*!qbEJ% z+4mms_B~j8Ai&!5TwbdPca>Y+<0($fLMq!T!Hz#S^lge`vuo?ECnaW;HDo)@UlI8= zWIpiJ*4woaoA;`TK18SP(GdwXzWYYOYkX_}wBBjgs_YYsk@c(Wuirppi5>nt_hsLC z8jw6}qGdPvn_gU_MtILcNmBM%n^i{^3;#C_f{%Ci{in1Q>)GB@?%EdA?w5_slMdC0 ze|q}LE-3(t*!Yy>)LAF@Lx!UMX;!h63gN>vx*|!gqu7mj?;eNtwXo`V6*92|Xi$Cq zoesGl33|CLDG%Y|IcHfNzE;4-v=O3GQcecntj7}bbMO6*pAv{#^{6v3G8KHD%Ywwc zwR{Vs{*JIZenJ0qDXdqqOy<5Qc+)NV79#C^O>6Z>6$Li_)%#_(@7@sfM8Ab`z#g0= zlfMi7)Qpcvdh9UqW?0UqtnFh~HVF;ucDa?UnS$yrvp6Pjv0t*6sMfchq)?kZDOy|Q z%njqnG`4K*>#IyW__dY>9<=f3kKPorVD-7+Cfz|N^N(^Z+B}W>cy{I~BH9!FIjy3^ zIg!1W?(`P7bM{4M%9d4r!L;xgJmLVuG2)GkkkNzAQg(}xfDaL}kp=J3uxIyN|4?Pj zDRDfWYWtGX&QgO5wUX+3w3@`S)6!!0{j3S)i8RDr<)^VDvj0--u<>A&$8$`&qf5=> z=x&oM=ah~SAMv_J4QH+q7HSc5pwS^p4?;Sp?TkG8Vt2>XrFC1~{{_$fHYKPHN2x47 zl(X>7*1iM1ON*$rIjokFY|>Jg$@S|V`Sn8(&l{dn1mI?l*zPH0MVA{HEgT#HyhjWJMer1Jv{S$mJf zNrIj%xE}^UT(1j4o0wV%vm&(L-uD0Q@YJUD!7pt`NuMm^wa(UeCB4IPvZM-G%uJoL z&y-mkN#=!huyzC+%_QjG_`6-VpY#Rm1y3_lf}yg1-RHDZ$=tXly6!~l-%_$UV((@7 z5=wyc$8iOD(kZDG5a~8)#7li(hPMo()_zz%ilLJm)&Mj5HoA#BTF$Vn*7)(~8a434 z{)%n+P6~cTtzqq5rsjQ;pL?liQrHfmao)FzVr0grEQqg*3wJAHIduBVx8#L7Q9Ma$YFm|7A8ML?oo`^>*;Zm;e0eG#@xt{f|(orzL9(T(4^0`_;h$O(N;W98+Yc==Hzw>yqa zA7t~W^Di-*>l{lR7&t7gK9TG(=R$))mRk6+!0>Vu_rMBzT2KS^B#H8AY`^kr*@_2k z;eW~FP$vX4JqnU9@^Vd5`fG}`8TMC<&=v=q240==!cVuPSjb(+!mvrB)#pN^$1aN6 zdy35KB6kXXNtE>Vd&<*Kn_9H~-eZcEajtrOuW7SSs!&G!FJ<8onhWf_|xA>_|Jkr&Ttb?01dA(@;*4yDs zP$4x`UQYc^9tQyqn>>+C--Ga0ISS2lOPL;=LULtNx~-z98p>D76m3h(KVPV?E5$Ns zW9%>+D3yWv)s;o!bITv@mR`$&|oPOr##(-UF+diO%K9t^Qr}LNB!g7*~^r%m~P%}eV6Jc6O)5Frgkyq zUm=pIzG~1=Z8ws57K%w_`B3dgj#Tj<@3fHRNera&QukQx=yMVbSM$={w#$DzM3Kjq zYE%_GAuQX7HqR^-jx5H!uw~EgdR{{!5*`04`}ibdWUAh-7^SOqf9{E`f|vK#=XRp6 z9g5xupNb4*ZNRhFBM~a_`WgTgI21TN6i8BbAeI-(6CJgl{R5@5_HZ#ADv)1)UVF8G zjk0-+Nf3P@V(}U8wzGuDC78}RKx}z{%YOVYFb;oM#_+4jUMfMI#RAb?`($Q0^p(rY z9C_;+wWyCmz^2uPEfd2M;T|?>LRFmB6aV4V0>aL=B<93I4Go@kR!SQI6Fx0*qPMcb zni6B#^MsEU6aR{)Y?V@G8RQy*ZY40>RiJ$DR*n}PC{^aylEGFU{p|Lyww$1+c(UB7 z!>=>HCJ>H&t#~B5NJD(lR$1~qQfJUo9RF31vK8M%>dD5#hx{+hHCP{2KbVcP&}$Ck zQn1GS9_udW@sn?*P|waQY2X{j28#QPypSY0A3g{qJ0^T!H9)l$_)O!WP_JS2X=Z+&2#AmOz{YZ3{)tFp;$g6gDE zjJu05-Rj4N&ZU8u48@3T@}r?7DLT;<66^kog1x1Db!h$$xAFHaUuM|0gAbD?!joI7 z`<5{&@u}Fzgpz_I=a0uf9w?!|XFkZkG|nVcpPJ4S zkvBNZ<)or5jCV3RNS-*7*J<6)2{Avq8*@wEJ#b&~xC3h8!uOCqhmP*}`(~U`?o8L- z)5=c=lU9Gq1!%5@M9IpaJ}=wHYVDC#XPyp1#`EfX7A!4HKIbJG2XdCaTV^bojIkvR zIon~*W2Uu-+O@xirM_2MfAFKgRTa1W!$f4tVi}P)(8ugKAUU;p@`aOXs|M%z-IHKx zZ^<_j34e_Va;o-kOR!b+$GrVoM;a=y08jfwup^uM=F~lZG}Hz4!1}|?Sc|%Uy0V+6 z?e6#OJL79lBcFThQVlI>^=@nS2cO8Sk|(@+V;azo)v(mIk+Uj(Y|gx21%Jdv|H0tC zCYPG!r;P_&WnXu$L#2>OFVw$4Y6J9#&;f8+9B6)p+ zw=Igw?UoBCDEh@iB=?cyXe~u@T^~1V_l#MZfcsfOvsJ3wIY&|NfEp*UT%7ic8qdtJ zx)$(PC4`mlgImaFf$p`UrzHh8@4gTf&>=m7A}W7)d=iW1q7m_u?yFG+ArpL54W@tD zo%7Cabqu68s7l8y=n;qSiEXhU<%xv83E9{n=1>JJS*Dp4C3^?EgTdB=9~)mu3e@4U z_J7kVkWxKQoUXRBepl*i@j|5C@)s+!jjaN&_ORG{pJ%VEKN~WCd&Vo_JaS}ry*lkR zb@E4`qiDd}IkhVd!5uYetp0ZraqzY~=gEu-TWm*`o3t%MfnlX>GdA<1NBX#s)a?dG zHE-yHfw!HQYu?xEetV7I@vA`DF%ci5UE`Ll{FEVj0+5cEg(uY2nh~VzWNu?k00P@tdD=3H3IpdrDOuau*m^Pw3h)zvWbHgXG_Bp` zT%3THxviZ&8AS;|axRW8?mDjKme!0$AZ2GOYhPWiAsj=6NE)D9ctmSE{D8u51>okdoc%kcS z+$2V&NXFYe==7Ec?{+07{-~I&uJs^j%tR?I@q!uuZy(#p{O4OutrTIW8jQFlwBNPy zU<-th(;Sy2;F}qj;63Q(|Dn^(_wq|7jQm_XC30=*^fx8)?|a$#C}hk(8G!#{^Z#GI z)8EN{(?KWB=lz>ll*>s_CIU{*H}ba!HFJ?x0&iGlAPQ0&qLh3LIorb|heN87v!Jf{ zpJ(SQ-B`BoW#NfeqT-O0wprBXK6rGlfH!dP5p|Q_ksd=+`&N`x1bj9?4Sm^N5>I=T z6*Z43Y_-n&B-E}4=i9=z4xuiI8QA!O*jdols>ZlNBK9$ZAy1^PX@0_m^#jFhKW>zU zNEiBM1|5l?%{zQWinkZb+N;(Pndj=>gXAxg$2c7`P^sZ{Q#1`ebV>wfN9>^hQfiaQzy ze{Esd=o8^(G+)dBDP^Fe?i?)km;0f$@EP&yhJNU55&Jv zRr{@xIQa7hjmLrFtTnF!^MjBj?cz-ElB# zhqi+M2#pnufS)1}V_Ov4h+hqQn_OU!$<(pn*KGK^6KU&HsUW0W*R|j<4(PX{VCm+ zQvMnay8Rs1=AwPQ{W&Z#xr*qxzs|zT366hyhz$LOA#YU1@qS7vYf>#<3yYx5cQ_BM zPZ~iPwo)J$h3)>DUds_JOnr*c1V@+BVFum!A@&lrog(s}KJA(S)0h@`;f;Sb z&O?zmxHR-WUgT1RC__g#Nq#Lf4Gbl9TP2({hry!*l%eZFcJojgc|ViEvuYHs@oCop z7}K3h1kV)@&Cu5Y$``nt2l_?BFJSpAUO(hc%{9Z`fDz|9zt6p!haSMt8(d$L4HF1r z4Y@+=EoL>y#%o%~Bp0||K!yfY3zg+@sVSIE#htPzYC=$Ci(Dd*#c;WAY)Ri98{&Cc zqg5-+C?P3Wf&i=^>}1GRdz}LjrU_O+S`{zsCFO%UM?T0VL=xBWgx0(7vCi5D!x6Af zXs3s!s4Nn1Fn^k2UzUAq={Hy(N#^^}W1u!Wn`EP_Ik}Hb85YwFzVv|0hJzWr)=s&_ zNK*O|u*Tn>Ls!3=X-GGKit zx8PzW`Y~DFKc_*lKJPo?sI4TcKG1AxRJ-32xuETB;b+D9mr~b z*eOi{{5j|;nmnV`9204Z8_v4lcy8~v{u?zCV6p~~DI9H#TR_Jyc}{?MK?-Tz)03pT zkJbLWa5~E6OgDY(bF@+HR8InVd@}lLq4?8&q#-V9CIOfrnb936U)VOANOU4tlj!JU zYlN5UZ2ELEXkw=0l%^6+_#E==V1IC9zWk_(0rG|T^PUh5O{Hw>xRhzpC{?za2L zgj87EbVGlL-{0@eHdodS758PRr$8kyotePY#!z9O;or%_&%7|w!@S=?tD}3TG~w{v zu<$_SA2C6mwb4?!zaO(71=Q_IwxDzaywX_=m7sPgrnQ%|wn&(__ksf%G8WbQgaedG zYgnE8V~hz?yQze{I0m+`LxO21z5VURm<|Q8>`0go>ypC*r2P~!Ix*Rws3eW!XoDmW zEC+~CqpLk3WylOlluX;y&x^!3EN6#n`Q^i#Nf3Z?cR`BB-q9=yS<7LvhM0GZ&O4kU z^=FaXH6-NPrYc^aW|s0dxiGx!u$Vu|8bsBqNCFN}lDLvf=kOspY=j1Rf7c=D5BNk( zi@=KnB3rns>iUTZ`F*w9PtlqQ%?c0sO%69izUl~FbXY_7sc#8sn-YnQtBVP)LpiXd z8NZ`IFzl+COfdBrGXK0gm7sQR(AMrBP^gA!<_cwM})7CF7cz^`q(4jbLs0aU02tZr|kt7Xp*^mGtt zc(pjN>^}1QcJZF@o2Tf}_>j$Qs6rQ)7D@^#_=ggCc8G?F<;U-<3{~};A}~UBkly5! zCmirG+UeaufLM-J_P|Y`?B>$Jg} zSJEOBK`3!gC%xD!%PH$P$|dl~-X^?3v9&?o8#xIrJSUh(>xcJ+++z0Os(GGx8i~E~ zx^4~c(*LlO90!7!9?{DBij8KUr^nRJbh(A_BQ1IvzEocwXO1Qx$a^0=p+gwL!d6_V zyOYsWXOqN%>w=XvFX1cCNGyygx>k!Exj>SR`Hi|-) z_V&+aIOjD@VGCTTpxf|R@!HwhteKMFSVTLDETAMkAZ)M5y*7HW8|zMnb|&}EkwCI- zI;?LwaO~wz=k53q7kU>DoYP>!;IXtkcag6UlN0r)%!<+iS-5)SQdJrD5(EZ6}SL)VeR9m9`jEZ|QB@#nEnCzTzHgjGyQeq{7UU2Y3 z>X#yeq0b39;NQi4O^bgHmxNMkEjr{Nzi0(~gbIzcoQ%20II|*G$RVfp-Aw7=2M{De zus-R6!WyR5hvGzn#Nd_Uz6YOuxME&(?Fp$th{RaXUWT|qri;$`ErgJSgR01#P3<}G zEs%MQ_Fx$N(%aMSTYleox;GI}Qu``vP%~~RJ z#N8Q{I7%ejx`P-Cw(QCe2S7MB$pgKj+I6GI z5f>h!X~Mvoz5{!XyQtdU{UZE884B?21Befl#72Ij9To%tSl(TnUicMs|w2O zJ0-V16`!jx#tVl_1v*Y?+%~%)pl$lT*1f)0wrD+s6-W5D?dbE^I0EKLAo2?zOenop zO9=`*Q&$`Zo)sazHNpWQ`%4+_w~zLHNe0MI!@kj5FuMMpjxGJtRrmte2&$e!5joov zcw&!0(muOU{BIJ*q)sEpOVTw&{v=leV;VGb5e8y{XR(z8mC}|NL22nDL6Q)L>xHO^ z)uzCUebm%5OWG!5dSYUa17+AZ?x~2;5?Cy7&}0rA6m)r50ABH?@H$o5bST^{M*o@x z6u&+lLLJ!-qz4JEG?~sw@G%5kFC_+G(F5L8-b`rn0-_Au?j=>1E7#S{LhN?j(?j%; z@7HD&Hg%TD3)5JA8#0NMYE19VNQpl)EIPMu?9K*Ij27-Lz=Ql%hmp)6>_bSO*e^(u zwh~nSN62-)9Kgc{xOFL#Mg#^RgJl(fH>%yJz27(3Wf4F{!OlIclt^QMrHjRz%#E^L zH?|X{_9E8+>d!IAe}4l7?zX#?>V?<1OAd*Kbi%x5ag2QTBR9BCg{@)erS3h;N;0&k zuf$ImBVldR9kofnuQ7&ncaxKrxELVl#iI0p=aq+}f%=L1baxS$q~X(}(Vq{dsvbel z#KpW?m7sOPb{$%dDqbXOzbSViJZNKijP(L~JkR7AUD1g7^TjZ@83bI#u0EIR{I>Gh z(mopFTe~#{UX{tA@%ps>8+;d1gMMsUcW*0Zm;a9j$R_i8m%-F}wL9pvV%2O3J(4%G zdf2sD_GSM!Fz)t2=mP1Wvl@op-lnrSNT#XOa}k?K?>N@|D%~S%Gw*AC&fB@?(%{LcMX8O&+3RZSqvxq-k&6+Icfq_5*wzS!!|SM+@Kz9!OfU z@VSvuKib&t+8e=e6Jf+$9B#g&1Jn z8ylyUpbHMLUC0U>`~ zexw4LtMw@SeO&l+F<*?mvo?cihL^2j#e29XlE*=-fx-LOQMK)arc49W?E^=Q_2Uvj z;$hIe{bISLNOl-{b@N!cE}qNk+P&(4tN0Sar3aBhq^6FCu{=6c`H^lAfRO#r!#pz**gjt=ByAYdB<73q`!^^@C>K%L(5^SR#_r&nGl z7p*JsUJ!j?p#KFLShf4^aJfx;D9jq|P;>1daNp@&V>~UxQK0!wlYb4i?eb)<4EpD8w`^j?VYx?~b(jj3ai{vWV9a+x8>X;mbHw#jwoc~=Yf;5kZVFc}2*OZb3sFOoY^ zu!hBSqV_H>ynkb^!imur3%^-Xr5~(%eF+hP8z$V}b1w`*)7}CyfrUJ0##iLOuX05v z%h5S@IL>pWz}e5aYy`re==yOL3@>FH5nUl-Qw2ZWxCDS(NQ8>^xGIktgBLbg!f%*IzClxa@Go~mdg2$*Q4hQYuSEb8eNz_&l;Pr;A?GQ< zVwKB~tEq{XtAsQe;fRPZ8Yc+RQ2)FlmM9cD6;^Mj7` z%&O_H&l*w|`VC9aE%T?igor-8dx=XLa{eu;MgWHY0nU$04gVs3YCQ7}o(|JH=kgJ0 z1;l!g!frP!0Kw8|LlLQ)^Iif3$cp`{8HmBm10(uJyta@TIE3;NtL>!S9~F*CN`h}+ zC2qVOPy2MqF@ZIV47hh-NTF@|5Rlhngb=Y_hMi;Zj>`wSmQ(T`hfFz9(B9gpCLpm7 zFiDy4*50y^<0_TyO_jOi>-*tx+dM9M1M*|DbvZ8!rCa2{KrXOz>P#QBAkE+ zPpfMJ>$}#hx~3WiE&(L<7vR|*?0FcH^gz6*RXmy!IfPZAEDWf_aC#{9Zk7|hugA2V zk{6^GCnCKOBdMNrJ%Af9a<%wiZy+P&(~W!K#z1I1uXjCQg^BArXuszHfQO107(ak+ zwoy0L6A-~5nEV0jVinneww-ng4=n$%sz~0=ooILGs!ODDt}!(SwBDVpl2Dot2>4w1 z)IMy$J{i5e+BADMOPV$dMlG37FKqh0MJn|Z0;{U%7#w(ZP!LqyH`fR~-s^_AWPX)7 zQ{jV+N}X~h`aW#@N3Qw=q5Hrbo8I50V5~=@3?TD_QX$_3z`SWV=i-7)1@vh2bj)3g z@43jwg3kT|4auv-ZlCcHja z$pm1zZ}8X-xy%ANAj?=@l?Vv-a2)mK1gxJatP~D_KLRzw%yGbuy1(2Vo(v}Hok}7E zB<+pcruZtLD*e7@=!PK(fQjh=LSKfXk`;--D#$+RL%?{vlT#LU>GMlb2qq-&)e~r= zi@0zf&#Y#OZk^B9#Vz^E6mCvtLAr!0^19f60vLd{(p*J&)QGW=^YY$Pq5Rd>nPu2l z_)^#tKwlIn8&SLhI4ie@?(F6ZFHIo&PcH-@#eo4{^eljn2uL+Jv*#=TfeL8r``4f6 zA%KOscPefm_~v2&AOb+oW!`xQf?2N?XQ$t81$1BX6C}aJ4S0@e z1Gj#>>FZD)&p*hG8f#c}O0Y1R5;Ap}7?ya0o4>)WVUSo<6NRkF>2DMX?B3qS-ITWsRr+|$T{Al|CF@fIU`UEy2 z+CikIjcl$t0Ni*2DdvSZg_{G`c01eA3j=zfx^9$#m|Qqu>B4-41PgZ?9|BD_2UN|X z&GHa^X(Q)`WI#bmNU;K>1wrFpyT9Z$nRh#x=w=+Th6;!1?#(=9xt$wMRveS}ZF1wf zm{zepY~UjX>i&b``~ybG`zds7)+*zG@&EznzVBtI{>)SzdErGTWQXyB1EP=~slR|U zb_WEZ?GR=v5tt@`Bhh!%9dEJ9LA%og4P>D}_U-)i#8}t00jgO~OzCF*QlGmopcere zQ$mz(hAjbZ{SFjPD+_gUC)vO052ng~pLNOWFVolLH#c)y7ou-^POE*B=8o~|bI?)8 zHT8yoTM%TPKy|H$=w^pBISW&uIbo!F7Y9sydWED%Rq>|o5O@+CL^77-e zh1D9q4iKuK)fYj-h6aYncJDt*_j>W@AFZ3%ljScbA<1N?E^EdETHNEts$ekf#eZ*ag&DEPz{sa|`Gf)AbV><;X zz+?(J85OM4;<8}`=Kh$JjU2#Koa50wPhVa2C-AXe)9*T{RDie2*diBIV1@U2U`B3O zI00V8RDoagy;?sW7DoNH&J7sWZQ$4Zgi2D8s-t1u-fW|~v*q(SF z6Xqpg-8%e0awDqIWP5%+OsDntfd{n6zPMi;+=$Yj;bnlI9(YoQbZkX^PyYsj5Gm5q z{PTzfNuzGXm51n}4W)$sGgqzAtn`h-ViE98z?8YaUW_a|$!P&of}WW;gsHLv^JC#` zg9e$07~=#Z02^2CkUR}#xPHePs2_v8@SP|?kKRoBkV9*Kto`a}@lC_rfWHsBZuq30 z9Lcs;oQ6_ISugZEm_S(KKrpo#Fc-WMxS9Z%-)!r}9;f0MagKtfn_~NbdugWsUI&6M)0LJqdOj3%@5b&e? zGc*;*EkwFkOyfe9e;Pt22 z3FF6IfOjK+=pe$Ts|g>pjR7jGVBhvC*0h^}Ii=PECWH@L?~C7*FFX}rKVfXU)y|x(lbHf^PC9-wmqgq@St9NtjgJU^`9W;u?so^Tv_UM5PCo03FNdQ zkgHe`lB|Xpa~cl&r_aYFf|&XZD#T6D6R?h*O~3soq4?ZLY;`pFpZ2h4;!C3u|NOjn zDvlm~7sUoZvEQ(77;%(*0x*lRk@Qfy_5C+zuWsU}h5{vf`|baPQpFdWH+g-rr4#u# zY#>4q)!J$UMz`C=H4um0w8K2N z`|_p{AYS`V@b)+5P5MFNIal!zY0sOxihA@zO7Ri62y#E#3eFx9R zRe%J>;ZA-l!Z~u zpBVF5(2!48UZXbaj%nh|*$y_m!Sk}Ht7W>k37T9_1QISe6NeIU1PyFmANU-|FVfNn z(uG z&vl3T0va#PQ^vqaa@a}%FCx@|{05O$Y;M#av1uoIo6lEe!;vRZ%S)(ZJL$dCs@{dO zFy#!nNuSIjxgL>nB}ha=TeDNrT=Tj7;t5>d528l|NNV*til{s!(KWU~o+ym!;?zqL z#NeLL&wy0Agw^&gh;PSj2N?B0SawP5!eix_K1jhH-IXEv-#-udq%POzY$^+C(ymkD zny$;w<|r#C1UThL2#}^y6Yu=Ae@|l1wu#}@J$S*N;vn$q>!}};agemr?YL9s2vNY# z^fq%LiqYcqY`fS-f`K%&8ouAQxl*|yw6zP-q6r>2v<93TL6Znyr>i}m>z=#J|7REe zyGN((*Emb$MzjC6G2)-i2`(1)1QHVey+MIb_`l9OX8gau2>v{pk3d%D!7GnaTK1Un z$xc%yHFZXPyPrcUJmtM(3{l%iMsLgGC0>%h2zgQi5)2t6lU;BCLc4y-xrQ}0v(raEoJw!ZIck!}Gh!{)IxASe@_J7*kNfnVs=3+;KJ5R!cjxi=$)&#-V5PR1vuRou-_?y}ct8EV;om40`?uR9#DxFn z?-bEYIZypPnt?EFsjig@rErBO2nPJMsEo=P-*JBMtNHEV5b5(rNlGbsq%U4xH`?&z zz5S7G_2TxV)OgMZo0VyJZVieOG>6Zpqm(cB&19|*&zyb9kquwrYv;s~`?1jw4`?+w zayjPGwtG}KXku1q2AA9Q-gU%P!^T`7&mGqdQ|=A5va8TmG-R-Oi63CwIyNPR9bL^}H%atw_!ToTc}Boxi2!_M)U;lR9fIk*sqz*Aie# zBK^+e;?gQO43kSU#R%-x3BQ`SKBc8=2!0*Q_c-f5F% z)?X1D`J>W$;HNFdP-gt9299KT%U2s_csuVKEGV0gNv!p}t?7%Bjkr_ysWe_r-ZdP{ zwOkbQr9K)j?ki(cFBkqK#otAx|Jmbr0w~XS=Ju;)&7*^pf`JdjVdl*f$&LS1!ras_w4RsjQo-q90jj z>K(U-70n99;w0+o(sfeEYP@vDef*eIBlQgQtgB>0wpH?|ylmi6U37(ZMf$P$ncou` zQ#i-%yf4-mm$~ItDqhQ3MZBFXA_1!l)P*q5ijjVN-6xR+eq?nApUuyhfmcmP$6){2 zCDV5B%pb`M-CMUg8E+|uc0a!P!DAINSv$=us|jCT9r7V3ZC`Ho5Aw|u0Mir?4bS^7v$*%4#+&?LdJj=WR%$#FiX zY_ev>O8OPGWfG;gcWT{8q>b=+6&N0_S}DA9<-z63dSvWMrX~CZ-)zN~IDVaGdL>HpGvvQ+fX`2IQmBUKXZJ5xxXGAC|7 zjhIB)U+eqy-$-#amb?paPH(TNd8t!nS>Maxqqr|-N&AYxjB)w{&#LjjvY*pXuF5g_ zaz~J!Fo%AL2{+gIgaP~CM4$6%GSRnUE|K78hgq$`q0M{peqXI3_a3Oz8MKdbxZk~N zQQE#-=26G?5V z2gbpan>p=~+TGhN`l&toZb#diy%XU9dE%&3t^8mxVv z2*X$1Dlg(6_++t~N_=*dN*p*1yV&!L$Fp08?cV0&QF<6~J@#eFJwfErqyCTYis-t_ z(BDY-zZPisjWqutcU<}^(_~q){mq#vcb~3;Dba=0-YN)nU?iDI`CaRQWD zJpB=C#@eWEV^T(Lh{u|`uIo;kjwjxG(F7e&2`*i_2xamd7MAZPcuS0jB=&$)H+?)D=3v^o*jV#Ecs2Yitxl4&oS5<^n464dGpBZu1>X(j zNa=ugy85zdf0m^Ha18W)-~?y7JI=SI13c!n3z=kjs-$@Li{TxgMfeOdN<6LK`Wk>k53XP2Cacfa&?0R)ZG(UQt*~x11bt3cGE<&3x;Gh95M8c@H>Li3InD{kg-+0N zH+^qs2eI5p*9#}`HpQD)tp7Q^Oc7ps&0Nv5yQp|`%)CeWJ>BiD@BDRQ@Pl_Zrp6>u z_W7U_5do%37@0=%XneJ2SmS}I%r}8kKP;Ol^gVeY?&3*o>`(qq`9NL#)0~aZq^vn< zBq$AE2}5)n%`tLmlbB~#Y2Lo;ph$BJK9lhJjFlRyfJ#6SB*BrqyGQ`~!S`v)tZdwy zF#5Z_e^i+B*7NkQ>RoU~MnTFi@g0JMceH;nbeP9k(s&k;nTZe}sKdy%MJNv7T`WeK z8fjm?>@)0>S2sx%jUetMHZ_}bGC?{ptCn~3n&)(FH2u;mCZFbsUuM$H`}8zoRPdVS zK6@vhOe=32KphX0PR)wPq}oSLJ5lV}lK+unn;Hyj0G@mM%#HBU?0vjl$hQ_(CzZqL zo<@bAV$+)D$k45cdPSuazRX5!OBdv>l27Ou zg^8=Erz=gz8!V%3yqA3Sah#}mup}bi_HNv|a!GQVz*96k)qmy1rRSlnD6MmPH-hoI z>noWjNdFXbQ=*GT`S&9uMEpfV0dWhsRmoYd&8EgN*pi_x-I*cupsa2x$uZqXzIRCt zhKobqE97;g!sfisbrvJIj6fuaYCqXt=KDu8<$K#H(UC7jn!EkDuqeDb%zmbFcy?f` zI&*Y++9pKEv|p9#$tyYM+UmmG=(3bfxMM<$-6`s6bmkZXUq&W`Kyc^*v%{hw?U~{y z!XrCrUQPl_XWf2j{z5wJP+E`L^U{;|C#`{5zMoVUb7k)?WL#=71G9t8{#CT`Wn};O zPARI-+Qp?R-L*32cExkzH2;>1$)zxXHU!zh4cEPt&q$Fh)+5uTIWg!XE8~U~Y({bK z0&L2WzxVN{(5Z&RMN)M3;w>1WQr~o_PnF?Mv?cr5>@rm~&FEnn%e4pI-W4x0wU$v( zi-fGOIbvq=g^2@e*>>QJu6{fNPrf-cY2B(}5cXh{J1Au!Sndl^z={{qCe8TXl2Y#~ zjrLqTCP-%QMUUIV%f+|)yx&NZIXV|xtix;dnfJwXh7|;Jx1TN_CG}w^CcTwsg64NJ zxsw?$KWf$LsV?oiJQ&#W*q`R&RqpGFIhh2I&$->n4ns&nVDBpI$|d9P_sA|M*N){1 zj;~@!rr(FE?&mJt~$N0#l=)3(_{q zb#yw=$9ZwCZgS=YW->YVB#yX%=Wl4%iU<~*Pz(l<*phKv=8tsefRFd9NZrqQ$c>)u zSg`hD16HPdOcz$MB@H$DX&FbZZl3I<=dCluHu@i)YD{wnnFu(;(ntai+dv`7U&8CgysJ*2v0E_@~YMKhF4; zH?cK?&go|5;pF|xD|$3_?B= z2K}BzP)QAwBlQQQ&r{;5u_nK>_NGKfd&@;6BH@ze42I99(J{`=^Y-MlPKJj!No7Ff zMgKltjJ9(Q%v>=GqSc(Q2wHN+XwHawQVgIl;_GszjV#s=jg2nF zf@nLND$LiBv`CpbSLC3V~^j~tpGrrl66KEN#Q*?om5>^*Q?) z+Ry{1rOB*Y1}vLdKq%WI>j}DJhnux+XN7y`V{6Uvj?Ng+U%`$>+yQnnsIa_$Qnt=4 znyN`eBZT%s-v^M+>ux~2USu8`?ZF$|cZ;Q(BdefaNeA(k!=2=OQ)Ku#HkY73%jE8q zT;$S<8xwv+9JAB>nWy^ZLHOQvV8b(S$_xdWE*^kl{J&?xLi<%1- zc#VJ)zct3!&Py5h{?DuNT6nora&n~1Ui+*_Ze+2suJr=4A?eF)G6A9 za}2|=RyHkKBU?6^CiR<9h6FpM?DK4z&!OAamG_|yrtk6Jk_<8-C;7bUtHPJ#=*cG2 zdTZ}OBkld2tot4}f+zhhmKcQ}hw5E+3;D5Pe1Rylb~+l)QGkJ45y_NN&p_~3f=>J( zlAm|k;Qb(kD6MznaN{6ef;X()J>?ewI!%sR1ee=a-HKJt*#jtnRv4)4m$QKVy<(Pd zYaO04-aL<+mna>Q+a?yU}VETrH|EeH^6d*>`LY)s|o;;~FgoA_oK}%F_!@ScMy0QTxl$ zdR57j5$}aAQ zjd@S~A^IEV__z9+^WPHZXuqJ1%LLLgL^#|#eA`eE0Gkjbeo(*rvFb%B=Pb-?#%2)W zVUhwh)dnY6ii@3^-On>2U%}%gK(?^ptlo1KR~3F+-t|@7h&z$ehxYj92ZB zf8?bL0Uu6|iD1Q_(zh_W_}27Al7{bW>L{}e$8t}Z2}}k|-z5__9@-1s1`q8xduq@b z|D+)2BvD*C>9nd zZLaRJK`H1+e_M3~X-_|poFt_!L^!nwK5;Q#GvLgY!jp-OzBs@2-Az9k#4=-l{+vtb z>P13|;SgU*;NvQ#+KZpue31FjXN}vT@58nBem1=7NxuM4$j7ccIKw%N9SE^?9y;NC zP|p~WSG`!mE1FZN)IK7>nOA2mD*+5D0sD|JqdU?Awp1GWES-nuR+D5P3B%YpL54VL zo8JrJ*Fs#g&6qqbop3C&$&xNR4_379bghmbJpIu>+-5&yrGZIVS(lC z_W#i4VkJDYHdgz=pka_n_cE4tAm%J6U5&FV{HFv)21R`Jj5S58v-zCM^L5!b4EJ2L z1XLVH+TX?4#>h`PYo6R*B-E^{o}zMd=s+htP`lB)34AhssfoPljY&_Q9%qJoKD?wW zbO*2HpttuYl`<z4B$!;$Q)?A(8~)?U=P zvm4jMeQfp%&YS$uB-+q&)E$m$=@CB^Pn#D^OBQs>ViCfDS}UgGJJSm7XA#vYBrB>= zGrXWovASF6b{lHgcCjJ zl)p(ws=P!IBPf2o^_huPWMqR71&h?P`UMJwm{CLYs}=K``uATnF_LnywG$I_kRGd?!tSTEoj^?m z;gR<~ee6$O5f&yy%6GwK0f%v?FJ;2tv{hy0Y#9uSNDvrF%j4+aixFvK@!&X5AnL;9 z7Hl3*NNSC{3%%gj>5@<*Xm3|HMhc51dJQkCbV#u-bKqWq5&nkig7V!MAX}6&D<9E$ z`cq7MXRmqGd-t{AaR;mp+ME{5GaK&4_uXGLUQ66=5M_Ssl#;Fs}8(aL5jINZJrH4IyR^Q?7 zz1l}oMnef^OY-?C2`{h826Iq&X8>cRqc2{uVgkC~OsYBe_~M54gU%!M+c2r1%oupT zM0Tlk6(fL8fYi)=sC_xa|i3ex>`7))Y~7-D%YZBl%C? z;JtpguLXKp+!&I!6YK)AOOiwcNp9!2D3^q;u^&n9K5`JCE-xz57)B!ps;9`o_k$~Z zOm}R*Oe5EQpapdebg6MCE7U?12Nb!ba*1=Bn~xc*Gj??HQTT0Ww3Z!Qz|%2IUcnY0 zk=SnIEpiXb_6|iSqZ*qU_W{F9d$XM;<(?|aSpA+%TEkS#Ta9fX1SiJDz6ihv@r+}) z-En7XpU_vL%6xDCynASH&bkQl6IT3e z>sl-#B#Snd#O%8B)EZrEsryK+proWU>mgM5Ha*hiEbhEo&n1S`dugY&2XNdwduVRv zm*oBFG`R9y@B94YF}c};h>O5(g-C{Lc9_YB{Dl_bf;k+2l*WVAp&c=}W{Byo zgKup5R%iG5wOZBsp7AwW4Km>{!vY!?ZD1ICN)H;xpI5T{ z&DPE*!{ZQ1nA9wT5w`H9$koqTPor&!CIqlQI0a^qB$skAK$A2%08xy<5dBXS?FTeg zWgN=dAuQf5?R*1gXWa)H4~yK3O-S?%rD4P?=oLuo1eh=VC()DdN88Bnjnh`-SILtN zaG&eGaW5VfA197lOF>j1&@f0Dld??~O)G=ph>e&;j7ZA7B*EJRt}ETV4$iyABvRn( zLR3USYelWpfF{I9)O!&XXus0RQAn)PH)m~*!rUy1)vB?x+Q4)PcQB|8=ct^`XvJCm zl8Z7;R7Nr^{X-i0#VCJNm|(xx+arhW1=)#& zbF#T|rLneQ?&nokVv_f$A6`&V_#$DNfdVZEgoJs%@UXKIn~cQ@?NM*tm6}RLJ+nMGNw7H^D93GJ>IlFvAQG!vB%S1~0c{CmP9oo`sn_Fw&%1 z=60DYzJz8p?-O#Dd1%j5-gH!fdiN(hv&v7F$L}K4wXa>)ta=n!>|(;D|xN=8Kha$pkU< z=2O6>$H<$`)@h3*t)nHGsoV$q@{%h_k8-}D!zi81d9aM@k(d{cwcGG?43KEnyjib> z*=rK+d)Kx|1NqchTbFaVWwW&%aj98gRPWMiX1TB5-2Bvma;x95^|j}bvz6b2)71P5 zT3z46<73tM;2_n$S5~z+l`(m}5_&&-dBM=|C-tZeUDvad4beV{sm>Q;OIgHd^V`-c z$fU-BOCin!y<0bJv*7RiI%Z}-sdVLJv#s;YBA>;zjvQ+}igcaLbn-|S8|x43nImxB zCDpIc_nwtE^n`}TYYkF!46&@^kxw9eZqWDttisp)GjuM_2k&zG@%|*xewMEDqxxU4j>ZPv5iNKy{Dd2L-N9L!0^1Khg^R|z zDvU1|G>F9jdF_A2kpSpo&EgQt*)vH@*~Aompte%){8sV!L34v*%N?aQBJYGcFTm zrSCMazlc0Cd{^xGH2XfrkL+CyoUN7ps3`t2 z!opq*OimqZ#9Qty)3s;PQ^VAt$>GIsZ|L_ry+wC-CLEvh8KMV3H*!GG$Z~u&5E&Y^ zU4@Fe$MjX{|hSIknn`xI1lIS!!7AiT|8p9;(0PB+zN)`N(`U!8EVe zxZf|S*(YK}gVl_+i<7HCX0<#;ga5!^Jmz!1bT`$)FR)fqgZ>1};p5Qp>_~|QFr0)1 zy>gS@ElG+jJ~Gls^w{mT=-aE1z#ebcN0bu@S=;6@t$s1Xgho)>QO+vdPreRYe;Znf zQGevyQNRmAi;Pz8sa@;G6)K`={@x(!<ME}gkXxZeaJeCZjpwmblU)%&8%bOJH*ix{gm7Uu$km<73 zhXPT5kOwb`EMDy8(Zm>8tv+#2w7FSxTnx&)j_biN*uZ%L(V~U38}X^O5!$L}*(36> z{Fx*6G&KMNpZF$tE4*-VvlqR|mZS1&yhIg49R<@7!*$aIW`K?FcKZ2tfZ>jsD3}jV zc+ZRVSqu`Eyf3y|^wrqQuw-N1pWS^4us0#1d{KuzYU_B<7iVXGWL|zoZNr~pu`RCW zO>%k#f_*)6V}IMko-{3T4_gjiboUYDJ$Cte^dMVgi|hMpTsLu%Sm4vyAWGEUy~F*2 z0Dq?U+|A{bBa*0Pr+^P1u0RB79f4<@F`^YM35p{1Z!@yn7%DzuU}oS=;Fe^;ch2)vk< zoxUQUB2_k1Nb7iuhBy%LhI_w`yNDE0*75sc;zWmsf!DS48R!M^0xh-Y6 zK^`LP(Gwdu6h#w$|K5e$ZxJEaB&A`{V5jy}ttsq7n7;Q@w(99I*#3;@UdGe|Uf}3O z{<81_s5NOf8(kvh8xS}y(k0oWN#a~2k=73u^SKsmv>WO-k8zVNlG}Vr)*Vv=x&+rSZv13bd(AX9w$_*Tymo1=kveysIJbVo`x_| z9eQ{9xxU26e<%xc!=#~ZbG|=Y-44Wdu%5w`pbO`L36%NL*}Iwz!`Y;WIKdbu1n47B z>Muy%nV;XIdBd-;c`Zt>IxAR{*gd!`z=!e%W$V@0*(GR#S>uk%XMp?s^=*QB6ui}z z5jAcQPDY}ITMIRN4M)?;{^qVbe5PT`Om76*lml-47yy@IoWNok&6wG3bcE@(SHulep1iRPQF44(&z=84Wsdd9vTW6oibMr86|lr zoH?|-Tuf{AO#>{OTb16?GR7-nMXol9^IsgQF@=jOBMaAmOb#Pig@FrrLYzS``r(~} z(uf3?>(XL#d)B2F-j3fb2={1EEBXlEa(Kf5s`_iAMArE6Mf0gmGA-m2xLm7emv<<53?pmutP2Dsi^C3*R)?MHqIy_vTm$ znp#$j24elpvZjZJ&FJRkV&9wtv7|BncLt5L=WPBjX<5=N zV7?CeD=2U}TrQJ3VR`Ju3aLjJ$*Zc4F8Dg^`b8*ej0S@jPn?v68Eh0(`>a9J(J039 z7_)+A2rdM{k^vGNvTc$wA`kR9H1Gy_RW7QR%LlQk%^4fo8@)>qA>*>8le4}{ey0wJf+s}IziMul=cGwb^-Lmbl~^$fYfpRX`tXByI6 z$=ejpG9JJ3wQJmD@{NFD(|1#AIy`$BSG_9Me$s8sNX_y6(Mh6~0nAe!d*#K`obt)w z;>q~div4isPlkXd)vs9Z`^lb5>%#q@Wcj2xLNf$sT)`m`NdDGvi7+?k=wxid{rKgJ zX-Q)%BJwxL!N@Ar$;@CjERWeYgRjzeWGi6rVL0lkbp0+RP^Fov6ougFsb2G=Jx`gc zhW$C%G_+cADsKZ@^101$x366s)NeYXm(9pQ`_$S<%CtebPAzOF#LLy*t&6cln-KcS zHquvXRUuD=u&rYf;b(>8TJUzn>xU5_Q6wb)aCwm9aRRu%@(lmuH^hW!WKWEIlSr9n z(Se#Yb4R+=FM-Q0c=Aq&k=Q-SaDqhqu!rr$TbEHVSi5(gHU#BG7MOD6INS=~*7@N* zdo7hyl$+x119Gzh*kGp95j#I`wFg2Dc?@qpVD4YjT+W<%(P0XlxW?+t-#wB9p%UJ| zdy_y*9=9e-bsm!a;#tl1Y|DK961#33n}6o&?3$}Wv7*^^;~SW`$I^|uF zt>DksYxc@Zt_y(mErrGUbObK8$}$lUv$@ zjs(AmI%6!gRz^Nb%eW=QNk13}n?75@)wY}Jq3&y@4#}y}h#2>AmZ=V5TH2W`NHI0! zOwe3M=PAsBLreUcJM3Z!T1gJZM^j}|yXFd@gI(}OkVP$9Dd3JbW7cBk7quEY2;H0{ zI!tu!9y4i@N^=$?wm|xVy-+@o6szwuFjNvF*W8b@eUHr`fyNbhi#0%xb&V>^NoZLL zXGIBuQ!N{=kmEH;v#K1r_BmR5#}nG@!4tFDc-ZNIx@PO1R*^l=!p-2IV(za~$exWo z-4OFWZ-uhJ?9Pz)4ByB9mcY((0ZliY(wJW~pw~3C3x!j-nfr%ji?oBn#4fnvu2~9dlZUyA)dGrq9qRp9@37q@n%arYw8s~B=9ALN!6o%YeMy;J$A0M zeVjn?Q<%0Q67eG6P9yl@rnN;6#Iq37#6qWm6{2tn)v{z>J{eDVdP zS-68k2v>5f#P>mUIK;V4(BrLq-(V-a>j#ADNnDKQ;oc~ESHc1|8p~yB4-k4?9r$VD zJhXfz^Ssp^!c9Rh_kDGzXJ`8q7m_)!qudNcLxl=g1ECyS+D%USd%o!45uRq-EXj7U z_4>vH%bkqGSvSRk?srBW_2}nKDN`Qjs>e>q;Kfp)gm36~{_67tri+AVIsV-2qihiU zK)CxZ2~zvD3hN`6GI3U1QfzuAwY1~eI(Sfc9OMrAv&I6;AOGw4|6QCZ=!8vHcGmw8 zXNsMbgZ-~IQVtd(98)Fn;ojHCFe{FK)=8(6S=3SKw^g02OpwGldeW6J_}E=0lC_8v zJItMdID}tqNr&yQ2T8?Q~!Q?P{c*nX%UZ}B}$QC_3g4a{%@Z& zDD=;gq&8dQ7`(iDpqU4sbM|M%x8T~LPC|J?f>@n&ui5NcJ?=?dS~q;g>8;~BXgN(s z(HA2o&A3a!F|qc{@-5W=HFcYi=nP7o@}7pF+{0Q#IYP@;xl-a8@Z%Yg_sQn%Kj^Me zZ+Uy6_iQujd6XOMF*HT`t<~lEx9dVXWb+n!e`uk56Z+#PHz+ofQO^h6V9&3wsQ~5L zl}Av~5FRK#YYAC{WVIMH+}GWdJ-dJwRjmKe2x48iY^P^^i>X~5g4a&I~CGuLYGU zaf10GqI{*L>mnH+$^_0#(ixPo@C~cS+m3s3eeZ@xmuVk861e_V?{(!bzhf>5e0|z3 zdsdL~`{M*_eaWVn5!vD<$G zl{;*vur5HU5H4g3?Pjh2;%`#@t5R_NIo%C^bpwU7`31z9ob>1(bs6dnb=DvytPMY1i>z13_paqn&`&}>ko}8@`ZU2&=iEm4R z0o40wq@AqpL_DysgKbVGZ+$3Q2JDV&Bp)nqgfO6=)hu0C?B0ocqx4TFL;B}8i#LBg zSoyBEy}7U|9%|KYQR+y#^SWX4bJyqsw(kETY_tpbi}3zmTKUa7eS8@=mok?RemRp> zA0N+_tJQfovy4_KRSsL*ukP?cD}yd-ircy2amL}AV7`L@FMJ9|%PAv@KFqnxXXYg)pTP`bMnC0N zMX7^XgNH`841Fe!=)D_4#Roe0h>tDpkz~0yMcrqUqENFbUrqC4v?+wHrBf$GD-PaL z_W7gb2&$m6|Edt)L{;8IwcR!14{N*{%=NtZWgAn4`p{b_g;wcjvP{B-oe5@cG z&KxF<9<)2|OoH_Po3g?IPb29MY(rIjd&{TbANX$?;9oV=f9tzXcb>l^w60H4Rx>TX{CcKm_G3c1;cerT{(}$7`aHGR zNzE;mJ>TWDbK0;z$s{y^))yi_fxV0gNYs=TvxivIwzePaK5f{D-px6gK1i?#Z%Sva)`Snb9`RE-nL*tuSl@`-^Q(bXpZA<-|%9u8dJ-B^k+64MuXOcQU{!dLr zOpM=;FFHT21^f4?BdPkF)6OSDfQ{w%yuvEEzpCZBj)Zp*$P2|>*ipEd1? z3K=fB-d8{!Ycb|pI^cHqB7(NjU~H>At9Q+Y4Unq`{SQF<$Nng#bp*B9mC?sC!oE+B z*GjVb6{s?+gall;iWn@4UETYs%Lo;TB372oo9ScUW* zq3h{>uX25te396-+3-)a>cP(kCHw*gVh|g+#q1Lye+wnxk9e4l%_SwIMjL8)Yk3b; z%vxW)3}T?A+2ZPRqT5-Z`C0S4`(8TWiQ(dAIzPt#LJ`n%_AZK+n-kUhQQgjo*!lZl zd#0bBwNn~eldl41QrtOwn#JjrTfbsih6ZT~Yuu*C=7re`aNe{h{e!dnN(f-qlxyRZ z>JJU{rda*U)Ac*oz@b>08)$t7Y@J2nlUv(xuQsFJVF-&y(?6O=V=S=1%E}F6exS7c?y$`^2*2I9eMD~gRXP-2<``vwav)Huw-r5B z?j3Xfo@JpB{J44f2z6EULfN6I?e1U)i(2=S}vlIk5t3ht%!4a}4Q4hJ(VN^7O|}-rq-r|Gt3iFRYnUf8S;cCPQvM!y)b^ z$3~_!?6xSLSbXa+@7hB+MiFfKrnkipU4=G*%34_XMi9&z(dK}HuO?CR%=>Z&Zu7#820L+f};stzKJLRRiiei?PafK5zYziR(J! zn@su2wo9i0fj0-G{hb@$&pEY1Tx-X3HB;gIb}Pw|B-m9G8?msPUfFWR0i(u&qh|Ik zjfK~2c3+R8;YE_iZm!{5!w|C8r_FUwLzEeTq)^`D{9{{JaXGdm0SU!He#u!!O{w}k^|FlVe2YpQJheJI|&Bq|Qy zOOAlUoqXr%sVs~+;5CbsKc)5Z@Mf_@nb6AhR+h%kc?PV%xwd~?zR;a@H zZCcQ|m?Fv6zZE*Li|{;DtJY25xahoI&OM{#J=~0_GC}1n8C_kePkM^9OWSxmpBsaB zOAfr?-_M^tUflG~7W97N;w4y?KjmRxt>6M4Sf5*LvDk$L=JubIpNlOOy3J{RAir=9+ylLWMbs6pEz2kwHqFw>IrXr7DMIz)6+44-?|eM5C^t= z+^~vxcu)LDAN#>s0kzoidOK$IKTY=EzTp1;Rn&2Ce~s8tGY~A|$l*}$>gs^%CgT^I z1>Q+1ee5%SFleVgG)!vaixL{nceQnPJxM^V0-`^#q$GoDsv;mj;;9{bBi7B&;E=o3 z#@S9Nn@m7()zaLr+mVa^jG_OQvOkx@@5#WD#IC07yU{~&8}%9tjy;CysPNs*_h#Pf zl-#pSxV!?!z*6;wY1Ag@O!{m0!xke4)+*Z(L+3#ttGewtsz!3tGU|Lpul!kJJ*a@x zaYkHtKW)J>1}cQRo7EjqGYA^c?`{M`y z;rsu^$HQ?$Y7ZaQW=?N6UI0TPFLG*X{GlW(FT=LI{X)Or+q&4?e(UoSQDKGydzD6_ z&W(z1U|T*fwiuxpAw5&@m0eui$x)oZIOiwST&r5y=6dyt{s%Md#!VxpA`OLh)T+B! zTmDoWhtngG$BrStJkmrjy$pU`Y*pcfAEdaQ18}rJ9X?&`IIzty+V=;R!EYKrK%o&b7 ztIyBD<)u?f--#V(?jfgF+@(|WsJzx=s~7rR56AQUXwctRZdT)%htPPf7goQal^uk@ zuh}nB)AM@cqqa7GeNdm~?hLfA&zB$x1?Pkd`>oi2_jTVX?*|^RGc5JiH%jYQhYsGAVah8cmB^*=G0jxtrur zqms~CNCAR3jx~>q*iiRm2DTpBLMcQwlU`2lskuN!SYg1iwPj%9^6j-QpyTN`PngRj-3;P;PfP;T2X!p9!}ImFV8iukn|yr|Xr5i=^8%cT zt0V5of9_`JU1wwf`to)t%X&=AP&eW8yPI&N=WW#0hEhxT+I8H5bU-pwDw7E>LMCIE zPy{&2DOaj#G-yvZLj`Q&w>JCew=Pf2gbOv{iU408r+Q6jy0=4U)TrsCBx&FTCo!Rs zU#^_n#T?}Aj#|gP4Xqpgb0GdV;QMFGx81D3L4wL*zakvB|Bt!O^Dw;qzDF5hQ3S;_ z0N82l^Q%r;eSCdMjmQ^Sre)bF={9?ml-}=r{b=LJ_Bd$tVSBO44mjH!;fwFglR4mF z&yhTWV%Jf|xK+mf%VP(>EIX%3k1HCGyiZ2zi*@;+EQCwcJ}ud?je3)P*S zcZCyVKx-ANBVb(8Uc4A;s-UngE0`UMD^5Vh(18FE^Y<=V;zr+Jq7?E5LLQ%Xpsw{h zSomn~%YUbr>mK6EB*%~B@twFXJPbbH*U$pRpoaLTHS&My&YS^~x&xzZKMfS-O=&Jsp(2I`E)i?7!p$anJuV%%#`<-Zg|mE*b3CUvZTI)mz#7wT%dE0>N->RdqshO>l@nw?N(LcD~7 z|AqQLVC8oa9Q*R=rWzm7FU9G7095eR@$~R_{@p5;RDr*D!Qc}+-JFjGb_8U??Ua|& z6=SOd;g9mQoRcDsh87#Zl)QySDiN5@pRPlR=baW#3l+tQ;rwBX9;JN zCC=Su1@X26V>>2a9DiHtgHrO9@pIzgoGltI_5 zroWfz9cmPTiI&R;OvfA!8u8F~6ZrCh@a<=~)M2vH$kmdURGyP6qwQB#Dj+t&x`N9r zwOKH*{sdT51QG$}p1C5jjGjnUljgT^aTh=B$hdjiD3QFDAkgI2#fvqyr5SP`5WO6l?7FQ)}8sSOY#4+-Y?r9*KquOVI?o; zU+yY3Ys$sV*JHHYY3zJ1dQThC6*#QmnB1Xw@PP_ms*L!D*U$GOp(f#K0mLz8eh;p> zMBXUfknthIP$`&)+uP@hua`l0WL0_uNy&Y#-DgZ1O}gQlJKk;r2alvKHYUr5Xxfk+ zi}LWEHJ&ggIBNYaoYyF?4;GHl=+KPSrzv_3cEMK@;NV))yXMHf%K%ewaQTmEorAqT z4i3S|hn2m%?(D-kcR%S^Nzbz_;?fhE!-nN z-pBL9Kqj*f7)^nR4qvR;`tHiNJgtrsMWvqD21jcA4lP`{?3>TxDA^9eKHQaUKEx5o z+rh6@R@{qp6wNp0S?7`4U>dj3HSRrhhc%V5!{hnj^4N%_baR?N58e)wN#hJWuvoJ`I%pebuE}oe z^@tZZX{;0WzPr_EX~;u^Xk~b+2OxkIS zKMOQxBHNuuBXbf_O3Fl%Wj$L3DKnl)cB(~Uivz?1lU~hTf|f8{F$w5vXqWYOoUU9` zv2DaIJ_Rj_`M_R<`n>xS2Hj01bnNO#mm0$y32CdrCO5S32t&2Ptz#`&L` zzgKA#ksNN4HAYI3dhR`*tUl*Bu(+IYsu>a$>c<7S!@iKBW~5>^Mo7z;{?3zNy1-+M z@L`&KSHtduw)mt*p$;C!R0E}iR0L=T+qX45P68?Bt@tU@@cULsB7*IZn!i7zW*}a8 zvbDc$p;j&FN%cV2_>KqF>`TYPP{*z*(qpD)`tHm~H!)Q!68pfDg! zut{z$yjAIB8}4~K&qt4enH*z^&Vsgxpau{)2WN{fAqj7n@bM`z<$W38fp*ci+ zAc4dQx|qDtH%o>Eju3Bu)#hctI&jy)ATc}jnIptS40i%65;aJ~ZCU3EmN(!=-&r%O z0sTEYK8h;v)Uqis1;WNrS3%lBOvoDq*Aq3m>hmH}ETS_UJNN5m$@3JWm;8CGlHOh9 zbA=Xc3(ip%M~o3Hk%U`uAw@Js36*w4R-QsOn#WZR{jejtlCFzYQJzcOpzXMJ6(>a` z^Mb6|6%vngwrwUgsu{6QcR&leZ{ z7)IB?LS!N@~Y1S>~0v(_hD)cqOtt6NRh&p zd~s{mGUJVkqm;=01#ZVKioaB~yf4fBBV_ftP4+07j4i^bkEq2&tgilAF@2!2V6~g*zFO9!W&8b9? zetf;($@G>3seJY%^43}b>rGcr+m#sryIIBE&Wc$ zn*U->Y&qO7<0-2uczDw4Lo2pJyA|dM+G_}Wk!c3$LDa1~dt}{ZKohgo*PEu#BKxJw zFOmj~5^$eSxDhc0QAZ6)A??gP@S8h#VuM|p+OpnPN?%+pw;Fz-7ajs(WxesI=}CGk z1BhagRtux+);9j3d?6|`tz3C?7HB8tS zst9q~h=d5=4=UEY%iY*_kD1nMNB1H-jcp#tW3%*6UYf)FM2*d^IyG*}wQ-f!!0CNH zkeOL_&f3|C^7Hu~*hsh|^iV>?;c=L6c1MWVmD2V{gLmtx%g->8x2*rHDZ~C}gWhah zzfUmw3xnPYO8wA5Z* z|EE-F1*1I``|JMzh&C zxc~B$b?B~xZsU^ZA*Y#h=`bDqY7JRgm ze`fAiL9NZ$|5Wx5NZ!Aeq7n9rSqBjQHv&HVha_x=RKF!*YVHz=WwAjCT)8G@eu6-@ z%4Kek?wt1egQmNYEpJb?>|?pcK1KKpe9~Cx7pK07>CSZ*$Z7$Vm&cqrupphBE?RkV zHPqy^++W?^b|c&Hq*AYSLYVOu5doK=f2Q&`KC};hFZW;R#RK$%Uj7qX|H?S%crRl* zZ@hlDY^Xmq*lLXEqSHO}A0nRg`G-us&i=0={v=PUC%~t_C-c~Ld01frh&VaF4bZhsgQtzt-f!@ zQEP~Zy>|xpX{P%PXQmQe$%@OFZwR(vG9`B)V9ccaTJ$gtfYSxu?Kf+anZpO3oOjY# zyp%CI`Y~X2DXdkD76wK23E2f<2REtYL{h+wAvP|gzbMIh3{rIJ$y?L6*MI+590#Sd zNcG0p_p92}Y{$<1d%nZT?Dgh3nDXljACP<$aFKUuKH_$74Txuj9=M=JD}jLJ#Q3j ztt0B@zJ0GhdjgJVnn+5%@ir4W8i6l27Rt<=2uaGmxnEQq;tsmW#$oR4|NI;J4}Sxu z<*?kzkIg&@ZqWS|-Qx>pJ17*q{s!J@Nr1|s&BfK1H5cjf&KFI(SFk_aJvRSKJk?F)*f_TK zmy)BCwgICu6r9qlOPQ+rTwP#?Jx-q+1|FFsSQ(?%9}OC-M|t;6hm?I^bGmbh%U~|l zANkUvC#R2%nef&&AG|-Z)a}{&FuJjCamu@Cte#uP*0$H@0_c~_w=RjH)m9wXvNLk33&$yPr3qh=E}Hx99{Sjt(&yV9sk&&QUqa_`}|Hl zJF^y495q{1I-Uv3@@4Lf!raf8g+YBLK&YkEtme5NbQW_cn6Kq+I`JM`j>_q|Gd}d` zgBHAK0dGNl<3V%N+U_@7dGlU~0N}+~9!LMXm*rxflP1kcYc)#08>UHzoHo=j=T23+ z0pS{Q^yKbZrX^5*x|kxZFIHUE-3C54ugBv;ao;-5~&7D#$KG z<1ArD)3PSX_EacxWwT2z)ywAm-EU6XulX0o8kw2fif3hemZ08B;}_x}@ZuoQHiKtvh2`N6{cT@L0RPzK^d0;8E)Rc=+SswFfln{3ngP zPqInVHXi_}%`eM20SPMe7USGkV|z7q)5xLC7e&!gz|*lkg8#$XTgSzfH3^^)oIoHr z1Pzd&!Ce{&PJkf6-91Qfm*B3!8h3YhcXxMpr<;Dv%+A|y=godQ``dT_=wIL4=iXCx zs;cYOsVXe$aM@N$@}3C|L%5S~#%Uld%&7bsg@W{9QKFixW8cx_CFsyEnB!pDZf2pGsdO|{dyFiTkVh=$>+&)YcM5&Wjg z^5+qw6`uo9k2#yp3VMC-cf0cE(_1ZOaGw~ip2{4?0LvTbK22y;iBG&=+38c#zI+=+ zyVwN1^e6l~v(Pz|o3I2+e7Aeo`G>g``ORgRYzg&U&$j>e%xOQT?lGt8*kjk_aB{M% ziSGBwg-1Hm-W3Jem`zl0zqk%m+H@Mcr**oxbpOQ6c6=hYb3@}k>F&-SbSZTI2x=Qw zwSLevUKvTK6tq4#XF}N;{55IB%sVhT5a9zVh%Xkz;y$u;m{W4R67^l3yf|0dRRU4? zq&`Zr8)JR&?Vei`dq5{)#X~P@F?3v^U%nHKS+#PUU_YWAOtbGXj1V$z5kWmFn$+x~ zPq~W+@qn+EkDqBv@)?s^Q7fmelAV=Y-*HgS`k!8kcCAL6ZdhzckD~pDYhFezS;A<`+ zR>mlFhUj(NM2SS)viO%g8!-Wr-|}pvU)P` zvCyIo)+{6M_k@|)!?>QsI1HkFKZBc)4&}y>c~y1I?{#ni?b+Zs{I~f0U96Fvo%x?` zYsgVnwOe9G_qZ!tH~64XScC2ZRi?#a(<>2E$9{NCHegSR5sh~xCq5Vsad)hM`a&Z= zIqShlrsG(#R~J$=(^HjDam^gxqkCM!PK>69V=yV@)cz{!rBIpcUK^To`F**|)iGs6 zUNxDNA1?ZcSroGN3SH;)Tu2UjCy7H+q7c7sbc+9bTaApI#NaqOsRr`<3qQnqb{Qf; z10)OvBh~kl2uMCX>7`@@4tdji(ng)?GW&`ODfJP?(iB;us@o-AN5Hl9%BNU~{`k>>Zl|^|Y0$n@o8&JMfZH z)o^pnrP1+Njaw%awy9I_ZDo@%E;~hFiv`Q=jzSqyu8}Lb-j^n;q&?P|w8B5q zmv<12#78ZOu-mvo5%1A`p;&MMtIj+j?|6-H;ZX>+jeqQfbIUMzU;E=a)AF(Mdr~bz z(>xVpwd-shD)`g1j&SY4qRSx_YRUK;bK`h+>$Gk#btXawtKuOPacQxw$`f1%F0 z%T+CqhmF4grU>R7MFy_nIm=K#hjNG^mgw`C)Z-=8vvi|38rczx4!iv~mf;I(_mZEb zAP|4xEAHNzJt=Z8@@+pmYE&C{Szyu7!k3Dsf!z1eAR*ukiKmvYi42O=TVbN$1`~86 z@psu(aG2sa2m>kMG>82tHEYUu`A;UfalI!A^(p&V_Ub}V(ZjPj>E>xL*}o!B!!Jc1 z-PHkpPjaZ>Un*kh9$1tS!8OwQAB4>x%M7Hn({pCgvg*P&!Qq~yX8U3VC_4o-zI4GQ ztw_cd+i9VCIJ!@T6>KNv$ZuMp-_ZuftK3I&H3Bm<{YWRV3Tf$-<{H#YIjY&@)ihX- zkj$b(vMOQ)vx%iQK0F_Fq>*J~%cO0oWR>1k|3t~IM8J+}tb8hEr*2V2=icfV!_b2Z z>&D~HO+@kQ!N6Bxv6(1!jSQZTAXOPRg5F<qO=EbTbtihYCeI2&5@`eG>Qz#o-@REUnWRp6!}rz?v3o+sGSnBKfe#EvNxQPcwFgUl9h zR;o@#SlwicY0h5%Xvb~To9l;;=SfR2M|}z?Y%}9=Vgob$qxpUdhioxUa%JIepHjmL z>d|33(Lkfc>ENrqP9%H-D|@Z3E4#Fjn!13-Z!3Hw&R?Y5-}~otREzvbCXn`e8woEM zfF$||vczUi>%NLQHw2Mx3u6u`Ofl_S)d$$>^R8PJ7tYHnqtK=ol&e z*nahnz=H_u;y`zU-MZev1p18;W30bdfhtbpSMS_xbgWE~ zP2rpCYQsVz2gDv8Gn>=m3>JFze9rvT(BXr9VH{K3lP%vtzz*+G;M z92UaTM^sO#>~Av1+C)#px3lm##;7y+B#2JCeFh-5P{dm#=G!5qhEqqwyd+zGo<*bj zlWJG(FKd=1M2rYCsA4Un9|B$5Q%4fM`bN<7y%cjd9SjrEWa0GcUW9Gm{fy@n3zZ>( zOvTKKzOPEKl3c~>%Y=`WQ5jYxxjjQvy*D#dzF`6W0c}~KHCJhgBho`Au3_7 znL(YQfzm;16FFvk%uDx*Mgs|P&@w?v-*EwO zhtZ_}clM0!UjssaW{&v$p|uf{mbYPiqC%FFEzegWR1loTOY4`*|Fgb!h8bR2W0c;_c`SmOGEi}8 z4$GSI<$hqJNg!riy>v%3$5d)p#YN|^<$5{ed$T*cMM+{UI zX8SFg#-2aOc5PVv&=RK(QY5*|q{DtN$W<^#NU+n}O{WcV7ERW5XuPX;{J6A}NoMU# zjx+*Uc)qyLo*vZi7-DavqsT-3jH7wVv*vCX>>e)X3EiLU%Yp`lF4MtfsZ`0g?T-=c z*1hwcGpu*b^T}&a1_2A_(@p&67OpY=6JUT%Gjzz$jx)%0*wENP_Q;wPrT{HZpKeYebm5--e&)>}?;)t->fW?j z3q4FTJ@12yJ<{}t5gp5WCs+(ciN^h&ESgwfj}zbR`4NPb6H%f;a_0HS6B}`zp(HEh z-Y1S&s5U`D_19-caue!A5-62#MC{5@p+8t9*4uV1B?VWCSYgZsnNY(yv*U0?9h6@z zIxlN-Mf9!M{z@mvol3kFH>6q>s3O@Feie+>2~TZ-;oqTL4kCU{#pPdbt#lJXlXnC6 zlxW6Cd|g!MGC9Zj9+eiEKtx3S1G3KL`hMrPh6;~{R@QX8m#+Fl!Z2)Gl?5tli8P`#hJ z6POU(kLA?Zybi@FvUPT{P8;0rKb#S7xT*N0f$v_=;-HT_wBlYt(#lL-Gh2dTLlY)e(tDJ^*=3A(1_qQahv_XW!qW%`&;r;%9q^WP|V+q?Ww ziC6s>GM;zxPv}*!8!^N*69zE=+Erh5pEQO4iioqQ)_YLA_P2 zhbp=lrnQ8xq35f8h1|tvH#)KaaK>l!#M& zI(!JNP7$Vb`x+(*DSOO%PHem12q5wJKYjq?E~eK1`WT?n?NlHt0EG4KHo5+PhVTD_ zFw0%kzXJ<6b8Git*wkW2_-0Y|?qMU(RtD(*z1+3YE(LJE@$rLLli14l=Sm=?HZ^au z&62_uJB)OWL#!Cs&^Fl&OLX#cNZI{H3|x@r>jwYS`h4Cc+w<&PMPadzzT{@0jv|JdT=gOtW!kla1n+jNe%^xtS9 zF-{KXqFcJZa)`Hl-g0KSuD##blK~i?Iktk|-%a6Las;Rr-4}t-d`HUa)IsNVcc7V} zkkQre`V0)5Tq&1hT)ARE_wqh8K+H8D>*Sl3bMRx#)4|S^%EfgIJ^1DF4S?%c7|>Ap z`z7&y>Ql+N3(ZylP$Il7_`w{c{Q~B;2fXUkje>`U=UxEG)g1E^moqu&9CItL<2QU+ ztN|SFq&)5m_V;=}ZT`l7)nZ_ky|U~>K?ZQO?~BKM&rj)dCwzPLG1TvI^D12B2Nuj?fbwH?b(QnDoy~$Yo6Z;{0f}1GFIsa z@EVj|0r7IWxxWo1@hH3-%K^%ktgRadKhUnAa^4qgfDB*g06MRVe9lX3D_m}~QY@_{ zS*hGj19EEs!rDB(eo7d%(q`ssl5z$W`vR9}Uqy&9Sc88Vf#*$* zM-CsDJabMwUwE_|K+5{C@iK`NWDAsW69DSIcfM1*-!OIo^uUGYK7;@{0a;?H-!Gak zZ>9tlz!Q*AZ**s@G1-wr+3l9TpSuhmgHK}EH+xGAGk9)p--O1jORGa20~(p@$oFq0sdFP;#?p z_7)ug`uvDr{lei^h{PTAFlc&2c4v#hBL$GMH3l5rDf|eyx?q7EX|Ki=GgTqS_f|3f#p$SeV`yADQep4S6_g$& z_|7q8Hvnh~j-6vjdV%R+?>CU#ATKI=YE=6mKH@_E`S%4O!4&%(0F3oW@!b7gJsSIi z>pigmOdiW^;s0x``|&|swf6<-a|y&i;KHM7=gANn;6+wuGY)qCs(k313M&OYKIfToR0B~QykSvgM}#g_&DA|esL$@hN) z;(zi2*z;5ndy5hNVuI=Sp0@sE+V7h;4ICcEtRGE#AAU5goc=V>|1s54xMDz_gHLl% zt7iY^InhKGe9Niv`~%P7rgiL*OZv^)L;v1Y+3*0%ekl2%Y=TS15F}NzDH+gkkYWWFWu9G zm^06H2%owbP0a)cK~?~J$mb{gWfu3Kz&{T0mMZ`q|zDk$uXgFzEm``J# zA3hGP9+~`GbH-LU>(h9xgO}cn;obtU=l7Y_@#^AZ!)cf+>SahMJFgIWp+>J#On;=c z|IlnBDQg>3Z=r4E3~iSljy^n+_kMMOhC`XG{&%E!Nmcnzcm2;e2@5CdKXsc+RFABe z`rkrM+k_ZT(UHsXT1iC80_Qt9qJ+Wis~gVtjdBH!>a1Y3@Auj)`kw!Y}$q zz*5@q>YD)++b%kaFq(<+%?vBb7;nw=_v3Kp6lw;}$9uYxhvWGQ;|4nn$`a9DF6SmT z$EbQMiF4hOlh^&*zjEttFoSjK*AG*pn;+lCQY}8GA7Gwre|wg{t~LXb1q54Ah*-XWw4&Gvz=mkxF;gL2fAbmWPYkja!-8*?MWDD zI%z*s!NbaGxIjft@>JHdc8Bn*tXvYiEYpX;UbBxYe_VXXg^0>kmXr9+4yu~+S5n-D z1bRa@O<$c2F=RgZi3Y2Lwkt3SSwD_n(x`pcHA>$KZE9PdO_l4jA>noCt^X#wKBc}; zXDgST*1lBa!KZ&+JIhlL%F~_>qAT2>bZA5L*`=P`8fREH9c|MM>8Ym8=))hk2o!9t zjO!(WaG+Oo_*h`%zh_jzdot9BZU?Y7_T57X%H0)7vZCGxAvvN?8f}b5A#D-Dtb~TD z<*rArM zmXH0Wk2E}IGf~vn5+L9up)e-LR6%{Ys%lFpCRB!@wwj{K=jEnAL2*)EgjC8TVjU{p zWVmlrzNlcLkjE?=#=z-D-C%tv8JZC|2_^Scw8f#Ov`&+U>x;A?P?)vskR3NyXw0!c z5w@~g97Jk}*QSY6F7@_Yok^}S6Oo`ks3kv+NO*m1+BkKme+{y(9ypd7(0MAg=`s72ZrV6GC73|#f24Gt$wjt|K7 zv~0-FjOsM>g(Aq5V;NA$Hn!yBg*i#Ww8v<@=j_@2`R#_NtPu80;h2;6v_CzJ&Dw?c zS8^qkd5%TwvPk-{>8sbMUxeved8Ot|aqh{Sc)3_~(@74FhMZc&S&@~>vw_O6$F*5<4fa;Hz%!vg8WT#a4?g-={LHGPrj?Fj6Z7^ zH;}$s@2;b2hvJvJ$mvBcsT~$f2TxH_jVnhBt=ij1X;Z2+h{tkI1DEKQ2_i9e!**dPxJD?T2@H_wY&F~~bb)COE2(lFr38CxUAVDsM7OSP%ExeQ^oN`<8Nll^9J+T)>me_ zF~Ih{EkT%g*3@e)Zy_(9ne{?mJy{JE!Pf>QIR(hN-_GI$-iwVqkuCZ-tNO9FkP1n^ z-|tEivLnb@JyX7RGZy!R$`o@4fbZe)iA*Qze%#TwDD4C0>gG7Eox!^V-kw(i#AqOt z;=nbV(VzuSg_{7Y`s%c+Bm0PPj~Yy5WX6KXsvKPd;ZYWD6fJx1=9pm#KH^5?AE5?m zs-O}wU9E|6qeBlVm^Jr4vE~*Jyo8e0+DFu~*6*zzB{?q9$#x6H-Qt%fhQgN3pUu$3 z^{@^w=|jM+f-|sm7iar#;m^-@HL!1-YNg~&w#+y;w<`b%n(oN;K}8&j1J@}Z;qpw5 zZl;FmEb$A;Vt>6Nm)MBTmVPU|lvIYEmTe~Ah!mp%BdrGg-KU7;tkk2Wa)OI)P=pZM zBMm2{u%*!x5$9cEy8(`HwFmpqq6pTp^uckrm{T3QI^ zD+5?U@kV1!)v>>?D3zjiPE3oOu33drg=|Def@FIGqy0+z^4X3W_K58}q!}83EK28b zt%QtdF&FW-84kaCafsfsjq6a3Zz`+^imqcvWuP zt!yFJF>W1ave(CRpg2#nbPvFI^9A28y&`c(FVvf02?O~X;-{3P#wrPt$&Zg;I3hlu z&hOz7l_9cJE4(h>I}{!ev#Sf(6xJ%O+{K=a0Ok~$(KFnBcn=u1`;_d&mqdo@Q(@Dv z%U|@srz0_X(&<~fGC1)xAe~FX(GkN_6?A#p>M0*3t%kh$8Fek@T$TB)2ewz&&nvkb zCS>2CyYpAOBj+Cu+COkszsS7oq-I1}zQHEU=?JDCaIr^;hB}yjwHS;NzLiZtSQpKEeO5I8 ziNJRK;?#*$NRyz0`;_n%_x2lldzgn0{YUa6n@*io`PqpHZL0kvHj${uqL*3(bL(bb zO4Bmy(k$NOXHcn`oGS~2EmI+wn4RO^rw4pb)*GXtri^>^SCVAA4nvZ9;^gg(T_Ws# z+N}4B)#sFj46}uI`CRT1W|`FIrIDT3J_gfc{(R29X&kXUWZ`*G7C5M9+)>x9uW@kh zok0sb=6iq(*%tr&cNAj(*W8tV6mtImC;9ugQlOl_y)*llkTVkp$3O0G{*wHyOdy5_ zzaI8Qe#uO_ha$IRrqI&#Qz8Kz!xh9+*%&>`NoZ6qenr)JRUgMs^Y(H(8Q;e6U!%-X z;oED=+hY%LDfE@!H_S`rnP-PXa2d3koh~+e{_aazxRcUdSCLy;R@3=kMufSh?tJhZ z4#lzkc9zz-d`FDIpI+HLRr zrof6ucX<~DTGCV>u3WJzSEzrquFoGat}Fu!%jBQRe%zYH+z@UF2l=6sH z6@O;5a)uTQ)qf?SvV<1boXfgWxU$Uu8H2wN{tsp;TaDVr0QLwA#?0~66Y9qF%TjO^ zv=@c{j>^vgb3NZ6L$S&STDZwEA?>{2?jqrv9KLDK3$@_6hkK=2$>TO7O>)=j~S>*B<@;t+C&96q{w$jjJ?P(2p@?H==)m#l8h;<6ajn;FV7=C!jO zwyQ61PIT)Y%PdbaF31?PXwQDsB;OrLYKb^$p5$hLB%<21cV|kjIGhR?z_>5<8?~B= zgXWxTh-A`^FAn?f&qXo`O5Tw;7`i2{a`STVA4|G&dUSDYoj7atzWZ;*^Dja$lRwv| zWxXey(mP-S<^P5%B;jT;S~W9nXq;R9N>!zt=MA1Ete--Z(kk6piQfLgx-KBU8UXt7Jl>TF>QaB9k+6Z z|1vk9WLe6~rX`ww^9Spasc*KGdA*|)sq7r!mf>$4 zhV2}8z*h7J63TZk^DB^P)i}E!pSUHi6|n@kF_kp{vh@$Ny5@je%Hhgn2~^h8M<%Nr z7YV07e&^+d48!h5Ox$#woPx~n44i~`Jg7tFb|p8=tx8*+est2dK{jVrF38II9G_n# z^Is7AQ?(GsDLc4iloW3!5wU5_4i6=Xyu{+Tqb?qT?_`V2KAjZT%xG$HcC8k5T$_fB zlr+ZG)HV*i-g|WC_H;e0Uhkb8Ra0Nt;bsR|g1}R!$ya8EB_*eqnc%+Uf!;EC_nL+w zHcOD%jW+$M?U#3LdRtvMj;g;!`*w#+T09-kPTw8)By>%feX8{+NjvUY8Zj#&5d}1B zJUyLQE7jN#L&BUzk34hoM8}HEct);`-7J`|BHFt9gaWHdbu9+XuMYCAoa!SuItQv4 zU0R7s9i~pImTJ#-cQ<=hx3yg8M%|vTr}iK2clC;ZIB`FlGsl-&w%?+;bq215L3#pC zo2?{mHskzG)4qgA>T4aCkmSXe2A^zlG>0n`IV$Vfgb!3z;PufLvhKtkniqNUq}>-P z({wSYOx?uAhm2$jFUYyxTeww!)SRqV?&NJKDP(-~ux^AjUPww^nU%JBv=Wi{@r-~H4ex8 z&*O0PAr+{wOo*@rAwDL)IO6%QG$my)XQjIDf>=1pBXYFqJIvrSkuZ@irp2Srj6Ue5 z5iC{jSO=c-N$iIe{d6&<)32=XFSXJO)79c(?c~Er8{_b=q>l=cNnakyGMni|))aWh zao&3N>7`^H0(BQW)Rhgk4x)!4mMO;HqVqTXOITkLk^X5cZ6&V7YKalc-w^;OhMbQl zKOMmaCq99a%56fO^rqU42s2W|I~yZgR(mob+>+EJv{TSq2CFo@|B8sX%W-9MxAl|6 z&|+i0L><{sbWrYxu=UWPpZ4aLzWi7|#(<_Wj=N0$s9+y9dGzqivjus17UPE4XzrS1 z`1ZZG39H>vvRyToH2iVX-j19s$=4Onaf&{kn<-hRs!#MX^wPeEx(NE;=nBM)jnSPe z<6n0y7B<57Obl{KO3eLA$DIfjLuYw{{M;e&(Wp@KY>YDCqeC^%`K%Ca=-k5v#xc}! zwE5L?xk#Z*Oh`_Wnu21Sq*`g zW(}^VuxGHIu6C}nUl<$DHKCqU9BDbeX1kB086?ULNdT4j7|-Rln&GAU?IsEiT7JUH zViwz#-=$#5Es``IrY**2x>?op3vhPeJlE?8eW*!(rd%0$Y{&dl*=q5?B1GZz;Z)1Rj= zV)W-%(wEZmtbgP%{Ld*2)eVMDMAYOwWl$>OtZ@egkx+?Vtm@XsvXp zeW$~0E|igrsOUFNX6CnI=}}cVsU`VQERpZpH!Ps#7;WIWxL+mW<2S7LLtzN0r<04D zA~W7X^+TtIDe@hAOG8O!d3t50#;~MTL}7}fREeg(!0X&u?#z7mvr>uE-gpag%fK+r!kJTbZpgoToP zZEWmn1rN``!Qoc;6MQ6)(I;|u)@JxDtD5h&NMlH@@}0;B@nJrfs>ZyG#{2q7`MYdNo%yt$`p zZZDZ^pVqKhDu38qf6%>Nqjz}QSq^|Im=iNKG8JX{G2beXm z7uyzl`yJ;11xrKoJ)@(WHhC5)b%m<6;^fA8miz@Z?s}-4&yX=33ZEx(76fh>JZJ#~ zsPUCdzT2lGihx-?i_VrESVOu(!x$qlaL63%i5_oxYfei^; zPvnHkjsGy3PKIEKRqvT8wUD#`UH?o8uXJfW1EB}(0Fq;otA)RuSd{QV%oAIXt3L;X* zm?54Oj|>g5Rc}r9{VhgHywkBC#IkQ)NmPB|aqRb1p82bw?4^s3Az3gZnL%;%;{!9(oDT1G%Tyc#_yndyUFQ54v7{8QV1>=BUSHR%;K}g>~`Sz1^ zBk7=nee0o6^dN$Sp-}J_fIy^n(hCG87(5CHDGrqHGzskB$0r2u%K@-py4xj0FeM!& z*b|*9_mTuQxZo0@!sq_cR{&h_+pmW&*MYsB=(tqAwt{pSH$smsUo5F3P+}aj)7)^G!kk^j;=a=jC-!i_eD8(W=jIf+(tB zI6+pB=BIrAWI!a6;s{XAf>c?3ocLY}&20Wu)*-TL#j`+gW(o4u_r4N=TWt~CLxq|D zE&u=mi+&Qw^JHK1wD92AS=ZY`X*?TFfnZcYfb$n^vtB_+-Tm5bOy13@&(-Jepfw?z z0jLJ21_#f1l5#OJv!XIMp8imVk1<}T(^Xae8h?8eg6`GCiiy)e& z=dXTzE*`bUBSxQ5j}MWA*LP>KE|55{GIB zktifKckR($_cfzhM9vN^zoBnf!zVxby$<-8lQdLZ_LvusiIQ`aP(F$JDMoF~r6!}j z6IadPgZ4^K(>TcBRDT@GKGqxlxlyt>0hbc)49PfwNtg6?BQ^=z&uz!CbVmI?Vfj_x zV!a~NyihqWH`{EObS(K1=phT6bU`oJJnDt70-VhhcKgx+=@91ikXeMd2+2@UIteqXY{<9#A~9 zj>|b(6HnXc6T#lS+0!5mmp!~Q&L81B6FpL4Z#c&W{K`(tBW+h>AHaO53k)f+VP9kO z;cng@GviBrmCFN~@OmKEgyifs#gA^bF?p)AiRCmLAFUvSf5zG^ZsghV9Pmac2hXqZ zy?3vcw3J3^Ce8dcA^KSE)U#Y4xxs|ZJU4zX_cpRMpL>2UHsA>)e;e59+sMxq6{>Qh z%Z=_*foS&lTv^rP*&oe*bsnN&qaubJzfweT>0&)I_GNd!;)EM3n4UgNKf`+{&6PUC zmtp)p@97jj&2404<5qWsB|Eg&0H0s{WiaMujuv>2{WEv$SQ1XI+&@DMY)3SOr29`W1NelC z(o|942qZNFPQ=!GYLrUL_F%@fYFQSMb#2i%h3(_S?M#ETw&&KbvIwdJj0?ILCWkn|uJp^RT`8MzMr>HPs)+0bAkX>JsdF$p4gL?Y&@! zVEIZhBru&cG}%TXppbwai&D4akSVol?u2@y7TJ%9m?mvO(zyX#shhxyv=Y^WTuh(w z7%jzYUW7aGc2J}R2mTRvWubf5VAel%E`m0-LxhnzX?R@|*gD%-H(bFAr#d`nzB#N9 zK+3^mFgITx;n@7pZR{dgkD=Ao9<$aN7n9;cl(*)7mPbyvA}S~beJ@_v*1y^s8B|Tk z+2(hAZqpIdiH@~#}+*R05r1Bm5Axi!t7KwXu9ydG1#L)I!a;WXx zYe_Q2WlW!=wm+1`)P84rZJWn~>qxVPv`od0OJAEp@gbV^byCn#Taldj!Kb{i(D1~V zvQgS;Q~ofqRpp_xgj`Ph<^i-XRQhTKItwxRJ3UAa4>l}%tT=5wZnH9!LM!Q=cO#-A z(s0)0w6aGxM$ZSGA|7tB6t%NWp&A5$RC{vcmIWKnNqdO#H-2m-xkXvj8nd+s7>WbU z1~w+ZXVJ&V*KpAN%0!WPZIhoE7C38=a)Lz@IyTOz0g9vfMxUP<<3RcAi)nrlUCmFE znwts$-0sLQ1|&xQZ|DnxBGKC==X}S)2WYK&_9G6e?Qov&^fhf(5|zhz>_;ftwK*OY zt!7K~OI=Rg{jRgx9!%P;va=#<0yS2#GQo_2MEt71c%4eBD8&Ic-N$w*^EO_b@%iWt zWi-%=?v@Al38ID_#U{$`a0_YXN?h7?QI2$Yos8y#4%t64RzBIt=8T*s&jYjhv!@sd z=rP9QFE)QJ+J886_c|6WVjdi!*#T1H4r?o-lk*ZA)7cX~?n9?}Kv zdH!s+Wq&plVtJ^h;k@E>;d3kbv~5YnqjJxO4X3DT+mgUqgrtUXlu31o07GyjZU49~Cic8&2qms6 zDm+>XSP6j5bu}PH9kgX(#Lt+FNSsrhK8Q^vQ?zw9981SNa+*&#u6G6Z7K%bWnU@4P z=EPEzj(P(d#Ud0sBhajR z_@{^sS6>vPY22%7+6drEp?}&-3FN!C9HWLhYcg1=t;xKh{H#+9lx*QG770>o^{e&I z|N1%Mrq!s=h7Z4KtFoi;R=0W5hrMlxl2wWlCxjx-bP^-!qyf3XiV$|PGwcaUdrr3k zPe=}_>GN%kK}KOZ)MlX6m-#Vz*4%&p2cNoi{~#g6>#? z#0(z-C@wXD+8e0osjXyPVM8U~F5E0T1YlGlRynhDJ$kv=NzK{v@qsK0S%fnP zy^%5YSafP@3|yHsK^^$shJt`!M$?&HMl)wi66cLs`4O27mXNFmB>)KH@`4&4sbm4sx#TF zod!WI3+v)E=(KQeYDtF&e|loif5Yz(#~K~XdLF{O(9EP1lm#>~y_Dko#z+16Ovw^< zejp>`qsBg|MB9ls;N)%Vn;!GSTenNy`KX{-XZC^fT6g}Gay*8p`*2gwmlvr$SKXiY zprec4j-9Dpx`^+vr`AhJxdse~(M! z?=(DYQjA{j?`yr9l3*L_W1$neMwF>ybw%aUbb0jUB4sRmp1>cy6PpQxVNfQEPI2U; zIk0SI{}pXA&bs@(nP^dqbbqk}vsPX}Q2fLvBHD#8>#&&H!Id9<0fcfS)1ftaB#)0Pv#!I7orjUbzn9MFlV zY#{AyQ{%V_3wAHw#G9ia4ynW%y{r-Ba)dKknmd@9?M>mPE8BRgQ@Fd$LbjY{7F--P zHayb&95*mZL12VJawUJOLf~=lhnlD`oTLk_?MHcCftCkMb3r1gl($8*qS+0&U_{G z2j699jssV(Pkd!E-c&-wnfYd@{xJ9&+WlFpSsjrNe%4tkjl@S09VRN5A7PuRE zLCz<=;M66+krNx=j@N@6ALW8ypNwr32n?F9tgKbDGRG;vhG0rhU&%-{G9yAUIB2+V zzEunN%?nJ(iy6hqI$$(7Vo9-vY z{(KTpoKGUFQJqapimN}Hhs9Smzps9(waGo6hOw0(lbxcxZ|I27 z)7{?=hpo40wM5`$=MykKGeq})<+YQbzNJfGq2)fCK~Hhru)zZgEL7sTtlQ%p86^#( zPIc<}<(_Y=0{E$JGXfjM^-Q$dPb}hDtnsm2LhIM*P^eS$9PyKdIdS*?x+?9v-O%vD zq+-m%5)?ES91YbV6CjM5nb|t*>*%)VESEJ?7U$Y&FN?$PoSv0G9bz$4ev)@g__zjr zEAUni7@nEFrmd==WULM)-uVUSmDp*>g*pWl@h4YIf{W1Kz3*T!4=ntg>sO7jKBjg8 zAC!Fa5%jK+6W>p^C0X$XWkYQPJT6xeQv5k0B&8_CvKj?{{jJeu&$Uc7bCv&RKxsXZ z1}EC_eMO0M6ef^$TV-H&g8i0lw~-tl-#j>Nv{ksBOqm6&Lr7iVI&4u@*<9W3!KkdE zbO!c&)!RxA-OEWNuphBU+v=*R)L?d)_|wZUu209wBA*bl;xAAstCHWrkGSC085|)$ z*9h+0h@81|O9_ zgrRTibZ_eqVyI|PthzfCXxQ04)7|yZ@x96C zf?~`I@&c|4T-pfC-(_0yA7V(>5|!{*Dm6~FLbBbs2wo>rcAzq-G$znb`TcGAQkLBMm5$!m)kwa3h`w23(m3R7XL zegf|q%Z^>*Y%TV=M=SnEToy5QFaEm#Ft2Bu_*sA)F*a(gF?=;ivTAKYLLxb5W>8x( zZi*FS)_Y;Z^C0wk+UmG&{@o|K()%`#lVT_B(yn*SvBPjC^@T}}W3JN;?$k2e#?pu4 zrTq1W?}IWy2TaEwOI}-YmDIMYd2GEaBo_QsFirKCLcl7w8XD z=c!<#*pS^=k`T9<h`=)|{ zWB~;DYLI9FM$r?NO`SG$`mVEqZTU`z)Xg#aD!V&cL=-TQA%UdM(8e)qfWWmT=<|>~JjjyBEc=_{Q^z zxIyQwU5Y@m>ZGgn29b6@RyqOY$oWRfNeIdyD}QgD+tS@G5za_7@IF1PM`XC0<0%~a zEQCGDmgM~1B2QM#mma*@J1SzhI+Bodi{;KPj=RHV`YotILt&vv3v0K0Wnf-tJ72q8;*o}&Vx0q>}N4emSl;uM9x|8 z)~J(JQiq_}0n5$}SV_g^NaN&?)_LWfCY$5$^4cMY-E|4XSt*3ZmfMN^F<(uJ2Bv#^^_d7#RBNR>c;sIm1FNbfVg1Jz6xG6uU4HZ=n zi@C1Yq;9QFZ4z4m$>s#M6E?|u8~w;vsICzG=aOL8+`dVy?KifC#cj0%$@gDEHW5@_ zGHrB8%ig{Y8@X~~!o8Pwk9%e7ZMCIj1y4kAhK~_caWNbjCCS!+ z@u^4V@Td-0_g&PGM>@XQn6p#-#LgKny^oUwo7MyM_ZCj~-OWv$60p5;S{bqf6(I}R zhiqVr#pe_6Dj{(MA<>nH6`mKjQ79NZAEIGIzKi8mJ^1Cwiv7@txa4BP^;EsDJ>A^R z5H&FnK3b4(;FopSD1qJo13Ey(zs=G!Gw~OaB0pz+PJy58Ha)}2Bh{RKA7!!Sz70Ge z4nmwLhL{iENKedbtLVdnPJLw#3?l0~WF9^Io&?XrLj(v|w_fg9zd$huh5SiM4X?Hi zAmXqSA16gI=41Ah4bH>Wxx$Y#g@auqLI+@@UOMU0l`Qn7WW)Iziazq~;VlEkx0tGx z)IxGSEK`Bpkmw6CDZVxXpqFl~p1ZV~ZS0Px8ag(D49e-$sOxZvsS!yRVpL<4Al`dJ z?MnKW9%U&q-*3E<3_rt4EfTm~_{G29q#`$}en~7cJ$B1&L}M}( zr=r;r^#i=l$SwIVa-?}Bg84K4+G%3J_hTm)$C$|qV5_)O^+cW&2I#ri{1i}=g|dir z**fX#^+-zh+uzXXts7sGZkAbAJdkg?N%DAU@GZ|}c|iyCW|*f2d&NBVwTy zY!;ErS($R6Im7CVc+={ZGf?Yn(MISQ?z7LXv~-*P@SBQNZwrH_X>#OfZ-Dy5e&oxQ z;W@Eisq^a>(nwI&a^i<-*@tJjX1HrRDygo6gr%j*yk_J%=Yr_Z`cPqc-w!2<%0dw> zC7?#%m;?(Z>pOG=v#-z@XdD*4{i=ocy0Dd~BxwUYG_m3GwZ1ksBXHlZmle!vZ{Cmiahf0)N{2>p#WrePjA;<4O!mYQ&s6?DU0Kqv!k@fKMK_>v&4s2{N@ zRN$y1P=5@|!ou=k#T4^Fl>k2^9n94l&s+7HM0m?$$LI-W@fsz;JXVl?m`~7@ITo|^ zGa=_o^R+~mK|_XSii~WbY>n7G_iZK%T*G=OFOz2X{P|~_T&kIe zH8Q%>Q-XQ5QC!j3S1i2Hn+J>)Bjae~mhJQ2y^jM^A1&-q`TLg*9{^2rDi?remS80hFrj*ecz9f?F=-|=ykS6I5!ZvOluw`99K~zr zC}5=lHJ0t4_O<~OBW|W1OVP6R3oege4w09@RdPbs3Stt0a6CHo8(E++{!!p};=-tq zp&J-lnf#(?@HUZ%Z-}Z>+WN9LTq$lF5y||s&i(cIv~Dd<|Fu$eR%!E_U(MCA`_#e< znzMjaU#f9VFhL-$(VJh}_{a}e%ETBO#L$}nmgKxS+GPe?l&~Tv_YU2*U0J03>cH6* zqiGxARudWAYjlO_0@)ze?MXeRRXoOC+JZ;Gv^r(4t@%fO+S{=#u6@*=;DR4|ry`JE z4SNPI1V4Xnhnc3qCm1{#rw$+gzU_XxA5h8__i%f4+!Km!l8ffQ-y?-}98l3_!;O>-XdZI+)*0JK;UA46vDu9|8lx4*= zQPcbr3QH1w%fS}P=FXf3Pz39?7%%T8vc21F?%*SC~3ZP(_H99VB(l+*Oe#oTR$ z2qTAPR1f8Pnd$h&#M^`R9AiKH+>>%dUD?*WUT-e_bV#vx*s@A=WWp46dN+LMT$8?O zePix-K8O~!xASG90A<-qNw9CdG)uhjWgP|&(Zx#Fo9jT&Gke4!o>pRA&a|XYTnT!j zashmVJjY(!7tX_q^K)U`sE~e8jN=PmEvoAB=Xd{i_kV7;h*r3cx;O$aRVp)IS99F; zfwnr%zW%88DbqY)KyevM(u~#6pnlCi{(4;8(gF#;S>7dxt_7XGC=rrSe<%D}66K~1 z^A{DmuY9CKo`r(V!FtP{=NfYDuMGm8s1&V5f6Sw6%Xgb`!r()6iGJHwhlE6emE`av zI*~f8DOAD3uYw4p)fE6YR|Xk}P+CY%IWA-Tsb9)tiBSRDoPYo|wLB_q=(mLQnzsEtO;9YUHDylBJRuRmD zkY&AHq&)foK*##*!A13*Pbk&uVMbr!LO_obK&@C>_s0L01@Cj~NGv&YgSN!pb_H#a zo?00W1UnQ7Thm=MMY79L6kjD!nbVu{Use8&S9CyFk?tYvlFNsJ+mb%Mb_8d;w3`q= z4k_tBB^`u{MuTW71Xe5Cxspb|OA=e$A#}z*OV3c)A$)+yTww#m>X%jR+B?x*01DtR zi(hisODfpCBCZxp@s#TnuUp>6r1|i>cAzMh;+@)E&jdM7R;laN%IbV=>in@(`@ka& z&qqVpw36g9D@I8Y*KGPUMt6g)x8U~U*8G`XiuzaA?a0xlKyiwno$hC>nf#mA9E)15 zXInu{TTCCFiCv`rSLI*tJG}IX#mjSo{J5^rm)TZxocoUCqu)g+QQSP=5AjlgO-Mhj zzjv~ahSex@?{@r%0);gel(TXo_X_qASK@LmOup+IjK(^nNdo!`YFS(xgz4q`JRVd` z9f|H^h#SnW7$Zpjq~W+7ljZR~6-4Ikf};zK>KGNjZn5uQuuC8Z%;M~V`S%n)LM|=0i^acTZ>r%T0TcoO+jx#!ezn@Ct+(f11MK4bWD zj#FUcxOp&&eGMXIpFU+>Ib7~mWKm%Y3j8S8LwrI!r%E;JpfgI;S5Z(DOa`{stM#YCnm7_01GvfQUM)r_%AWr-7P>|Wya4b74u>p~^{Nk1-t$QfL_zN0PEq z+TQR~u(Gd@NM;|K2y`8TF0{duj&lGM(lUC8x6FxWf15afpso-;oN_m{K1f;Wp~h8(Ftz$ADp_8`U92y1`lyuh-{ld z=AT1H1T6Kn+AvnD!p)g61Z#w3#(Y%*Dv~{+;1;KRPHtfl9@_oh)5EQjboaqvs4@~U zVTP$4#yqFKVk1HyjT{}b#C1m_A<>GiQ7Rf~rR+NKkX)J=kEmCB%bCxPgso&6tz3Ys zGwJ%{IMiFX6XSIIoGL|G+f9Jj)ylW}TWQ|FDS4BuNY=P=h&Pu_H~YJV4IO3(?*7i1 z@Q^%p2UNgP`-j1@8@hR3rT3l>Lsl5E5TBm?AOdUN;A}O1el=YH7l2Q|&rf|t+@LKB zrWJYIO2;hU=GZfP_7IsV+!CH45ubd(TchosfC)&*fx2sdE2_P>xwX+wyK}TH4Tu|G3;K zCVf(+5>Uo+v~&8&U6n1cdg*DE%1|x1t9I(AWZZTLfgCZ;Bdhr+q-pn4=>5mV;d28p z9sKH`!yKMgtDeal*8bY!A&T%5sOpU<@`@J+Wq$4Vyu9QpU2kO$=?s^Z7u0r1MCqvt z5A_m&w>>C@*4jMornL5ffC5rT#c46_554cS*)Pw&yVQ_r(YG81=Ey(o4t<@Px|YGI z;j3pGY$}1Y;c=2~A@-<^uSP^6l@STK zRgzl;-{(2uAn%%=|Ngdx5yrRgB2+e7I1wCJCA06ZG3K+WyhIl|lWxVU=12dCL;jh} zz%y+kGqG4~E?e*$&eJIs(bwlC3vY!0Wclavhl4{f*^inG5+v6cyi>kNmYQTTHYua~ z3Iax4EJdzEwJtGIH<)^&F^UFGMF4AKyhcmZSxx(q5ZH##+igsjd*GAU#eBQRd=lP} zELBw*Uu0!I<*2iY!EE*osHA2JB1o|j7uFjNYBQDNa!pY>LF4d%&x9~VE(|pHkCSmF zuC0C*g~cMiVb_i{oso>2SEprk=cwoqo`5@w^Llq3;D$Ebd0QF$ofFZa_bkxF-Zygd zs;XE!x)t` zO}%7;^@nyL>ZRL4JRYouRA%GUpzGwmz_aoBc@LM$x0cv3977h^%J4OJm;Ll& z5Np!z1?G^v(nA(*Zb&ZA0QrXvg^IN6G+AhMWK&>AG?sS2GLcoMU$Mx~LXTt;SdZld z=9lIC!TF-m_a;4d+kK76Nm8AQLS}Wj5F?C87Fe_v4)y@bi^>MwuqvXVssc4SC^iQh za`!w{R0;5ys~u>?_z0O10OCbrt;X>EC;#POanB?*Eh8a0|CcY}69pYhQO|lXC_o1f_}-$4(0Wvu{>R@$sN_KCoUTbzMN-ce`$h z{Rohjne!Xshfq`>Z*RL@jR0l~*S!nj8C$4jvXS25$7TwgTaJ;{e|k^0%<|(d^u(+7 zGjWhI)ZHt7!NNG*Yja@#uyzIut2y#(2;KG(@@pd534d;gGAs_$G71sTp;xeV=DAta zpl+>35!S;*7Ki`&EY04^<`_Oh{we>6w*Yu#MFnS3g4Rsz(fl@wb6mjY2CeU1!=NxF^hx2&rYa9JP|fPI z$FGE>onSMUd<^zS6VPpKI(XRPlCXk7V(8GdsV0d{MrN0&tmGp2b;S zTl!{wn>N8I#Yt{o8%#XbO)gR1_m`!U(*iH;y5Ju)IR$k5FVnVHNlW;n`gN8%&H%AsOb!r$`>a{Yv5JtD^hl}pM zdJl9p$huyKcivgQoONi$Z?<)P793MCE}W`#Y=%uyKFp2v|DskvLHYi1Pk7b51wk(u z@4*(8`Z>u7Sm(2f7t)^4FqpYj@o?PuE9M&`^=rNVY_0EII zGCy1P)7j*(Sz7&VSUYz@Yuv@soF>GDGxf`CB?!cKpk=(;* zy-seTdmMbgV5;=b7K6af{3Z)`M-twZbo1Cyy=_<-ma9R&@&CLH=W~M}l`qQY^BB?unY09*ZR75b3y*%f3fho+GwKeV`>3p}#<}8amb-+^_ z>RBm(4@zMzC79GE&Zfb-a4Se~SM(KHMsg<+MuWn;7Y-S9Vo3*l*w)%A~sd&tM$Pcj)^+7v zzF)F8$f^89ksN4!iLNEsu{fgaIzm}d)iU;>F&r+~&5$*jnE}9HxUU9A!TS2Ti;{ja zwPBrFWV5qnS2N#vtx@#ZybV`LU$vm5t#tb`AE6Y?j#4IP4$NY@y?@vk)7RJMTQe>gj&`AVozoMetLkGHYoiZueq3Q|XNzc6CX^B#-|N zThA~;g#M-#SYtxcV=!8lsLj0X#4!wU96aK*I1GB=Mf_8w30?kf2gRd=b#pqE0keW7 zM5(MFWgfs3LfSFj1*s##4>iquo4OyLKgwF^3u+_|)4Eb^y3xj-iWLGaG1ru> zI94)7%a({?tJrh3@5{K!u|fij-Nx-l2on3!eyuDzp8+-JM>1An+t@=X{g!l~mVR8EqRRew@pK|asM1bG7oLP$OfT=27bYR?;& zi3u~wIfm?YcuFlo|I3%wkj{BmH67VS6Myal4;_*k^>RZ+O1n;_51cWv+E^YNUOJ)( zq)bsHSnp=d+QxcucGboQf0&rQp*P5cUYe3jH=sf25Li_%*7|q3tulO!^2I+Z+5rvM%?^Y3WSs5(|J=OBgO57@WQcj5On!EOl*6>*&K3*2|?;pExnkt+j66(QB&Kro;FiJiR%d%qe zQIGe~zZOqtwAZ6K%?-c-+yxibk%PEHEqbSj9OM|~Dt6{1oFt8v{D2TudKsfn84WU> z%W3?2q8}e*BKAWtUSpKNAFYs|7dm{upKVo)1;MrC95O1=d{w@ct%0vL>nSx}sP{!i zRV_P3Ozs*$9dq&VQYu?@L{0n_6P23$b(r-{F!830h&R+svPFZlXatwwh{eQ{E6SnU zbL*aL$I_^xjC>YB+ubrAYmy4hgzkq?)A zDJm7$c?uIZ;j5)jt9mX(XJvuhw!@-Gc+WnomFs1(cvzM5p>Z_wdZlF9xV?l9OLfzj z%w{#)xE2xQwVSb=_pbMase32|oCZ4{I>CyTxLyT#%QMW2yBf|y@^&lMYyYffS>CI_ zO?n@>m>?TMmJrlW^PO@5l>SAqk^7Dy61?E0?pzV^b8k#p)|Rz?R-<{2@Zbe`8<(e|5Azm&mP)5sx|Eal8R{c@;jC1SQV>gJeXhKL1Q>48~buCmsA zt1AmEng|2#SP+=DE=hYYD#~k4&=}M&z`O^8wr5dByCD3|Kj7oje!09CIN+gqT;F zjM(rQD?3mfjuiMsT|ePRHaZ(i%o>HS5E$fWp>?q?oW{EPMv$x@EL)j5tCea$!oGH zbg>hqhbny=C400{iEF{iQ#XS(5epKk%a@3)jaSCX!$cbU>Ykl^=}LD>LEqrx(gYkKUhczekMM<^pA$B+Zhu+Cy)b3)vUUU!7yT-ewi{Lp^=C{ak@X$laA1 z(J9E%aStaZ_(((ANJXg)eR6EnjxMNHufjZVT{D6#ClfTI2cKp8z+hc=?y!iJFZwPh zbOS3bq>Kj&cgKMx^W3z(i@2sn+5KBsYZuPWLsiucQ2r7^O5&Q3mn}lU_eIv0L?UwD zI#l;Ij4Sn1Qza%t9<}Aw*7Q3#ikG6IcsTTq{?!{n`@md?Y zhOm$es=uVaqdAWo0&ib9JGYz0B|@FM-^fC6CGgQzp{m%n8Kp7pc~3xNEb&n${d zS#!`st@5#ff7jyjr+b!s%8IQGlS6SPov=|r0#a{08f{p(anr<>R9^sN`v8W-7`&w>PjZhg_SPTVcr!v*zP9XsEZA zJ6jq88;2Oh2^)6g= z8-Rm1M=N^E>b321DHqeB+M%gR(3|xR1`_gXh z3l^l62i-ZC@|n=E&&v@(8s+(1ao#N74cBCy35!tRRMe71wGxm*+1%BIeELCGVwBz4M|AgSQD8wc3{Nx3Lt+_yU>`>gG79ojcpODppHQPf>-4GFi2N2TH!Drwdp9`jR2+O4tFSJ+ zvvYzmI-1(0uC#%yw|VBq`|_D9i}XBlL|-nN2n&5eZiQ7rLlw&5+w;))!`EJ9=L#UY zrjNae0xLYvo}Fb-tc+yjv)NKFK z3wagnk2w&*oi3W`?qe+`xb9hkb*|;T7U$b8X>=XOVA5WI>2$SBr_?@`oQWeuDbU)&K}__R%;jE4u|JUFYtAm1Uo2S{Z2-1WJe z@xOn&D{gHjVV5aVr4^@|!Gc9e!E(91&?WFsD89|{VsXbNg;`jmkh}|$VuP&t@J=k8 zQ9B@-uF4tf5q=X{9jOG5i#Yir9Os^dlmy;H%#A>R|1~E%zRJR_Q#$WtnNVu;*|R+J z@x^4)1r}#)$o*^>7ha$RB4l7_u?xB4izsj7tMI~uhLb_RNvXj1RbzW)?t0b`PVmW5 zpVAN5?kj~zKUl|+Hz11MU)37h@z{-TVV%0T2fgR>qdZG7GDyVOd@oYxoKDv|FWe

nD4OcIV*t!)q5O=9XVcCB-^I$m+a+xC7l9s?BGb@axSDypC#Y@{Lrt zC?tK*V8N{)mQsr%GzZZYiOG9vI{+&*y!uGCZ;J8KbYm=erD4Crs2KrAe$U<0czlE| zr{hrdxFTb6YRYizsUU+XJ1<5WPWs8?@Sp~BSYkSOgBHF%6l#P0Pi(f35$CUuQ=Ap3 z;HU8q2A9$i(HK3pd_hEtSlWA=?o|HQ`2)u5KPeEiVR{&2umucbZ{CC*{5w-K(j9>5 zzS2}C&NN^`SqrkCk81tbg)Zp_Bo<;Twfn1o9!pPIP1ySHN5n^bZp-8C?O}yf4)yFn z=m>N@(@-Z%WPBVhqp4nKRIE`>U8aZjuZ($Fyu2lwS+&q+0iY~4v)yl99^FxyovcoD zzn|ipAq7jxb~Gr*XfGZpYuC}JfEDqOzvSKHui=1z zF67nbIjF`)rlVQj^LN`d&LK?G3z6Z)gkc|Q4U!f-K2XQEdNrCBU1WQL!F~+S zg;gc={iVk(!Zib92YcAXc{P% zWz_zVIwswVZ{ihI_O(>8G8>n*q;B$zP;toG%Z9R;mgPi{uc=STuQnqNrF;9yqqaWD zWXakEN0if3Oqbl%_)AR9bEA@^` zCiz=qrCK$dKmbw=6D7Zbl{dr3mdkG*yuC1O_m({l|GNAnMY0w!_D7oWY9JR^^G21htujYCis;Ib(nY!AnjAU;J( z>N)1WF><{C#wsOBl;cGvfBf0qO|N{yBwQQ&L;J7*c$=D%1Q$kP8!E1VP|TXF9axd?jM`8MyTgrnO!OL-Igb zr2y_!S?0mx@_0@f?m9)3O}b~erDx+w(wGvr0R5R7ODgU)(X5@aCUZ3+Nk3prEtLnh zNj7sW1;fdh@5eWk+)jLhZCClB%WBjg_Z;fgUj!i&-IUfXm};u|1}_*q=uvSGUnwek z>6ZBuw(+}PNG>_g>rT&ixNTzO(+9FD@CE3E(%tYqjpS7p_Ht@sqg{ZA8i_9OwI`_r zThX@ z#D{lPxt?P3UU|x2co)yOtR-5VW`w`aD)AD@LYdZ-qK`FFFM`1oJ%sEx>ZKSWV0gMz ziW!l9h8#H!2{FN*L6duPeOwR&XNm@h4X$Cm=eEr}|)c5HqW%2FspZIF!vo{QV8_Zaz6fV;ll84N?82qi* z`4PvLOT7VPaN@;;xEIFT`e%%4fSdiCgn%yyPN^`*Runzt*1i_L0*EKD&OHR(fVfz>q+wuyoKykGT#m+0n ziB1>~K6H@wNmWqrdMj~+_S-JFI|VBodiv!XVA#7>WR+G4UW@Fu(Fyk^v=X>I5Y06K z!*ac!v>uBn&A|e+_E`UbEla+hI%nL-EWpx;vNi^dA&ppkjf&!_nGJchzxt7<%_bG5 z>JF>!d+}rho{}~KxV`qamCiyvj{@TD@JC#6<+C(c6yQqgIl3SPrrO>Ign84!T#u+t zf&i$yPi)-lr?d*6h?#qWN(dtEE_Dh#F2j;pjK05a7s<(BA*1y(RJ+eD=D<`?kHS4L zO+sw_(sZE_*y>IeMI%AvcwVm+QeFQ*fh8QCS5f@QaLXoC*mpxBGtKQi9rK6W@1H|C zfAziSKyJT*oh(&GG5Brfl2?AFoYxVYHCTmCL1EP3j`<9BU_O_dd8a{u)egTQo9?aU zWeKL@y0;{V^~R;n=zJFXrZP(C!?5k8RW614|Fv-rOQHaX4sF}EZQHhOo40M-wr$(C zZQFMDo6S_c#r#2*NhK%jdU)yFs~c*~@akDi?@-Kt&8}Y(js&H8ldn$=3u1oj!5r;* zM$OPCGbu!OtO>vI79fL&we%#~JpDU)hGl)KSNB{=D1^c^`x2dVs_!m527&N4D7g4=cS-v8|Sxvr~4p`vzLjoBuH<{JCRY_D<;h zk8JGbl9XPLM)qv8t_&@W3BB8tk5kkm*uP!-I3n92f`9xWU~aSM_F& z{?3PnGwsU_M~S%#mxcUsCMVQ?F9=2KjGb8pAR1qH08+#wLk)@gm`-e^o=!WwY=M3 zp#b3W7NuQS!^%fVBgcUqj&HGb{PGw9AW$_{T_R$%9kr;{Dh2OGHt|B8Hr@=(&%`>A zQG5X8Qy~ja)>s2R9i+_NCKIyePbCfY7Uw^u9OhUe0jm0B?2V954jg!^N%joLNG5+| z4z48}|yb$eJQ5{SyJn?HO~Y z=%kr1Jqoc&HX}dxN=$Tw-!26%q_C%7mNWR=YWTIW&cqpDO40u|gBd#oVNWDl$*mMd z<)_OfIE8-Ib4;9ZLM67(upVk!cNRPO#d~BP#cZ~_eNMC zU?~_`31!~Ru(JfVrC5OXX{p;#`x$bKEyPXmw%hhW1@Dweiz)0(Bo&`!+X%nki7-)D zuq<`#?2g?&b#y);7lEn99aFI*eyj{|B6^w~rKvbr)@^rFKljI-BEL#pgGw<@d&*sw z73L#1bkshSj2t7Q_CooEjmB+z2%O3K3V?>76ra+e(*(}5qF11Mr_yA#&VOI<@;kPD z%2nsH#OzAprf^s+*Bz&s&Eu`GmF9kuBD}4isyhwcy9?7#IH$prp1|svI!%Q$)Q_ts zmCmC1yn7{RE`zgKYV6OqW&04sY;woa_%ebrE!b6(ui8CtM(ex)`#*$H!j{Fh}Rd=U<`IJ9VWsDZE+fuomRi(=0>6m|2)i{X|efohaiW{v? zzKSkLgOf6Fyu4kZImA#OPJ+Ohj-=>>0GD>(I&RA=v@|uR(t&M+VbB~Y1pxri-+}V} z&RLM*4g*AZ`Ov^nWcEwoRS+Z^)QO0=e-gwJ1fhm>CQj+8`@|G1AyRhSjIYVc*$cI zB4*um2v?m|UFr|{2)C%fM`Feu%a?4nLK>(uDX`&_F%8apU3ztOsyx&8B;U}q&=BAX{Qog#v?&wZi1D=Yz z2OkmJ_QO*pK8f;sZrblH%&g%gn+x8v5IEtw#NNx zIoba)R5&hwOWKV2BjVy_;F%Uar?1>G^gXYizCxCN~pM*od&QiVwI7wYGt^pXz zPeO}u!n5rRQqnc>zW0qmamrJdwAcYZ;3oU^mJ?Y!rOWcSWE0NYB7M;sma%sy82G-V znHAV)YuW?&CIU#&;&{~0cj0l(!SQ)T84>#Y&+SPSFgK*G{5Z@P_*2zjb32b9Z+u2* zMB!pVIW7n~Y*3SME za7~tx>_KaC;V}PQJ$_Y!;$Zv2Nq;&r!1{&h@Bgqgjwn96I+DPe6OU8*0JI&+%w=%i zluC$K#;&?v!GlfrbDCbU3#~`A=Vu~7mS|;VFKDh0XK*-PI7Yhc!4J^SowCdqv{>@C zauhkLm_=kA$n|qd5sF|qmPK@#l{X#Fnv`)f_^YNK_NJu3l7}fkQj~X-X-_q2c^Cql z&kDy+Mo7-IZS|`TcO^pM;w=ij78Ct{_?|plBRv~XB+1(LhseYO%ammB<~6kY09T% z6S|P^=BBX(>8Sls?h?EwMxdPYWo?vW?KJahljKpYSTd;RAzy+OS!SAdWw#3oe>F-} zn(*vVN=PmZKJ4w!oHysA6N?IGU2#qu|G8WOzuvF^p3r{|&9a@QyqZ5Dm!c1|_nySb z&3@}Iev|8`?gW@Q`=DC##(Ry1|2eI!m78X(t6Kav;IcWCMpVe?#6?J+YH#|Xv>WuOFI#rh($ zpx@+CTAzTea7{bPB1PNvtS6{Q)3C8OSVGcq&c++BFD33_6^!Knwg^qxFAPV_4F}m=BtyMs+OPyt6SE%-{u$N*b)RzUGUhH!SmDc zI=*|!QEdvi3C>}RZ`D>jD@%v0caXW4nGeryk#g1+Fd&>3Umy-8y#sP)G{h4+RcYlw(fD8L^M~Xnv(Q6u#v>_vT zeLL1!4;k5K`n&XI_BuvX3yY!8w01wOHdN9$HJ(^h{kqQgI=C-uIIp`i>|2!-^`djx zSp;sNmeP6;{CpwG-0b{XbKIt+pUKC(Hr?w!vtQhb>^tG^1X)0DIG8=T4AAeHzmJLO zS`}>4kz2M!kkg8wps-9mhfGvT;p|ylM6;1%#DY<}0a#?vaVSPp^V7RABLLSdg`GT* zTfeiPW=y7Vhj9>ag9Mfd%_5cMWabgiXG1$C(5Qc(HX$9$LGiQu4gaeo*x#;XOVX*b zhGyWJjVcQp+YMZ!&k_Q@qt@MZaw7{z5N$Z@DLNEm6>0A+eKu9*!1pk?i%h7NizqtR5 zWPxAV?n;6~_jZVbms~A`XSwDN(%^V7@lK|l6@sWuwk)vgt(}%a(?ANFRGPp^@nf%E z@d%Mwz&*8qGWSgewT*w(y_eaL<;fN650?{?Hl@`-MJgiboVT47`w%OieTCjTuROw!(ByZ`_L%EA?;TW6fr zPeqqK2?#gpkRa3&DEI!Q5646ofBNtW&;O^dV>1LAKLxaE0Na>7!|G4?OqS#=_f}Tr zx(CSaJ*7PMn!^A2`}0@^LWHmKA&mN0K}LB<()nn|HAv`_>G_CGMywd_jA9`|%AeQ^ zyEV?Kg{%2;xAMFEzy#kk$u(SQY*?$sIUtSL&s1iY^+;56gNkcLGW;MN=uyTZ%U*^E z4OkyJ>d^@TCKn5bfza@6MrkK!XK;PE++K&Qh3g-X_LkDmt z+2q)TG%^!apMIc5GTo|-b)Czqchm(F=sd~kB>s7sVgaKa+Cz3PU%YTJJk-jSo?}^> z%L*4vg8K}YzI=`qvFsW+g+I0dO8kd!Sk|8Ot0pPC+~Wzhehm+Bnh|`hy5Ai3bJg@Z ztEsSztuix?@c&83k5+nDS|jbp+rxO_T0~@npRgvfSj7$ceE-c#&%Y~c0;BngBN;cj8q2`1nt;N5f8eO57Onz9>sqG%^I|L?K5RI~Gnd z>rvLm*f{cZe2;Y)h93vji>d+z$)BHGzqEYXo&d~|IcTX8iHDb4uPDd6^!U3xm}Cya z-%2~W^OXPiYn3FBLth-)q~i_JmK3P{OJ!o zTnyvt3a>`u=LajD&I=M11+UKLchh)xtV2|Uo{z38;+z`jcD8q)ZOqlI`tSNmq>Yrm_y@;4^;Bl%7qy zfVWlH^YHaWaR$FVCCg^n%5pd_*%}HPXc4=_xhp&-3_bn7Ee}|rNttZVu*n#bPxzqC z?y^Pyy{af^u4!d%!XGg-!QqecULIO|Jt^skVE1z>-UR^FcIcYmn%Y>KIf%t!s`Ryz zjf^_{yORu@lu-ttMkz%DF9XZ^92;SDj%{gVF{0)lq4@_&e^YdGqkK^i)?jF0T4nFC zIhlYY8MHkvQ75eyQ}djYvzZ@f>r720T?)`jNLqh7nI4(w3j*Ub>CAPNMnU4o@tXoM zIcW=`a9()&4S=LtiuKD&Cn?&j64=1Q8t%$s2o0@rFVLUsPGtF-Du`%ug5uW{?$EX` z%h(A(cY~cn;zD7-3TLct&t3(WArbns0*D{EE#8+?-^?hC!$Y4AkBUq^>-Ft7IhdSE zE%8;YM4_MBNC~F}?Dyo01`NucNoJYVw3i-z&8@Ap5Q$v;iVYj}ZE06v;22Ak<$b0) z{}_h-XsOp>Wm}xQty?c_S~!Pym&^`JsMVoTzrMPmxNd+@zKWTd=3(bycWhDLEkT>R z*Vd`b<&PrAJBMGbFsILtsvBgmw~-||Y9Fm!`4_i%!^Vz$2ZaRtTU2nSw?HwG63IT@mwb1uw}iveC|AF;1qGwGWJ~H*|$U;T2nF zAN;_1$)d36S(Y;#(n)qj5EpSxSnBD>%oFV25Os+O$T%zmX^-!zu&zN55{5%pSL5!S zR(+vyNLOzu9kBu-UC|P<%-+699pAvHQ}c^4BBFK~8IQqv_f-4hRjDq`b`@okp!Vu- z4Eb=2Ulg9ahIvdHA7X}8eWYmdMhR@Eo`lFAfscpi?%V2KT1R}j#@Q}_Y%|q|x=`5m zC8_OhH4Sn5@YKBwf@sZQ4uKAIo^`nk>d%i5PJE98jDzp50FTx zVPrKav_be!^yAs#8r?C8WH1sr>SdGOlDa<#4;mH8bRUUT^7jR zfs^%woLF18N6E}>wjNvI9TyRx=^&+1;-Q!ED+&H@iOe?Ooa9xC==1aE>E0n3c~j;` zz>^ReWwsCI@h4MuGH^7X(%Y&HbCyJyZLmkyqEBi3X#Fn2dlW@>R=ND0%fa8u7S+I+ ziPH;L-Hr~ROBqLuYO}lSIID@^7zteOSTwn>XP-u%HQ)o&s(gmP`5-!0>x4qK!hK3R zw}NsUv0JUs&fi)uroAnu%%Ruz(wWpEytI!N((TJ_oZ4-rQJ zUcBE+Td^XzVgf(ty4q$}SEmNGpem<>Cps`%EZd)CdSDxSa z5rJFU{Le87Z2v*bdDyyQ+$C|dB4jqptb70WJgBt1+yHga$rx|wws4bX!Fp1i4u&lO#Wa^aU~1=&s8oN%u|blxAFMT^Ano^+9qFCOHBgoI!*l zj$C=u`0_V90>3z73XDu20P-c4I>{S4fuN1n?EwxMLb*ZTw{zm*teN3BAj=j6u^xdzhqoqvrU6 z*JI0dsVKLvItWL-PZtiDF)Dfzal_KEvFRdvn4VsSj@lQagzX!jlSbJ+uH1+>VCsJP*zuGmSkdNGvIFyz zjlESFVr!$dEF6J6FIE3mGeUrwNZhy;ruWh0Xx!-zkA3?J&I_TX=1Zw)ZfigM2hlYo zI~>}>V#;1ZZq3mMH2g=rZ5kPAL0Q949d2>FPi%T;nKk*lZ5(Ab3?FH#`O~j-r6+ad zZ)f~4y0T?|Qvyf5MgVbiu_hVry#OrrePvA0-MRy>aB3`tVziHXCF*!t09C*bzAaNop6|5kj_d!HD;44lv zy#u~i92yrg*h$-)=>_{n(!FJ$24vnZ8adjtTha0e^H| znw7JT?A_sf7pza3>l|yvJuJ+dIgVgrrB6%`wyjtW=Tb7r|1H^!ZWHxa3FG*(J(4rL z)WZ|#Ld|QvZclM$w*MTCrLS5Wey^#*b|4k)RI6e?DkbshLS9C$*c;! zuVW(w5|~miNPoK;?5~DdJ;w-p=*c89N9J-5Bex-%Z0eajWmiIUw5dv}>|7^w9k)bt z>xpmyC9?3H? zbuRp@<4MWIAaPb9(CEiE=kTSYx06JN$7{UP0A$K|RI`!e){XD`Lio-2+C=oqHI|o~ z@{*v=wkR%CF^DD^bjdn0xLVF-=Sr}U(%TQ#7RQ8HNt_UNoXT3Q^RuJ>sqQD5iWn8z zcPK>@4T;FIMPDBVmjUNQU!SIt;!*j?6D%qDHxqY0V~o$yru!5C-RqsX&F+iJKSVR8 z;>+xC=&<<~g&QOHjdFJA8uEi3vvpUp(DIcrq2#~RpI+=|sU6)?pf-s}`kMLn%XYG8 zKs08+*>idvrN+C!`dlE1Ftt+S?#SV|;V-U?8Kh{#_w`)Mo7fDT9EELboI3`*xaKD| zII_;4xx=cMUctQ0l54&Dywn0(9N@B!42;1Ubt0@{#0XR463wM?)!haOm#BN~VH^tq z&<6sE*oKhq_4|f(kqmwEB7H`fKVvNCQTbkMH3{Ru3*2tDr^Kp($DV+} zsO})3=W#x!r1QBl<|96X>K)|q#U@n zf#$H|#M35`MRT#Lv);x^Q`Q~<5_wvRKB(LCIE^8hOEU{{)lQaTxn)ppKJlyLlwX>2 z3zWa>RxZ$skYj8J3Nb0n-NazVtA5I5Od8N4?pF28?QqMc4@`*q>ZqrE0d%9uAkcj^ z{hP*Wjqp!wcBAa&1E3oQ6)T2a=kmt@1*#k>C3k|%a46?Liaxq_oKuuA+L)2r5 zr81=#*A$dVf`QMh;}SiJwj+?zq|?UX(ydMzI{dDnV#{go6`uSmhtCPdLT_k50?$95 zhW@##M?*U_<|TK)bk|-!Nm*e#C}v&7r21DT$j;JsG`M!+GjVpKmWWRVhU=Os?Nf2^ zFD4yr_tAX$-D@eX?D{3QMc;qkCFqgIk)j_k2(7RYI%UGz4Alte@h7yvk4ekttv707 z`299~@rCweS<*K@ga?SvEx<4%_AhUS<5(5=-%MQ!{j=e)e3v3zN_Y&qEsIISTI5HmmoJr(`Utb@eNiRD5e$GK8GN zJ%&!4{7H(ysuXFhK;9+$goR|LRmfDXD9w~VJ;%m`p@mDb4V5s=(k!8Ro}}hz*QIY; z!BS6wgwnlTZoz*pSLSp0oH00WgT6ugmzjH$laROR#GzSPw);+PdBN66-y+msJ)ES> z$U#}_*M7)evvaHG16e}b*Hn)dcq4SoEqFwRqWRFzBvE{oC^=YuE`m?U%M?z*gkf&f zQmRJe>(`|EjQTrd8R+AqX2u7X-%-mF35B)NDMlBJev)Mt4ZFcrNU&3CW7o{&C0!Fr ztpX~UNE_60qHP?{Xw#AyS$Arez8$t0dMZH56ZZ777>qV_gK!DUFs^@6W^C;yJeX6C zq8jrVDimf@ihkb-Le^=?7|TY?$A|xiXI_#_kaO8S*;CZR32`KWpAwY-o5Gc+9EZgN zwLXViyut-T(sOkpMu>&^cNRmyY?u&x7ZLH1!`9(z#0LkUb=(0b;Xr4u>Y4H-%2`nF zt{UmW6rfqII|-63rWl8$tm8x)cK#>Qh~jHdp^X_P$(t)eo?QHS9-%-Vzm7WvWw_J& zWc|~R|~Y$L{)2G?YOPZ}J6~wkB?XZKE~v zkz^i*O}I5c39Q-`Fuv52;GvV0wWi8t#cmgNwtn@y4vIQJp#}GA#!^#IeW-bR?BcBT z-QSC^z2EDqL(Cix!iD5va$ zT1Yi$(TN$minBI+L=niQhw%BrbVcT&8V?XJ3{$JoABDDW>Nv*#_YDu9-;F(fh0pNv zK4Hpc+6Wo{<%wG%_CQBHL(oCd77hnH~~|*l1@A zp*XlDqdhzKCe_zzly3>0CE|cH{5S>bAo%)GgyM3~C8J5SZV^r(i z#rZ|h1Q6D_knftgr-+!ypzFWIEc)I87HY@9TNWK)R2r>J*uEYspc(AVr7SF+Fk9Tm zlyL0F6bB<<9^`K~;a*CK3<#ND-qFK~#dFrl)jF74>FbEyw_>Y4ip|P-GUQC{=K*5; z8!V4~jACa^5=W0b=$93_sx|&JOr0bUb@oR}?JsHrXNt&G8=QTVIwc@p@$1f)qO(TA zySr6{TAO;66Gz51@_fPH0j$S3uB1Xk*NVA>*#H!8-coR?R=cCJU!7OrtDmkf~v zA^3RxdGtKc|H7t*QnC;>TQ_(Yrb%L-Ql~3f9N9BY8hhF$k7%N|@+zmT_vjRfzRXw@ zY>nndBmxP~VU{r?h&C2;ZjoW6c znm!7nX?V|R#L^04bXC`AOY??Guu|tOMI2zP)V3OL8F(qELfw3%+kFw9-2HIj$fdYv z$fCHHI}kp+v*VLIL6*YH?5wUTgB>Qdu0cA99b}hnQ3w#)A7kvWF7o=4`Yv4Qud6c! zoiU2C8fYO*=YG}k_*q{+uXHXjEfIEI-HlOvHzBh;)i~Cdr8c9p&-~!pGj_+-4~myK z)QIkzkeUiOo7<&r0Ls7KEWYSFd*zPyQzP^UibZec6A9Z0fvvwuKVV!@Oe0#25q08> zh!oR&Q5s9XePP4~fz1S;>DOOlKk$6ByvDW( zeYuDiSTkPbZ`gLYo znpR(Rr}F(?^)?`R=8`=q{Bu66jeR5ybk38_Q0-brAZzvvSHsC z*A4PE=T@XNxqIhjnaGc>!Y20UD*?}lavF!bl&YQyO9dpu;2q;M}=Px47(vjZl_ZCI==3M1RNveFIzPFQlw~2n`7U zK9c!FUGxT>fwF}YnR!TBCScuyANWV6Md<3#!{;5yDi^#yV|jckx?v2^Q~fl*jH6k@ zcxXo2>c@hSNl3w zywpUZpTh*)aQu8~3e7b`G4~mm@tW0@-F?A;YiY1TwIG)6T+fB;MUepVL=BHW;=*Rc zs-Z|VMQ3o5Jjz-eS8Ov9GRRf_nz-H5SiUe`;}Q4Xe!2uiHL+0Y4Ob!emknj~Dy&WP z_=O+WfbL&*LF>B>W)KMKwpg2rV<&@Jr(DIqmUcHTA>}U2Y?@(A0Pdk$t<4xeS*F$g zJYZb4qYE5O#r}g4%l2qKGQBL_OdHO zp~2QJoQ(08x>TNs5{lX!&P(QfsMLoEJZsc&_G!cZ(3gm8lP zTH3n@Yc2lKClw0D5_-DN0P69rBWq~b8OL6wHUGe46Lflgf#L-z-hOY!TanO=!3uB{ zJ{jliX>0>aNzj@rxCS6hvgJE66k;aH0B`J~&4sb-?I z(`4FyvN5a-PeakC+(}Glk82P@N1jm01_v(dFbwqc?b+_t+H3mUleSP#e{)Qkrv7at z4Yd~{sj|g+b*3ly^KWN5NHnqZA#!ZDd75s?1w^QE0+*2M?%y=wI1`V~x%Gcf3PMGS)tI;SjHZ7c>x469wV> zJ0cds6$u%zYr5?%3u)=Jhs-m&N+CFSiP!|je8HKiK2$yaP}}#nA9WE+lMz-pHGpISN_4UVT<1SoRqu#YBg4&r!5 zTEe>R49&sr!R#xw$8Nnb9et0l(SwP_G0`Kk1HjbdZcGT z-Rv8@HwtjX?~L4E4BOLz_(T<@i!+Hsci&RQgV8ONMA06cXhGs1=^`G*mI^3W@kR z|6Qv&8*5`6l|r}%>eGpYTnTFNkX4V55}KApUS?95^IMI@NoLbVbgOtEs`Wi5)Q6gr!wxnIEHR;(66eWxRar70fd~@Sh3} zby(znpN_Qe_PatAMN|NZmHyD{hIkk$(6+2dIH{rBzLx!m%v$_WF7oBL)sr=TvJZZluFlE~%p1_Bb4!p)0>7_*?t+#9Zmzp#ibnfktX zLJvLXBm+I>0Nn@zCWIf+2Khn)x|IKT6~5k9tMU{W)v`c2E+r9p_o=#VR)N|R(iqQ} zPh$qm7*St8t76BwtwQX0QH+pqv8wcA~v~~#h$5|J3c1x4Q7yJtHxX? z{w32R5;Z+--2|OwKfk8In9Hr!%_JRhOkqd0R3H0tF8YEzFgz|s%+F0B=IKp1)fAX5 zCA_9233++9X7Mx3J5C7SwD6N|43FdRigGl=BeL6pLRkr@HOk#A2Q;LF^mU4rMfy&r zGQC{U+`q4WEPyrGB(X2Fki5m;Q%JP=lkihQjO)&rR&MK)!7V%2Kc~{``p}+1e`&ED zC!b=dver;E{GdnQbT&J%!#ZC|i3S@DKLsIpwjnvGrBsqdMw7!UV-*jcGn+sDd~(Hg zegK+YJ(JiCTI$6j{3W0%U?d&+g!^e!9K7)U=Hou6&^7Jo`(|U14j;JNucAq9mwJhD zmu&IKb`j*dG6t&XQjS(c?Jacl81jUrqAz0YB-O;mxX4L^-GSK(1h700$ui@8r#B%8W2^HvFw7-4?CK zQ&3V`YMxBgf_Dv|aWAV572_Fz-E3ytg}p2A6sqmd$bq>>z;+=GGRSdgH+5P$8}1^b z1_c|B>86g!DE?jjWE)77H!kdyRc0A>Kfrobj^+(`%P1f1Hw2s*XcB=7B>_O~u{$C3 z6QM3AMpT?-yAl;%U{ul~*Qi+}3S~Rkxnl@)$xkHEPA#*N%Q6&0Ji^R%y0G76sLmGK z19|!uu`R?&LrTEl-jLN*kYwyk#i4NkWcJgVT|-f>k51*v&HE;>@x7RoiZuS$)ekD?NCzztX^h=NQ5`Q68u~ykz6%L;I{#I1Bxhl z^DH}>;Y|VY7Try4Y}ecLb;9_jgrnk)diH@H76Y)-+`(-!fG0OYAwz*E#RxzmNJUC? z<9x+!J78oVK0J|?DQLZ_#nXj2~c{wy*zXXP58C{(MF2ARI$#}TAJ2#^_L z(ox$-1wAjA)Dl4F9ABcm^-LW&%ldgH3dys@W>h80=BiM*t4l9x4;xB5_JNf5ZJ3i( z4mSe~a9s?RF&ZfpGEm<6P==(m-~>CD5uYFEp3s|lrwJefTS}b9(D?@0vUh*y(KHWe z+!@pCc~mob_g8p5!A3lqsAI(wRK~tSF8@(#L&k9@30MRvfkNHbx7lm+GIwkc%C}p_EAJM6;-fA*>xHQQtWb zg18Bgl7bJSCqZqHA(Y<^`SB%)L7gTjE+9a`IFODJ){x-MUo+W&cpb<-*XHQ zM1_6N-DvTMI8RrCEUgXTDN8yI>=!zM#C)c{b4ikhxA^RB)#{X5#sIp&F}9htlR6*X zm@gRGY1S;V_E0-UVgKgyZCf8#rtVQAyjI>Lh5G%>)($A2mQcT^z~X9o0nDS^bHbE= z$d?g}!)TejD?!4oxZZe@^>#@;SK99Xd|vpId9l->URR+VL}}{wfk>3sD-KbYf!~!5 ze36RlMucdJx3M*a$aQdt9pR{CCUpUDnfF{aJKoKLkx~AncV0b3M8nq)*yLDNKM5_R z2G9Jleb28Xj6iWGmiYrZiy7egfwIn`J)NbcSyS76S#X}}hN|4t?cP&5dp{5AViKluZdUUS3|BJPO0$S2No=#3&I4NP}-~S1O z^Y7obl3W5CfhbzDq(10flIx3LqEQMxTjZ`UCZMoOxuflkabY#MZKzLM zy&X%M`fL+m%NP-$az&6F2*?K06@R6!=m${8y&Vaw=4pHBmY&v*j6^nWxIKwhx8lm>v$xKIWB` zK7$9Hd{XO00*)D(NsVromB>dwVo{1Nywzv_P}zaRUD?S7{F8Y<9qb);2}xb1Kpy0w zwk51|^!8+lDWV(s60X_HF&%jmMD6b{4ZdtzF57Eu=@Ovr1d%&E3_AO}sk;-M`k`@N zDcWQ>!1wEG2DIfpRr7$qR*2~8=DaMcuZm%pM@zQT7#C?bBZ$|WrckOM#!-OWD@z&y zCv2xQ(CjmNr~l#E3y&+!-z8*X=mBGT%oI*8*W)HaXVWPb3Y-OeI%R(}6+~A8@5}zC z`-&RJYsBC2FBHB@!lETQ1DgTXw)dq^eI0sXTb*rr1vv=)f9CKXPd_JEx;{U`YaNUO zu64FyiDG5_jnM=3mTh&q?qm ze$p1TomTT#X9qtU)i7a}BCj^pF3cKNk<94IEDB`qB4nMS47=D(pH5PxK+v9}%g}`a z-q;D&uW4w@Efoz8Yt8c=$&I_TZf(V!uCowpwf_6({@~x7+7LL~eoA%6SrWLVxt}?3 zn)?V)&Z*ZQel5d{{QuRZ{XZS+j08-~tepR8%>L)nW@qI1FM;-dcWEmz|baVppTA&8TCKCe_1VGJ901eFa^!0s83XyC7 zkQg)qEkhxgRxD#4Y6ysASYG}Xw zYDj?)#T>*u00(#GKLBNdEda!M?fK1=0LD-X!2eX*s>7-(fC*F)7g7~0QFg(q*;iKr zFTaJNDyp(N33vo#Re5A^0G4yWB+IJHKEEo!8$IXjh2Rur-TxMdF3=yjRCIM@wM8W) zGlP$AFaYR2O!Hg)`^=Z#CdD2~;IH_X6iZ9^hWAha0h^mM^U=wnql1G1^OK{!F^C3s zqo&3m@!X8|VBkJbV>?hc&t~vHz`qtziPhs;+_=;Jzf^ag(f|%z7EGpNk;%|A!scz0$RdahphtQ!s6uM ze*OF{t#)~k%e^2#fCX|dFTI>czpEy{b{W6PbU+?j=`qcrsri$>w`uv+iGiPb-S4v| z*n!r#G&nf8KdOij49#_5KWevkp0hO8zxCtL)DqHRl@*L*-LN^dfjno`+$~WtH8H;; zzcu7EWTVh~rY9hEcXhyNyCEuDD=m)&hcgHqC&b#Cn2GTcd z0_E%}2Dr+?=>{kv@O<0F9v=tPNBXh(KtBLs2>%k)wwv-rVE2=}f->lX9P|o z`T1wiPCUXNg3?d;2wVe@S@501?!3Z(8ligy@2;Qx6|}p8`W4%<^OE{kko)kjp!5N} z`@Z}~@Ctyeng1+W=NsO`X!R{{XN~^{`@V~CkpE0JavuL_1@h3B_=O$pu$6jGo(1|b zQzNl=FZ0iy-kogCIsVv(+Vs)^{cEOAWAC1p`y;=x9W;Y$BQyBM59+{59^a z74rSBYU7~Z9qr}qUMnfbH|TE4-WR%;R=}j*x2IOXrQV$lr0s9tFV7#T-dxIG?er<% znT=l_>)-y0m=^--J#py7R(34k;4y5j4Xrjy=H6{r8oQU)+TU;4SAGbJAMIsdlgKbX zZC;GE-IFZ;UKZR2GXD z@zCVDEnX+rRiI0faFNtfb0pA-^3Tv~u%=bmPM5dmyhktN?8n`+V0>qX9A zAD@@f+%{Kpel+-6w}tRUDLo_QUx}1M?LjXa(z^=^toVs4QQ8bEHi6I3dLN84qQ{t| ze!Q9RxPCDDUVK5EjivJCH-H=+3~Oo6{Z(25+^n9h*=A>p@Nr6i)Jwsz5q2eeh%i6 z5g{6EE|Fhn64VEdpOK~f?ihnS;`2g?bAoeYxQFeENxZNHU(6n7+9hKChAO@fJ?@(inc%ysPZ53= zIqU&93vpSx{S8zE9qiw&c=uUN&f5em2X*{2}%L^kgdr{v>SpPqX0_6D?A1m1G4$BZ=rBTsjA>0H6t(3W6?o`C1g(fw#*7LOWOf zGvFrq!GA2D3-3Ty@)zlw9p)w|r3lu5ygiHKP(Eagp?`62)c7W2?}oyw{Wy8$3usn$ zp~bLibqE~>nS>HX^5UR5;6Ebdys}giI7eemSZi!Ll z?6av3=U(DEZ7URQFIXtgbt1k2l2@L(Vnq?D931XX8t?+bf(;5&C_aP|SpJLiWuJ1W zac?uNaMW^>TL=3a0WEL+Do^L5%{ks)AcK9SY%v9nnliZm27f7m;wMBO=f=hWD@4*l4oK@rZw z1>E_^AT=d=sp6Ry*zs+@CP#{s-7VUFrbd2Fk1-4BN9$5l2TvUauY@v(myda=QDtiY z;lQlRQMTi~ZeORyL?jxVGLB9hi?;#=+O)vLxr-XA_dpoz9UkR(rqMX%6$YO{@Qzc4bl zb$)3B^pcMpZvnqH}NWp_mE zydfC=#YUdgMBV{jRFnD^g%64@gN!cRKLnylPTS;3@Hhc2jCWaMej92y24>&SGRzV< z{gF~u2{rZyew9})?A@u^sg}QzaG-+!uRCwrr>=64h=xz!Gj3LhS}NwEx4Mr+kc3Hv z`J?$K780PHEdgBCg@eaWhu}*er_I8Wq`hxwh8-8+GWY$s=sx)~`a>+5Ecui?JcaSy zZs_a&+I_9-6AOKL;ZS^BeViT6#AJ`@rVIT`285RK-Z8XawL>oGN9zUI1k? zrrbHeto?~eH z2`z1qEUqWiX{8Y=BcAex-20M!f`Gt@2ys}IIbv76v+jIzD+rd=HFZGmkXHLhVJzm9 zC0poKc~tW7wK+uqMS{(WaqPUEGo5!aFM!pcN9~s4kl9cI*=C1rT6UFrmhX=#TSnUn z#LeK4dPgxoOIp3#{mN0}y#PzTbGGn?0R^vlO#G}r`mJ(W2qFP|0(U4&jnUz(gfe=u z3xEbg4WgNR&PES|AzG#2*c0s|Mm&0>N8Kd^D@c7YAKfS7Zz6|&3pBIY9OS5~k%-dI zP0S=`dLX2MO)C2e8$l>9qjgpoN6bWF7khNJ5zbcG+35pw#f=fYe<#O)M;g|p`Lb6h zJ`4+&Na^vCC>+6yKST>7?%0NqZt&qW=&bk-o;slADel!7Jj<2{gu&B*2wgPn^MG2IJrcT7 zslRKUIi+3eF)wX>>glDHq`nnsHu^wnes?y0!gz7`UrPAP5`h#H;-AS~i zt=$|G?@dKo+bAydrC+lgWlGou$LRYf#r3wnR2<8z7h&31;g5x8Qwx|(VfxO<-5M-d z*1Guz2Kb%KZK+gyj>5HmkIb~6oGUV+B{xecX@?FTSokS_xhVW_m^R1+>$St2|Ect} zbK4*d4sqah2?$=L-BKtI{_-{-B-C*9dI9?;s4ceb{X`~qcJ2un9WGc{&k#fsK~asH z;Mtt$NxB(FvIBcuLfVTo8#7mVm7wp)DcoK>?K|F_QQB=~GHrjTcYBZC5V^hIPJ-ZM z_v_PiRUT$BVp-OCbbgcn9Fu5r-|-Z8LF_*=+$AxRuJR4Ld%zKd66V z<`Uax`ec~PTA{HM-@I5{kXb{q;YvdTg`?;mTvN~L+=TEg?*8TQ+P_o)2S ztbBwGQZC%;rlR)`o&0js72;6oV*52GhS1STbZD|5?P0@<0q!ESQL>mh35>ZYtCq)N z>DSKVSrI}d%;VbB8rZ|6nIMrv#ymR|Qi)GdZ2g|8vAa2Dfo7CncV!rU&?NB)(q-4% z^Nk?|bl3uvSu1w_&tBKKY=0R=rJyA5c;Kt!eC3V?3pp{0lh8v}BWV7$X5#)ym%K4| z-;0%XId+$gSs|*zMRThd#LPR-+YSbx`TcfP_e@Du7BO=Z-8l!?WlrkjU_rivT^`f8 zXbZ5!tMlNsfS=0inljSWbMj6XYRKJCYTyAE(?v{sY4CqU4hr z$CXKc15{8|R(7(}Ej9Y3J(AvuL%k)_thSGr|w2^qJ}kCZ(`t@O4S{FlNnrJq+nPa0Py_-Nr83cY|;6P$y1rD zZ@rm0m-+UmGrQPx!6299FH( zQa)@ec>?jYuj`?rZVfW`OiD)bEDNu8X5b<)Cq=wMt5XdX^CVL*p=x^6U?NLMjn41R z04y*vK+PwaW)v28pNh{Z!;jjMHzVSioPLG#u*yim+_xK8r-Y=F*HXFkE1W(b&hFXo z7IzpR>}EDg)%2HlQ)p$r{Q#i)Ek;zR%VDyx@V|@_QEw*9(+;P~xhZDB8$js`+h38X zGKVUt1ARlnu6i7Ap{XDeZb2d|P`etl$xe~@ioJM(21N=2m?mYUam6d9cwrQbF0;k2QC%j_%KuK@Vc~nJue`nrf;|4&+8kt{MiTVQAlZlCyN7`9loLXf z5A4jj7f5-|6p0+?e6C*zn0I`EM=jrs14Em>j>sKgkLqVUA@3yQsS1W>!F6G)j8Ern#Xsgr2Dq?3 zPf>_B$=R8sias8@2{~Mm+sF!W2~g9i+T1Y!Q5IzW+mFXLX&T?8{-8>nX2H@CD_oCq!$Y(IJ)S)=xr!{y+8l=e#J#LT(acD`eSRh&BXePHY7@qukazEnbcaMQeYH+p(P8PfrQ) zZQhwR;g^=gNwN5%$mB$UOg^{E@gvlsO!fqkcr-L0KbUi8Xn>7o0Xt)E$uz_JFxd!N?kMJ^ZzAIFT5X1u24Xc%$xu#|Qw?hyU>R^q7iO)^c=MPCB+f@Yhe*|7Mh9$oA|4ICUC^48V=Zjypv<|hvzF9^t{+|D z1|%AXFxWGJxb*=$Y#-@W1T{j(d9s2FPokAlYZJ@=eK}LF*9SB6xy8+SDO=-5RfOT$ zh#{)STx&0D&81KxadpK}2_8#d1NS_lWCYz-@k2w{1sRR`P#Y}z8~!r#zL7C~Q!!D}~2|*rgCtIe(!ij1})mn~%m9nldq*8tt68#}-}M zM=S2uE2sp>1o$Iu@97ZN^1Qtxu*3LPgNyYpMp2!WKOqr_BbyxrTHuRjx0BR8-#3bL zSS(3ct?$5d09@r<{Ru2JX0CE_jUR;+bz^|4A3v5oyzt054{WpmV(J68crWN7v%cKd z@_d!4fda3T?}}I~ixQ!U%>&cLkL{7&NWC^A$pu&BCJHk}<5;KFF%2DXdpor3x{V7= zf7fc7{dV@i(A-(h_1M(=GP5w1j#ozfI_>aPQZwk}sa$X%t9#P(nSTqPbe8UuC6RZ} z4|C%q2Wl=Rij3anY~ifPY-Hf9U7dR4qu`EWa$L^3w*0g*`5Q(uOL9-y5CCiOBluCP zDW}la!2D*Y`%>i?hw8VeG4jLwFsCF`LFp{ETyC$zAqh(O(Zs(+pHQVjvtv5D{xQKf+hNp$rNwOYMS^~cyHxS%b&&YfZv$#;W5%A7bJDntWe{EqU4eC zJvrW-y#5MNSKqzPhuE2I*`BomGHpYsdh>AVL-MxHi_C*3udxtC!SCohdo%4g)2!(F zI#0okg1yGzDwWzb&kXHrzkJ5jt%mkD*=L4zHI--fh585jv>m<>lK4!;m(A$uCYk$U z-?Rs*%57(_E1d7teCA%T-H#UcxWhuD7J!=5+7i`8*PGISGMwX&0*|M3dP z)dI+gb(atYcZ~f(LI?pPictYsV^+g2-+`9ppy@psxAcB4=f${20MLQqPbVNs4S*u5gI%p_Sx72nRMT|TAHbA$pgRuI z2-~AaFZS>nwsRp4Y5{`IvxzBr= zfW+$ZGaKvY|?_H=uiuIA@Fveqg7xHM6X^$5%bn9sUDu%BvQr8;Prj&cN zN_l?9lc>hlu@@ObI^_L*teT6?cS-3Nq1nXp>pOJj zilwVc!PltM^_s^RIY>7{eey_-21^VXiSO8x*j%&2FGZ1tbgmuP0mi?wMT1mnOoJjY zNId`#UPqz;O277U?JWSZ960qGU|i z@v5Rw@lnVXQ{ACgv2maMaNpe*LBHTZnyM zE&tda2sK47kcVZb>vIYhrxJ+zuy7NI24vQFS@zh24rzok3+wrzjC!{-oP#cIP#h@p zmO|Xfh!50He@8&9cQ=kqJ9Px>bdPfkZl$ts>1_QX9C^;Zv#-%t1%M(Z7u@FDgk7`g zhPMmof6`{r9JK6|WVgt5w}FH8d&wOUcS)kG;o`iZc0mmw;X^Y)mxC%BdnKVo{ID0` zIKDewG@y%A?8?wKRJ-C#Ws67h+0cXkA*bHfh$Jhr^fYOPxJ6TI&>(<|n;}r1jAm|> z%Sk^qt46z@ueOc%**EY_tD`Ezc`V^57scM*z(0U?`ASL;mP|D89|P{`Jh)3$=;avq zsr7ucT20aM>%!y5XcoFaA$WaUBum#Cq}6guMy_Kg4H1qc0Aon z2$7K23xpkx^swgB1ejt8=y!$XH?969bsT}ZgbT+Q#f}~J`-qAT$Y|c+fH)oP3By^= z;)oq*5B2j;C%8~BcTCh*>5zso^xLv$f|U0)YBN{9;B~>9tD^X1)9M191$OpxL&`0<@>zFCPMdCi==& z$5#`^hv2z>?PX(sra5`OmAj8)I~-l+tBVY(Gwj?Fj>jZ#nGYpC59#Ktzcey!a= zM<;BtM4I@Iog?F%Lh?yID}Iq+?9`D-J)$3NXlH#Swur&K7`LDLsl=jHkdHQ;n;*)r zp_WjX8%i7mZRHd(rmi7j6bW1h3)fiuig{F`+@HWp59;LobH18Fkm*BgCRUCMM?_1G zJa3S2#Ls5n9uieSQ8YJ?aCk&}=L0?=^ogn`CPxsM30cl^iO$5$t~_jtuS+fAso=Tj ztJVh{QT~GuE zqv&?dI1WE?sEiMoo4=(1_6tJLge@ChFACX8nhptmfu<5Y6>86MY#58CC9|BakB0jU zHjltfv0$X4Q?Q#ToH3<(hyvdxglABb>y_Qn`KfbG280mvFxZTT7L^H+DWO2b$wfvi z^A3e!AE_!D-yMp|^Q#@9Ul*M6^S54Zpt~17b-7iX!&9AV!qnP)u#YUJ<_$jP*63fI-ps%a3n~SP7(>-7j(D>K?Z4qh6d6Yb3ju?PEX5 z+1SqM8f}}XCTUBVq!zAfB8Nx+Lhe73%bWHpsYBz>w6G)P*4T4gxM>Wyz zgcji7Zy|h8x(Jo6v#f1FdrLMPCXemY_G`(D43#*2Ln+?|dkLe;nB&J#Af$3%hau-V zCoyFQi};{UfP19zb=C3&+MfeboK5{1|o=94Q zYXU9Wn{2d@^2A-}6-_e;;8Rrj$5vISGsc+8$Zf%hcbCxi;Cxm6kK^P~22kF= zC|*725|fBH*6{Xc;v0G|;{#wLVUo7R5OQg1ifA&OTwHqiX4vV`3;j?!K1b?JfJ6mJ zU2uF0UZ&}S^ih}n@C3qB_{m&qx#nF_>{^>!(O!Jn0*UyY_-*4xed&4t1@F}(Gj}Bg z-vfP|)e9Z#Sya}BY7aPnp4C5ZivCQE)GCoi%%i}tVXb%jqOIr-lT=pDeCh~;u-QJ3 zuY?ePpy}{q4BbvRMf`h|UWLB0TKcU?O|$^FL`>#;ap6cquahSvAr}oB1>vFd11c13 zp_8@@HrARC3@O+m548-Y*O-38p*m%Q-Ytl7T>ZPlcDHkf@&RlEwVstVJ#OCTWW>|H z(ce#<@juNfzpvZ&K2y#6Sqwg0EaQBW>qmRO)>vAUEap}_K3$sqjJrnKdMVV%@TvDe z7`9mB?wxX)JYTzH%OvzNcR0LRNYfW-vjSG2OH&Xtp$jW`T5u1cTbkP?Uv+w;3$d}6 zN*a6YyGdVi!W8@tv9+$HDjywrH+|pG+VpS@c{jzSTwYbuu`MeWPQF!37b9E_#j;}b z=Ms*G;O^M1^)!wvi?U(dz#!Bz-UF*vsCs&`8YJeu{34B*9(@rxn*zW3bh}KVWQ?rC zGR*b8d(q}=nkjMA!5Clziz;B{Ly3gr{w`as$dPj@5(>+Xs_N1a#<+xu^bT5OZ-T7K z#~iC7H4=ryl+SGX{F_0&D@VU%fV8v4Df-p>Q63VTo4-mZhw7s>nA6g^;gYdmB^x~* z34xF_i`I}X!OsPPBV#S`5~h*~nCDA}R63LaO1{sugAw~*Z&Z?tkA(d$*XGc$@t|9| zxr3`WCr!O$YE?9!^CZ$P|Bt_3WH5Em5FXkWhWThD_E63C;Gd)2&`{+>zVF)Wzei{L zKF*!u5-gU-iQQxpOQJPt0;{;!uBdIG#?;DJ);i2*Si*fix27iqWg*YAp|3}K%6x01 zoHwaI#avTVQ&$mvY~bmt^;32ckg{3nQd?QVqeR=RshjRVRG-CEP*B*g0r>;!$p-a= zx0jnb9L=LM+kcJiA*A?2Stc2KVSa7jjwNqCMIa90Qf|N8CvW~=&Wr8CX~lVpL@u>Y z;kO?`yzopBl{8-GFzK7afXy?s)iU<-Cscgd275_8ZT7@%XqLNVMEp|ZytxOu=fIS> zbppf<5bprk6`kclo@kcCwi|h_G*0pH#`+V3Lf&#`sk?DCRifcd77i}Kvlfu!9ymx7XeTW+)Lr9KB1VIg2GA}M(m zK6A~2^H9Y{yS_p#sHGh0q*9%r(|lhXcbdDG?PYHI8)z3*SKKL0VGb)Sb4joQGL4tZ zH(kh6*eEiN`$EqDptZcJuqWI{$wxQ*B*$$Ek!XIu*y><~;8buFvL$=Ft@V20Hbj#O z&PC0ombZ}!9ZYr*EDMw-tjQnrg}TeW@!TcjX@qg`-jga}j37(#m-E&$o?!FP18$qS zj#q(>9zqsas}~QM8Qg><=R|kG2Kcf89wC;&$)%4U8G@}GFA>GN=uFJ z3czt&X4m)TSdaLJ3=^L^J6Tzlgldrw-iZx*~RDq$p z@3|d(5GF9u&e0my!(>L@G=BlBp3TLwso-vb_VZ>xUQEMyyR+a!$h)q#yxa)85p0h;F>EDptFTgN

AoSg|fnrzF+*_(Opkh4aMsr>g~JG(6exrA?TjOKwE5pq{jm+uX&VW3S0M zeNXVWl+g2r-p`h9Ze3M=?@uXsGQ}rgd4N@Iy#)%w*WlnkSh*w`0r#hH;p50^$0*+w znTEvHrH;q}<8vO)2|Bg(Q^`ZkAns#YU&;DFHi8pTj+{DL=^{ALw?*V}(xXqG2pyr{ z$Nb>Y_OTGc9Q(j>KC;e4$?g?UTpDPkQWdpe+k>nwezwMjZtHPCLwR3grh~!o= zpB8;wlJRnd0woel!k;C52dqEMU`?JcGa9r&^S%Fk%R_L}D)qk~0zesZ`=58lyz7v- zB)AY8oadGq;@KEXh;lqBDba+qzMw|$w(U+MUByXrd*>zt5I2+k8}i--)n~-@;~P*v zQQK|rc;wW(FvlO_sza5UQ6mKFQB*&07joDRP%`=FKs$+SGczaiD~?YGfKG-WVOdPT z7i#=YsdcSre_W^69tVbO&Q|q(6u9zNa}v0(2U}7X7evyv-WNw8nXUX>+m~LX4{gwI z@Ge16-3eaMQ>!^#iVQC{q>ha+I7Gg3p9UTlRn09m5V>mb1Ct@KbrtCe^+ zP6Ht#g3QY3MCUY z5IPF!p7CT8rQ_i6vp2N+Lo2wQ86?ZMVR(8o^<0@LOCdbqXC$a!0St@7$PpNEt=W~@ zmNLMjj8OKPMIf2Key1zK3-9&c-M#~sN3BO(Tl&7^2-4|a(HoAi)B5~@oT&N@e#&6Td8 z7E}!(6|uWR7tBqeP9rbA?VCnl7nce*n!3=`RBzjh7tIg5SWZdgXqqqvk56P24(JPr zhzF;HcVyv>dnIoj@UIZ`n!BR6U~=SEPAxq>S4%VqJMYsQgq@3;t!w;Yf|C{|NfASC zx#^~iSy|xWs%az?j_Ngr7∋>*g~kqWO`khY4%n?6xs>_-(lIKc%8&>Mpq2zErbg zN0Um^^jgom;O#W8mgJ~TVO(Dy5%ITI!0eC0YiT;UrcK5l=k0$($5%vjqL;JG`u64M zLZ1-g13rDIhMywdn2e9dN3~9$N$|6*`>+W@%qu!cD^@zu@8C;Igu8v(qdVepQ>JZu z%WxpJ+xTi>sz~B?r~b6Pi!!G^Ld~*^#K97r)GRD?19P58ILPE#WIz7JdE?9vJ`~4@>sDWG24SK^9J8XISBJ)C@!i+Fq9x&r zq#tyJIn85-J{I6bXsspPWfO%V7Jvo1u_4A?tFI6BI%u4uU0FaT1}$`^AGXp&)G3ui z{t%RZ`~{}M)xpA=Bp?|0y)Fyl6igRj8alNS%zNUZ;e5(VxVZmaJ6ePE#a?j2M1CV3-a#v=g^)={>Yj8_eL-)i)$xjI*!`M-)jyo2d+Xa-!P}S^w)#1UxL7`? z>7rwH|9Ao;c{8$vuMmxm&staLM+wK%MFpD1^b~GJT*sj_JCXa5lRN|miL)GsW88h4 zU@fw7^`e(TwcSZN71l4Y*k$sH{5~PDtJbl9hCKFgWN7L9tA5J3rwRI}Tb;mTe72z} zjjR0PZCC?Q3l#imo8}^TZ;5*V`0TYSKX8%gD>XOM%4um>-bUfLNGh#zl6-HEK z#j3{WVtq|ds@XVp>~T=eeM%JgVvdb26OFx6*k*fkF%7-~3am4VYY3hi4m2OjIrb*i zn!Z>9$UUSdgyCSJXvYtPMNGOu)z2#oRuJ8ZIxC1V>v&S`ZEagyVYcQ2aubY_1w4{h2fVCGj$~i`bPruXVv4X!25G**c9M zu*Lk9J^)jwPS+Cf4My`j(W0J5{qgNG7t1UYeg=3TBKQqHvsV6VYrfSjZZ(Pa+9Hu(W@*Jw?m?0e}v{O>JM-M-==p%H{4z5k(tj>C33GiO=3zdfz#p zxWjh>n~d@H_Q$J=BO_-Bh(ck}PEmU>6jUaGd1xF~M^3+}i}<2M zKOZ3HGh)&RSpf_X7u;!ng(`T;e%9jHYMJ90O@lgm_aVyF=U0bm0c!Ql=Y=iPt$J*q z+u@V6ij2)gSKwXmVuKE<>;jd?Dr0aYtjzod)Dt^~64OF^il^151_WdSV*a_*6#*aFE0~jyy&NWuQf)2vr-69%USR%E7gA1A_B_DyKt{HN5NQ&<0R4MtH z-$K1KNb)fOR`TP=J4Hr1eMGJIuBHRfj%Jb!cvSRbIj&!muP>n3u;Q==`On&092 zuoe!^T9ty4XAS#IBWw!`BZl@;YlO7CjxCm@DFVjRO<~%v zA$6p*jN^mNPJW}ie)(l&V_@fu>p51$;=@EXMJy?&>l~`kHx#iK4^$rlqjuQR>kAhS z{fBDKk$qaROe-G5zMI|J^M?nt1UAM4X;}y4`Ak$sL{YdT>|FG^23A%(&EMWukoAMi z!bb^VNnM#I(HS+dH%Nw#du#E|aPnSw&w`9XG*qD5MRp*sp=!846uxf8I#6b4fHO>* zQpf)AB>nV-!UId7Mf+JMRzq%m9i~a(B#h;jouV)|$2G;WD#0U>KfxYQlC=D&GU$0! ziuX$ZP$7V(a`vE-7v|p7QyRpMH@3CE31>-xVha5s2g_l?hUh^xW+Xz+vvwHEv?b2w@o~0L;<=;Su`c3FH$3 z8hz{+@{+pS-f6K(;VDi)=T%8iH95WzB$)ZliJe&~kL$7^cZs74o8!V^m+1LDw^xD1 zZbLzhNulfU2#u=8k2BvG#qa!9+3Kny`%z`l9gHEfsZ+?!4%G@S0mLbd*}xuE_~{Jo z7kMwx5_wTFf}z8K-V_RSNeZK?e72~!W#Qr3irDg?NOU^%?wrb-8PTa9i{fEAorw8VUC{r#;SD+W7k0j5thc-E<3p6X{NVbh4oVOK_vm zXwr~j*ic(9%jQ?zjAHAHp~byWX;Zv+*(|?Nyc-ye>%2@;Fo-z^#msr^>5?J~ReTta z*qN`o!D`*49Yi{r#`a0*qL-u1Odq3ZKp&@(#cOoi^`VXPFXUanDZ_n-4!4F-zf$QU zb5e&}3?9fzuQi{<|HSX3;Oj&IE)qnf@!BIIR`L09Z;2v1t~NUg>9#piqT4@-%2|To zEX=3+;x7wOXirk~G6Cm)7O9<` z)eRo@Oj@dz(}l*;hj|i1qTZgvE-J!68+( zhpS4?4F>+XDhIK&`NSh-lVK>5IFZjKfSpj>vk>@nNs+&L=&dm@;)vl|Vo{KiiTtIK z>YD9k&Xl*W@4l!9EIY%^H#5}%UfWN~~)*_BY;n{Lq8A2ZKfC@AbI52rfMWJ*FCc`JT#Bh%>ldd=D)C*&mV}+En z^SwfA`OCbzl;cravq=KT{HUWdET~Ev_*uY_wAZMG5=g^oL<|9`EEsss{S3QuvRugW zbr)?hDa$z#Q;<+NM@w8Cl8Te?FAp}+>tLMOhWRZA=UBgDMD*nwzlyP z?SQ=O=^n@%%d9!js6wLF(p1BA=VK6y1tRRg`d|s%(wQPn#}9d7fFY+LI27HiQ+GtF<`eu)|j^ ze_>8v)C`Sz**P|%lok@s)1<;X&4?)LVfS~r9brZ|pwccPia1C@SW&HN#t9CZ^1ORt zo|9VTILrJG5O+MQSg7V2dQiHdNL(*)psViv~9|qoUn&uF;>FYxmndi*0WYs&Y z>$$%}(!J*{vP&?Guk6goTaj}`R|#S8Sk#){=NoL}TPMo-6b4p(oK6^d=>wnfUT>sF z)|9=fL3RglWTx*nFaD$50FIE(n_uiu3uRxyn!_{|yG_{>c>CU*OI2fg(DGoB@lz}G ztH#l<8}Qs+-MKrV$fM$E#`x_QG47kJC-k%?1`{nm#VIz&e2Q9{ou5H`5HQ#8QYRF9 z`bmoEJ1s~fcllaQggI}&d#ySpvd^Cb&k{0kW1(+^J!;!1hKmI%)?fmgKMxUavl)B*`7?2Hd$y5 zp!;>8!Fp+&laKlXQ3?!2`jsk7<%sLhJ9286B`5&~U+DqN{Vep2U&+(on%t&yHPlCc z>?)E4W9H*EyC9(~e}{!4)2lM=|Ft1c74E<5{4*$i-O9r~c)c~zvC{z3d@ypX@Tvwa zGrgyZdRgUmzsgULlP^IhJLLou7wz8kX>kljKl6?)zovmm1LjlL84BlVStQie)YEwW z)mXlDJ3UhH_HvXCi*VGF>z$u<8eeIFU4JwZ+7G~m9}~5a&Cd?Lt)BXF0bbJVSpPz` zNFJu#&zKZqtyPA1t~cM-uS_APicwN*f|!rn!v`aSi{xaYtTO*IhsAv-*%d z+2@1py%*dGsh$GI$a?GaV4L$hxd7_XmC2R;Szfg~h08 z5S7s=u|dHwY>-B?>T#S($KkO*$Fa;V^%G6IsMk*eOIEhe2E$g2Dw^rw)jKqzbx%8! zEvIGAgC4KpZzmZrx&2z=i;4$d89iLLi{O&QsA=q1%h2$k5I=YhyV~8?DEuG)#=xS5 z?1dBXjp+RUwP_AZq9_UyY}>YN+ox^Ywr$(CZQHhO+jh_0yxsgmMPx-~F^dLFSC7n- zJS9Qy)Vf+KeNYaOTA&T5u>TG9I)(|gHi7Yi&YxMCtW(&Hr}$#9^J>zHD2)f0jkEL! znI!k6Z0XSPzIK2fqo`muWPPnDf%VLAcO_sGI}sfF4Vb1A6c{soi*FOT)Tz9XL9hsi zU4c?MB^I#rJ=2;7kU#rxInn`28C%PCdqJNVNGR&)fXH?~T0JNu#Qz+m9({OPVa%W( zs$?v_(8S&UNZb7F%T^A7$LVVTGX;fe=5V)0W3qv!Aoc>oqEP&80wh^pz+47l!srV? zUeu>XspHw`gGY|^c`5t0tHd2*j;u-To&YfUihY>ZAapRz0}*_ny}A|>!^k|61yL~n zIsx6yd!pt}o9S9Yqc?*cxc`P@5_)0kjvhO&1L{6`2ryt_KZ$FOL`IIQmos)AFkb?zYH5_DSGe{xB?qL+y7? zVB-LOk;|-X>G0*xIX^P)LW`Ao{2}iWj)lzy#u0zoa~jkhMon3r_?LJ7rYuTNcv1K?fTMFrrspwDekA?{?k4;8>jgt~VIeK~iX+7pr?n{0r#BcC zfI4|>-=sE^)*z^C_d-=`#={j?Z17B$LLJ;CCq>3C{+6Gnn3dgJ+<5#$o;N4`vH-Yw z|JAja*6i@JGI#U_=MoYwkMI}43>YrUzS75!d!nJ<2~%?(8d+C<;mk!S!VWPZGpOWq zK@iPz-u-px;Ipb{NiLB*4tq1REwHfp8ML0;3%T`ASdpfgw95^VexHhV zq*Jut>>s580|KkX#obOt4NATV)2hjxw^72NH!)Ux7khlBTXk;ukx zF&^!Vg|_UJ3*3_Yh0j~g>O%u?6QWyEXPB}+m?N^&dor)x20+c#`KabIMSTAb;c8_S z2MYMxlbDBOt$vwAkL)rrU@e;q9Zo>K*;rF?B5% zuG-~XXk_cn|CF&i>nlo%MEA6bzXn+9{3?|;)W!AMv#u}sk{&T#k$gSNCoB%XJmEYA zwhNy@=m%9uV@LN#3ny)FbcYpC=BoRft#Q{DXBdLt2em9*@6S0rW21dFhvOd@AqGXo z9f4zGicfk0wU2<~&63bj0M?+8WaYBmgUK%ob$1_VJNp?hc@zOT4{Ky+weki(D63z| zQ9Wcap4sPhB)U?OO(uO_>avbfQH6JT15geOL*flb^sOP zjw;P=O;KVy!CjG<^6-^~;${88#2ZtDup=)mp@FT4TwVLJ_P9867#thfrb7BwckFI{ zcIWsFBK!$4B_1zyg3Xp_pH?Kv=OvMGO}1xSXcxmUfD4Cp9O}6O=-j9~+r{i;a`%^E{qfm3JW>a`Kx#HSG0Kg{o4rr)Xp2wA zBB^|!Rk!?4H8g2eu0z`2(|ITJaMjH*1el+HYKXBrjM&b6_6s#gSS1waqql!KaN{2l zIm(f)CC+}NOekx2R7G~0Iz6ch3itNiUOEzca3~6#;|(=4{!+IOJ&>nY+PuTJPbx$9 zvZkK#4A-cZBsPs;CwQBN(Z>ft=xvG#*uM$U%roP&&`>5LQ!j8nN$y%&&;qZ;A4eJy zs#=I(S?DdS^6NVSJu;>*YO> zlaJI)M-856 zAqysTpS=FMLnRuD+6yXzoM&Po!hgvw{@5LAyro&<=~H!ypk9LSq-{foBRrpDIFGJ;j}|H~{&}(YnZOGc5xSz{=|vV*kMXsU?%L}+zw4irci5G^N1A^kXYcPUbczsG`zV z1SE9PtM1_+bS7Hme!7i|>I0wn;r6Eni_<*PFB{LcplYO)1zB+W9mx+c6ls=Qt%u(UZ^89R+|3awbNB%g`e9iKE#GG0k)^=Z7{bQV_f6AfpEPj5JV5AWES=*qgALhd$?=VxM#ks#e^C zM&bf20~LD?<4RG$a~Q#!KS2oQdNH@2o;n1HCh2yGl@Eyjb{hTPIjL7=ZPga4mTawe zLb{g}fwZ!-p5B#KH8u2LC6iiJKNc!Qbhx}Vj(9ROq$mO%tSgqgfR=$u3%NaM7q z0%a?PK}DEYw5KwGG>5);@K40&`2NG_1VPO>E9VtSbkxfUXJc z%WdH*ASn*>{v9{=LTd*xE;1M|pCIy354KmZHYWbsZuJ6~ln*F1@0LK#JGh)l5FiG7 zwW=skJy3C%k5s2aj6jW^DEhCAhsCqMUNatLE!ZD3V7|#J62th5U#FuFW)gbX?0$Oe zo4tp|oS+Dm@yXE~HHBfO(qrO}VhY$KKZNcob0;yF~5KGz6MyIU;W=D|Gv37 z(GX&ZF0637vZhfus~FKh#?;u{u28Z$w&19L@9xXM@W~f9MW!@wTdgnhx;7|z4T)nf zQt^#o)1O61)+Wnb&L0&sPmG&^B-MwwSkT%wB*MV^WfM@B5r3R(+^}_Bq})mIbrNV( zRkP_tcrLc~B5r2vr;fyNTO5W+CgfRSt3iinZJw^AJT1OzCpCm>qx%ybPK8*Bd#_xm zu67Y-bZfKIs>B5kyEv@ZER`xo$Dp)Qz=Fe7O4+UAN&M;{4yARV2x1Ak?frVo%sK2o zj}y~nxdc2E@rkp>P)w7YB56caMl9vX;NXQ6qW^8@3>Bp0gKRJDt8r2z0Ue{YvMxP4 z*H|y*N?YOnKR#_i+EOKJxmnGg`}sH{lz8l4o9O(dj*IlWHq59JJ3j&)r(z;ftFDVl zwdOr}X1!6o>!?O`L;cXQ6=`x9s|k=M_MF!5<347jh0l1n@4lgXTrM=s%{y0l!#uN* zOVuSEi=*8{HhqnM$B=mpzH3zAC+;h^WDzNZ#?O-pj>H%s0Nud0rmnAY=bee7%JWLF z;Ct67fun=x=d4l-knu-!eL|Z;NsKU3g+#~c`P7V>_OYKu5-L(9ph`&J#4g?|32i5! z*@v2JbHOD2WMxu-V-Wi>DxDUPNwzow-fehG)x}|E`&RVpj1UBZLfvqjY}ZCBi`yxs z5M>p@-5`NzvxE%Wy6_7-`E#pIhzPLK)O2O?GfLH@==Ezvr@0>XI%6|A-pPYPX7b1} z!oy6&&K9k9d9SmS>0^RGy|)~xJ;h*;^VBl#pLTl;^mdm_yU>5-sX2jE=2dUW3lYjm ztI#FyIV`ANhJBrei&fC-OrS&%)~xP?e%N9f7$%nN_P7e z2I*eK>q(^HirY8LAAnuQa)7J%w$Snh=qS%iV;V%sH^x3jqCjnaKdcbwd_b0Q=1EY9 z=M24Pg+48aic}5JkOmoNu8fGx64{5+L|5GRvdtJf`~@yFleHvVczf$@d!N_lqn}J< zjpGKd#|^d9v#9$Tex_^Rq!`>%M%1&tyWz_{5<-KV8O+Jywu>&0PqNap;>yoQ+e%^((MoL)70md8G1 zV2dn9oF1P%0-;QV51;QOxi{@YF1f^c`HcJ^SBYCB29|_uV=egp#+is2TPodon}IETK^Z!`@9>v(`W*L>dZNK$jBpv^8`0Oxq zb`oS)d2{WhjU%-P7?~z?Y9nePZg{gp91;RkKGx^~XX)Nk8QJfiTu1&*q8V>dt57Vn z@6>9{+JFsUs@ryWS?AmCSN1g?Hj5FGNc{{{#8`_@cDmU^77z(Nuu)H3k!5GSM6zNI zjUl!o{3~YIAUgrN)@UFOkU2-{#+=3H>0$;x;6B)|lKSi+2UefEs$I-vG`T*lJ0-LT`y7o+r7ikn(faqeSkz>~?~NMBVsm<+V! zQpFT8tq@hq!jBWW9JmK8z_|jNABaHbx*M9%pHz5xddoC>0<@4PBJ{JsR01F8xJH+% zH`(tyI~2!(P~{3s*03923I>gRu`Bm#8Rk`6OXaJ%DQhO58_6?363@9RVFje;dQSU4 zB`rACstc;5Au*s2t{W&&AInO*KS4VIOg6ME1y{}6R6x29k{ZX^m2VL7*UeZqF}^?k zw}#)_X$K}96&&ncAKgr{=@{=>k4d0H;_6y_G9V`qo`;IJg4_Yd+JZo=A}aS*JrAt%&zi6)3lMqoKI>?6qJ&&9ysUYCKN)G)b*tp50F?lR(u2m zE5QM8`r`NI$(osL`^T;yhr|$xO(hj8=@?SJt>X=&cFC_tU879f$>6m&B}fEv;Ac11oGB=VPY8Y* zxZ0W#^kYKj;I1?k_etVpYJav!&UNMUU<|fm@F5be#)1k;kNg^HYl(EzYVedSZHEZ4 z8*>ia*T_#sp3=jSOn4exs~R@=?qbb>9nDmR%(x0?a3c2VR<(S`3ZSQ=sut89s9 zPG3w03#sWH40&f``5?6d0W6VBvs+RM<`be>H+fROFuE>h944rBgkHz4fy;SegOVfs z^$|h+5ed`Y?L`7_zi#Giuje8>q3OLo*G7pMm$5z}t8YxMMN1jYA*U=1K*mRP+E7rIiq zk6-V1yrWHVOLf;|>{ZV0F2Yt+Q&N~+aho+#3ZG;3@|x@dwaeh6rY(qyzv zVdI0-@B~W_gL+@vH&<{|(4xm&y7;?{j2b3$>91O(VVu+Dthc zKs(nB%fVIw2QTVXy0mV80x-V!$Y|a#eFp~P2DGMSL1?nGWh}=$#Jp4>kFhpOllRDS zxtM!#nmwUY@X;H`H{zh2CMu|z7d?dM@w&7p5RC- zHiNT@6)>%3(3y$H)>l_?j16|c8I(~Lu7u@aoKTF|SPs)%&Lh-zoEVu{;K^nLLpVA} zid1jY=m}Bq!pdHa+X%8E^#nlSPL5?jK|(8=SWO4BRxb((pRs|j!`dLjuNWJqNZqbb zPzbF0g1qw*Ho12f+c*E{$E9B5d++{T=V?49selS3z!n?i21%$*;anhCWARPt>m#D)iySOw1>|dM120o3 zD?H2zl3bZbya`eXoSZ65{2KvQn+6V6G9_DQT71fEr>>;Km_~Chb^Nl05z>#|~wI1PP0R$9ce#wTbeb_$WXDS?kmeXWW0IMi`)Ng_Z+L zV(_XM0}yW8QDEe2jOig?RQrdA6JImOE&0>b_W_^%H`@-MzG6&ixnxo{`?HS^1e^k& zEy{3F%Y}ieSxaBXnZZDWJnUO_vg6lpzvxV8Kjf7N98E`hd@@=X@ReJ+bs0)>e%x2DttIUPL8 zrI>4VBG#n(SF8R>XNQ6hO~t0T9@fE3J5DrkZ6+=ow-RMlQXrnT)M zngO~d@;KWfn`Ho=PQ3h*1bjX6wAQLtAfzf;{vLIYqEn2z4J-M3{gjix5H-;2#Z-Bv z^e)w-zjTb_E-6T@2~Q=x3QYyiB_GeMQd*CnH&Oskb~Go?#tFBfiY!gQ7^5H>Zp65u zy%kkm7bvAcIg{@k2`bDO`!`4S}5xKPidDT)Ep0VYANRY zADTI|%VrO6@ z5JB24)vwTH_)mgKY?Y>w8?UP<_%F>;fL3k=7|3`NU&CoO%hBRZxLow%-zB>Ga6{$c zZQwvoWA0iZ54f!!>84`#B+b+lj_p_{Ye`;(;iqpece-UT{waP-92|lSEv)luIgUOx zc&R=L>LWvQ+2`kfUua;EtPF4l4;2C$v>JZ0!j>)T~9Om5uSd!b@klrrg8f-=y>}jVmv_ z&%Ic#U<*jwT=V7*%fDA*uv5gjAB%2yJ5*9TMDKLtWmK!Xd3Eoj<15Xy94;=xSea?n zp7NkRw8(cTt-FP|`Huwr*@J=~-tx#(0GTsg(7?6B#RaI^8%!mOE*RICNHgH|=eDr~ z_H@*}K>I%%o`n$~`i9FfoN;?RFEvI4Q$NDBd)}YfiT#8o7>7LZ#KbltfXt* z;@G~0OFGO-1{`*lS6FZTU(Q{O@rlE}vZRERD^RW%%>`eT8;4AMmC#Onw>X}*CKgUI z{46J~WQX{6lYDy5pRt)8{pz9NJ3Bduq3eW=4)eY@1*gVV8|8r7_)) zy@<~ih)pbB-qE18RgRp3j4Z;JftEXZ>z~5y54^1}<_&@@0tihaoN~3-ySM=6Ek55z z4+%9jVh?x+<}=vjZaNG*VXW3ODvEmRE@kOa$7h1=?3tumq=cD z`8Tp2P#)_wX2g;-N%|ge5&>oy;8StYPt90=tu_ro2)UKemkiD|n$}+GGH$>+Bv{M% zwNkA~C2%Akq)+9W`edE?u87t+wB#K-{oLb zxt=Z}Yc#zIulqp1x^cmtCLoEMVKAk3l7lmYTG>VPT)op$q=^|Gr-dl`la zZ)had$PwPKt$$x3?Ygxc=p~;UcQoKo?Q#Eg6da$$0{UfirBm`*C1BbWn|NznjdCPD zAQUWxDn1e06bt{@=@_}jLs5$VHw{Gx zp)nfEq;-}x&RwtcaZ6r$%JmlXgv!RBR|L-xzK29jUD*%O_=l3uVIl=ud2C!iru0#> z!!(V|jPLoT*+yq9gOOBQYGA-2A#;V8piLpMKgTqR6ZwbX94yG-gr^f=cLiA(W|X6D zOCv*BaZ8b37{HiS%M7-ofxITMZr#H5`+d-G&T@Oy=<{-3-)$8?UEXO$_dCI-MBDw+ zGK6{E`7KiuxXC}!x-SFbs4yX0`31j42)WCDVGVDVQ}5#)KKHhQW6UX28dcj2b7`bw z>WV^CKwx5(T+BuJvEqx@FB(DNIWQvBq>Hy}SM?@h(V9`#3KnOYf$|!eu{XLnFg^Is z(h)9&-AjdQdroRO`t;=CeypTV>N?6@?}}tY4&ACervwKY3l-Q$Bt1z%%sZD$vsde# zrGXwO&0=vzqh>bY=Z~r)HTxSnm(J01cJBN!Ta2AUkRZSUp2xOr+qP}n+_7!jJGO1x zwr$(aR#Gp?;Z-Gl>w6!%etpa&jtt=lIkRD-C2IQGx+EPOp-Mf!4;FwjlfCYQ4kNrd z+UjSF%K*RuGu2#D&l|$f{0pv+$u)Fvm^h1P0)8ft-KEoSmyyS&wIPL+ll*P#d@)>@ z9tb@}sj zyHx47tD*)zE6vIM=_^WTUL$M!D%XQOz?q)g7*(I`f5#1lHxI5? z6DhVb_sko|wQrFCn+#zJ>$QYiP$^w*mtx~-XMI9MQWS~qBbF7?@{Q_|2HA|7kgX9^ z<|Y*lAA?BGtStP{bQ#|5Kv3n0j5paVBBa#c2wT~5$xvy^c8fA?t!3px3pmY5jSRBb z=0`LX_fy1isY++LPy*C`!I+c#<{*twr!1%jPkCx?d=uD|J^uE4RXZv0AH~ZhH|phJ z+EoMS$}ed03Z6Dx=+-#HDM7wHThFKW;5L!{rOhSPaQ0e0$Im=iWG z6JfV!Buze&)7>Q4t1i@xft*?zW=6fI-$=OE&b6)Z&Z4X_GCB5R&IzKV)Zwob{C zr+B-uIAn@K0u=||7B*j>a!jfBlG@`w%AKHmj~ho2vSCE@4a`{x6lXKH6kEQm5tLz) zi!Wn)VK$o!ts(J+;?r=qI>#VhS0Sfx<>B1woGgk_^a}heK^$XNPqt5l_zdZH-ZB7* zB1>~8RxhvE(2vm5lSo7N5Ejxo1jN@FJ{#ezT7|@siT+4OALEWVDWX1P$;S06j~#Bi z+s=gwm)*~)yVW+bjhnvgmE$S5uc}-&gLAnGC$@qyGt#G9^@v?Hz_G;EHMfpa;_!s> z#1cq*u(2laJbncgi^1=PL~)5x%f5_HZJ(Ey+Q?#>)aG_Z0W_c%mJNT=`Ykr6Q~<>q zlXg#0^0YBI)nlM`lkiyyVW@DRe%U{cmkAY{;6@Ueb=ChwLIv4LoUV3d0vxd{Gro-F zL(!MH%xjE_q%pTRs?o`)weK2>3}z_0{KeAEIL!+r=d)*$0-{H&Ua|zht8l3!m*erA z;Bu(358-piz*;_V;!Q84VJG;JlPR(x+-PM;J&SW_F-t5*c5D|eJ$O1df0S%2XBTB z8B`&Q0Oe*5FEy-gVZs@&^DQtwatS+SO)W3_)vp`K^FYx8M}c zRo{p2i1U(=ons_j4+MpOw;n~hSgKxsU{cB+WDMRC-sl(lgxN$Zs&rD7dKDFy_$_>C zjn0PzA@Ay~)wREON>B>D2Ep<1q}6Ysh`z9kchXAnLwgv5kZ5DBfW+6H zBWm3P&ZRi4k1MIub)U~mR>oN0YaZlWOC3_=Kc1;{2 zvIe;Q6UsCaU5l>~JCa0MIBNka`1Lz8k96HjkKY2d)bTxGSM#iH&1-tl7eQNXFNSgr z6VcvY^rP9Smr@@MfW?*fOUpHfzIySI_dRpCeJ z?+Fd+yy)*Mv$w_(^m^*;8G(gAHTOo;dxets z>g4pZ4lMk?&4^#(6GmsWQ?q80?XfzE0o$^sfd(JbpBIuaD2ZNYzro9Lj_T`P@2)y< zueixL2rxpmHUOo&2xnV*Xxu5bj?=B7HAc5;M6BI zKkLf0Wro6^Uwyqr{P|K96Y4_$`-u}KU@voavyh({Y}Oj1D3BOg^CuT4u? zZ4`-~ip6=}*}J?3eU7Z!^Kfj5Gb}KRtp!Yne8IDI&&NCRM*Za#)rkhNBCmH=-MA@W<&$3;(g zoXG~=1B)Nu@5I2{)Jx1F^Gen36n>5}d4`1P?Qr*}iAv~v-g8WM+95)%aI4&d0(!z( zwte7!DwUS}H1i&t7#uGR`Lsvw_Wr9W-zV1F0#hw|$~{;zyalxwnebL_oe}m1{p~Tu zB5Az)>qgnRGKk1OiP7HfV${Tk~gRQqi~S2X{`2A{?gVQLXiuyM5~k z@lixjq4uwa;p%p~CPo-wQs=RJ>tf$FF=6kVAFEe68f8Bo>sSrF^4&-<2T_-*x-1TH z1wOG(E^Vc==2~c};y+!w+Em0TFoD{Vzqj)iL>%Zorf3em6?~FKdWfzxAlkW=7Pn?x zF8yFNjC@-ddAxMV3?6Ne10K9qmXfr%hBE$J+Ylj80(=46QU70H`GN9yRvU`1wD2bs zAUGXR1QylCttw>GFpKt^u+T9w%ju&C$SzxoZH{_{Aur1qHTvl-r#^FH=r^^!k|#l| z0~G_aiW+{kLPl~?sHe24wF%0ZHtZ2vAnB_Ke_-q+n+>>GUmt@efBQqjFcoWO{(Wc` zB}#7DqZre$r>Adbp^~ z+zeZfi;Ur~2gp8;E@T&JfHVIZIz53|DI2t}H~lqFJaIxvXQLNj_2E7ky2lBR+`F7% z47T1I8sYxBgK8ik!LrFuW9rGL0h-J8NQ$j)wV=#fj_^BFNp@IxY#M*QoY^Rq^<8!PU+BYHMa@r^Cfb=@+8+{nuT+yesf zdo%uf^`c0kpo`nnhTTIUM~^xLs?7U!KEs_$q8nQ_s5?w65cpUso93OkEesfAyQH!u z%DtcY!e55}%3!Ydk#B`KN`tWqV>Zu2)Od+gICzbP)It%*t0rjaNorxf7Qo&PiZLaX z?xS^*O;6L|O>bqpM;Zz}IT5E&hSIkUcThv-)+^Q334rEw*Gqt?O7F9*g<4A~(i`9yO&N(qYy>;w0?yO47P8rkuKe|CegyWY(!z0eIxM6gq7SKSui+>_IxA@ zxeXkr*9-h8DEY*7QTkhs!AN}ZWy%ZQEuMW^ZW)3zLUoC}oy;b5l1t_(zDObl0bOk1 zE^iT$5=YyLavV1JfkpWB3I^XG#CNc3A?53nPu@!o82Je!I1Wkt8t@U$E)j%9xbV*s z4RhjB5nNhD+67ec_CwMiD*fpvRP6 zT*xqrGBjT*>pcOf#L^U$Z7)=dP9r#Bf6N5`k{|{Kv}o6$phB)3uTPYl*a~jybhp&B zcu9zw!)2!*x79j5Nw(S{uF&q7cR4xYRyEjQ_AdQ_>#T0#Lfr#nD|yg{O@Oi1c)VP) zLJzn_V9s!~9xB{(+UWu=X&2YydP7He_f7l-2NIc8{;#+{RsseBJ0mM79-jXI{ul`u z+1Qx=b7LZ4WoP-Xv_CcuhW`coBlv&Q{+d->G?HvSQ{gNn0X-}bz_7E4wzC5XoM15O z``07@&l3?4;}MGb`-%iRNQ+aF6}U)1eT92pKYLzzd;I)HW3ShJj^DFZJ+`~D#7YQE z!IyN~z@Q=k1x5NC{d0OCCvYQ!{ROEg*uiaJmS6#aheY}eARYDeR9SE$1@q#q6kv4V z;zu}uLm6P`aACs(`WueA{K;@2#SgRK`gHjP@M8lq-aYyV{Q;g{`yq})0JlRt_qXhe zkbxl4hlB!>eg_JytjuaCVah16|CR&W zLmLV#ZY^CMtir?&?Ykj?!A#lzMDextMLGMpFq<}Ni(MehN4WtI__O0&hqPII!vFE+ zFzNHl+gHM!r+D3k1pqXM84;FKSsEP`AA-CLK@`l-7U9|QJ!Ezga|h(3XrlJ_D%iKZ z+Z{4UfaBm|oVnu%r2_*#BQ&rtLjNL?1b!Ypk1a$>3vM+2frSwVIVmXF%(d@>hWpjV zHZNPf)!`*3=Iz6c6foM8pg;Gpzr%>g?$a-zWW7)09UEEcoK#G?$tT-Yki0A z_?N#lYuyF{%IEvd!a#|5e)y%g{rTBoKv{s^54rzPMug)a+S)@u>!vdc75Hg6r^dFh zF3Zdac(NZ5{}}TqXFeVCIw)SC9#|US3%+ChGRwl0R6Fq zz>&j1(Ug1kyVOFBbQ}cpw#z3Dve18T1O%r6ujoWFfDFB=_NfjwBl3fviHrm657o1IskRhvJ4pF{0NM*7kg9yMgb#-0rboN z`4XpxrNPb)T*j@l=QsZ`t5L>1FnxPF{|iwctPgNJa?>!hbLM2H^F( z8e2dGBw}Ds2>_mk48dL`f1KV7IzR%j;4d`x`z|4dF3quBcEE4Yxqj>NZ6hUEjAMv4 z5sp3kvtP)WFr!6qEi48(62wgG=hyS>iyf@{Z{h9lgqot@JD5KSArpWE4h^gTNim57 z0%;NOZz}YslfxGpMQm5^47xUS-+qrh#`_Qj6dC?aq8g$mR#6VWg4hr>b;C>GeQ3&z z@IXX#0GFCs^Ogr)o06c!^0!P6P;yRcmd)HDf=L3Z9Y@gSy0go~Q8qR`V}|1&ZG&K&Z?71 zW}@7I*?1|>ChY86j_GC3mHPlso~d_4p95fv#)qM2!_O{>5jdBsY!{!SHA#epXF^}N7oMo(Li5hR z?SJHk+Rcp2=Nrsh6L}jdK)#m3w6=2|J3Gx<-8JHRmAbKW=hf3YtvTJ!g*wnuz6rNQ z5Y|?FFLDgGExTRQ{7Cm>a3G z2i#g41aM~f9+us!eC;$T*jf+i!+KzCn9GcKW@rXS7hGR^ce; zp<*?=y8lL%PLwTpVou+cN_9OgcORYk2DtW9Pe5@&d zWz&>4Wh9=fgj$U~mB{peq1ks_C5t9bL&aYyDjzT87L2t~7NTap`C@(C|01$^^p1i$ zslIGHec#H)WF#$#PS2&SW7xN}{PvMzmRn4q8DrX+^I%JW;}tnIPC>2w#8;;olFko%XQe$e^V*!G7xYT@W5)O~qImohum#53I- z9#pNxM-WD3gy{|CIFWLS>It`BdX=@Do`~uDp4DCYzGlyk$ndcz@r9t2iNDEx+{#4T zGbSU_lHxkqJd3l~V##}(7d^d>P0KzL=G71OB7&F`(^<#f`yE@cu;<(bTcTit6S>U2Bl_5QWMbx{Mp*_iccgHDjzuIW0-0*bXO(m7ea#bPxJ z%QH*WMsI&)HX_2xR$g=ae!FyWS#n_3^3v*_CC-0Pg6?DZTb=A7GCrBdyd&x3THr+QD- zVQ5kb?Zqazz_^4fc#|VjdEt_|JT)<7YM7(E-Zz)oHsxQRg*tjygxyru(LO&nVzvme*S)vx{4rH!r=pcTXHb!MK&?IDr=pdy_J_8W#uY$W@Wa2 zzQEu@R&MrX+jjFv+ig@TD+C!dr7~=f{;ta2lHs3`Pfca2Yuxg$-6xV(L)Bm~EzvHw z*xqcqL$TO)^X21%7~4y*Ds2>hc>b_Z zh3%fCzE3QeYzG?lNwnVlk(=VliBh0v7NcXsjCHrn(gm{BVYqCtuYjtL3l@bOBS{p!l5r-V5 zzXTms#e16FzAmv+cC2SFoyo=X(?OHA;K93m^gzluY@PX$Jy)0#Kx)I-DDPe6+K5tH z^jra^90QuY>pLcuW;E_wi$%P0Drw=-Fbx*k&B;H3x*6uisiRy#31Ja zY2mdwFt56Tu%vs|FEt)T8y6#S%qrovdj6}^)>3hVmtMtEimBM&>hMhdh z+30-yF(b>q`PpWPF)~hahVWG!l@2(5+JiE&dc8f2-R<~MtaZV1&=VbNMm={V*Dd=t z%n@Sp0G$1(oRHrWO6E4XHK5Q~N3uG#wXPb%T0%h0&WBsSBs+>oKwV&)3?^%%L$A>N z8J%%{f7NSk#@c15Rys4@p)@0Y8z&XOIr*|^AnH}1J8f|@Xp_-9AR`iM+^kAfQSR5j zI^;=x8RzKI%LrQQ>#Nhs8gFc2D>N)MO@{cYld@XN-WxF5&)^}*+x>HA zh3KPJ%Lg01t`|Ac>v;piPfT|ZuVFa-D!2}^mGs-80wVB)H^KWNfFL^Kj{<>CZ~6_1 zyrty$`WYx*IZsoo~_ zf^qOJ)Kx>T_(p42TRl|b%SK<~$4DuodFa%>Fwsf`_XPeIKIq^WgJ~9tj`X_hUiy>= zq_gbyr*{H^aZ7T8B!JmDGBrm#!M5M zOW&;u9de`U*rZyXC6QgLN&@FF9%p1k|5IAB<@Uj(J+(mFIz=bcW^%XamXMPNSp2r_&bqV-H+3N{ z?3(OKxBVqiI>T~YvO=loBgd*A$f5J~shgdi+laVUc3dL!FY2Ov$K_lyBw|x|W5;>0KDZ|GC4$|Uqzn@8j6>cIX&1fn1U;HFx&^tlN08R5 z7oG#rm8?s#cJ6m*z%LO+c5zhH(~YB`d1(|v23O>YsRDmd3Wk1$)qdlzH#KZyk~c26YgDfFNr=aOYYSPbl@Q3t8waJ>^}!^!=}@*gjuT zGtf0Mh8nDmVDDoj*X`L2_haQ%LP8g& zo=6xbqrS25pD#o|k;dl4gn8$FWatnHWY7gV;bX6mCj3|H9lwX=3kSL)(mYEqtn#3U zhhKoVuPZrBX$x{_I($ooR%E$Ki$d(Y($p2$>Erk23^yAR`W6*nFrtd?*Gok!W5S>6 zg|uhjl2a#v%fj?RWt*p?s@J zxTSMGhh^3}X-LLSWlV5G!OS-?9i&alcf5YFm^#UUKWsr7FAB<0JTC?^xwMSW|HLG# zrm90pQ6Tj|>D7bf+#kRKoxd=P9y@h4vTYHQsPAXeQOJ57JdH9Q&}P9a0NX*Y#gWPn zqw4UolSi(dwcZ(NdE6m6_n;+C%`YXk$WbNF`4^JM)VJbF(SIIug<2*J7%^N-0vftIxB2IjPUj_$LZt zvt)3Qc_1za51)2{5+Pim>0sGVsw<)@96Aa;(Sw|_I>!$tp zeMc1(YG^(=1bg!;^d{?gfMz@M_+Zf`{MpoR&;!NTCO*B|Dt1{rNYC!W-mXmIYvQu< zNlIr_7?_5X%d65oZ#FkyK$Y~UG8rMm|Ix%bxxtcdRf|R?d^(4ftEW2$t99uy!|GvV zu~T}~y(<@=K{#8V_#!X%44KOpSc#DZ*0;OR*l}%$C$&VpW)%C-vLU~?qB>SdDQs6R z7fhX5TRr*9K!-H40acG3lCWJRwtUSFn^a{$n4@WQSS+~`5CEj&= z^7nyB_FUTvWjV6TXLxYQ?XNrFrzCWBv+1_v3eiB>WueU%4rzKt{XMRZ-m~d2(oVPc zd%?maO-vj|#hj>o?45jpbb~4{9LR`bOVT-e&o)TlRi^q*MsKfRo8;lQrf9HV+$3@@ z|J*^3sKT$)>e@{5zNIx%bER(_D2raNw-i*OHVxVLaF~hW8jT^2e9@d+qm7~LCGCSh z^xw0iBRi^IX@F+ixLzbLWxR8$GsfM6nU+%b=RvG-O1`Y`>JZ&G<@Kv@u^811S=!|5 zAjuqq8NBM~<_pLu64)UI2Ey>MS=;SL&eq$x{2g zn{B1zikPc!Xfx6D*CTt1MNzh692<$?4N*En2yD}*iD@{UCo^FHyLT0FoP^kzGTaav z4iQ=DpHc}Uu5MN(QSKT1Kt)ufnYaB=#Bs6hF~RV+rRr_HM`zYa2@k0DS?-0xo>)3u zKR0Ye&%HM1qrnkd%m>qD_YD@21sm$!BU?PZzh;uIm}#eR;2S`e07|Uy{8F7tht2dT5$GHQ{rOpgUqLeH0+LkaT4XhB5i#ZO*RsHGB&gO z2B=HU*?xisKgcEJgR|D87Vkx@baG&Vm8HkR{B)g-(@6iU#Qlme1Y6YB6SpFRA+;J$ z6>(7j`OrHCsl=(5@ho8;EyZvvnz~1;KOWLvy@k6SI8^y(2Dr^{;$NBmY9Pi;-4D*oBjT7NTa+7{*iCOumn<0h}k%bleR9_JHH_QmTIQ$R;Y zp5_x`QyJ!WQN0E*+Vz zZK&JGv~7fHG?AfTr2FShPjybxsOh21Jw1Ehn!zZ>u~eb3-^ae_*Vy}$7eMyp}u zu|$Lv3r__;4h2K@yn3~5Xk8$4{P?=DRJZtOgL&XjiRDSr(Yw(ti4e>r{IIDLG+OJV zhO;QQ9i{K9`zEHV7P4QoVGa=*Q(=z%I`7m&6ugXnl05GdAWY)SggLYXSH>~4A6zM@tMw^PLww24_>AvOA)si_qrL);WlU0 zvctylfFk3Tm8e=Jo8ym}3&iK~b2Cbm zuurp6>hk`#=q<(sR){q_+HSO*{)gt&7+QEY-7Q6&cxmMQR&Ju%JU%J0asa!jXBQ77 zoo1s=UIM-RtaG(rdb`Z_aG)!%o#;%#!EMC7>+B+#BbEM#efs3Ikm#l z9~7>^)$hcePNkl3!Asp4#DgHnU_G;xnKxjMNzD$q_&#fu-ECdcNNsPatW%F56%;k0 zO*)~o@0q@4>C&Dp5WDNE(bw<)vN~FWiERx_a6I*s*4#nZ3MXHR8jvVw6VlFb;6X*R+%qNOClEhJP1US74XRi!Bv zi2{{RuDwWv7wLU_=8>~>`}+O;>NRbBGQ)f8#CPg5>yhhpT!b@MLDqwUKNl!eQaNCF z@!^jP6;k|3ICxHwE>;x3e^DVvKo}fc3}lZXq?@fRHxmtIbdF#WTO9NOpp$t6p4U&n zg>eR~1UM!drwvyJGERUx-~)&c<_~nP0YeNC3>>-Ej13MIQ*#ki{EyvEhA^IogkmHH zX`~PV!q4>y1FNd~B6($@jA0No45~0-0Vo0>{Q~;OfC9rn9x^CeL?A&3%v*q;1`=N3 z4gky;;1apyLXdCVL*^g&vGV z_zUtAzzZh0@lhc5${q#DIMBgOU|@M|7N9RcA%3D}>UWO}o)OqrDM;vEMQTqANMJrr z4t_yh85$Zo2vNww86!l>A2ea*&G0Lng&haTh z^EpDWLy5kF;VjTb9A}pTMTcL$$sb|c8L=TJgTXQc0ScKbGBZ~R#DW{4FUSeq`v9tK zB`1a31l%~aPW1S(@hSD00Zu@1^kVYCD`TX z=^~`a!qi*po|J|6=Or{gA#<9uCSWQD%mMLni_vnfji82%O}fY5$h`O#H5*I3=Y$Ns zH}lD-?ZxOkiM}Tme58K9%)&vhNK!o(dg|Q*?ZWiWJ+EakWoB08%_?iI>2h}ECNPJa zZ00%ZmcEprbM4Jn7TL9?&MTFjvW6l4S1dsa3BpgJ{#DJ}WKJjcd8Krqd1AdR8Sq5+ zjyiI1#~(Iu)}2%SC)I%#j+k`J<}ty>rJT#t((0@mb>ytv& z{J-_%22o>~0BTKny`&v}-XSK~3<_OUUpvbe(;?>L(0*r-P#HzHjFu6D#aq4nG_~5- z!@u8S#MS9)23rwvnXNv@@J(LPgfi2eW<~IP&Jr9kBUwgG()AT`yh=1m*QehISiBLi z&l?k0IEk|*^Sk?>83?68uM=B@&YqeR50}%JvJcKrS4kq-)F&?Tz%w(WcA}Iq9-YY# z6<0Qv$x$C;9+^1qDDo zu)fD)CI<#kwrU0Df2JVTH|S0Ry*5)TDl~j~YXAM0hpnZTztQPs^kYgcv1k+TrmRM# z9c)umsy+shvT|+}o5NXi+MPD{OpSk_H#?r{7vs8;ORVra!6k_bT@!J%UL7P zzo4|Q7hxgh(=c-8ICBb-urkio(BCQ?%FLG&C6pFdjwOzZha-iy32gD)!7D=Zw2MH+#As#={|jp^}`@$*j-6%V8Wqq>Id zcKEXWf?4U0n1AOSIX^)+RaCq8FuW#C{d=L-R{sV zI>)zr^{?*8iE-@JBEJR6n`CorX6|+zH0oVa^^LOPEBqPv7&|4_0^)YU4Pi?mH|Cf* zAuHcZp*)((vgFx&ByGL}7NC`n=}1-wYE(@v$cxK4)VXNl6s>`JX7Akv*y5Z7%X9$Fw{T26j9sN!DWi5ZE5DZ5D9X~nroYScfexa z>P^zuZ{(!(d_ekcm&*6Y*`HiA3Sz!7D^#3xIVH_Z<24n{+%S^xpAk`&kpLcRX=!A83l4O=0X=^-pC+D0I~uc1HOLND243qnaXH z`?kPQzIPT&p_}NViPk7-(JuAMt6jBMSTgWh&j;qk=tSIN3k>b4^XFL&Da5>k(mOAH zmxc4nVbT&&XCy5!rwr47`jdXCvXoI#X^j_P6UTX1x>6F=x_2=mM0t}{Xs=~ie3^2A zSV<&%I}SI?P;Ravbsv#<|wP?DeJtn!YhiWwSZMnT^dK;7BmD3k~@;e^o8O-)Ac0(d(F%zdu zw~|JClVYvO7Sp)Kw)TitIi?#~Xe3q7}ZMa^a2Ry#0V&}K&lx&^}NxP_D)l*W59>6;e3PC!KZeNQcCxH~k zhunJ=RIJB?V@Y`h?z#EW@Q2NgA50@fITkx`^W8nrFIZmg7j$ImmZ+3DuLq#BYT|;g zL{VC;?}kHU3^Fdjb~)Y!5BE`2@0{XfV%gDYU#G18KCV$z=d3vR(mWSK)*xI&Cd}XR zsiQ1uc`(euUuuJ(g`y(Z=2vzd9)B^tx*oZcZ7SlcT)DF;vE8r{Z7k8;gSZyOw(Sc{ zR_te##GHRp6}&RE9mNLyxyKQbhh%rVbXN$HR1q{h`{t_~z;LX}8~bt~|F|AvxEv|Z6_D$r*9x(QmSQD)3TPFe2wXLZ788(3v|b5_e&yDqi- zvH zr}Y~y;`;;QYt!c2r9-VHyawuyIrU=q6j9NMX~uw*SBLJ9FC)nmT8eMz1KZoL=Ht(4 zeet1uHuV7$jJIxekkEuL42}JkRbJtY<_70?EYx#NX$cF2 z4W1R^{EOrZAjNu${J~jvyj@Tf#_Gw|L&vox9oBzRJElU~e7@tzh?%!uALaC63XByz zV4>JPO16x4oghEeg!)v@FXnB9Jpgs3Zn7i^E72L=MuoWA7D|?Aq60evJAB!Hy5}qPc*S)sxMKNPW9l5$G+fT}eP(-m2N~~N z!X3Xg-pDAPBkN-0bB*{epZ3i?vT=#b=&p6|nrOD2rlb_VS%d!06Msn-wKuV;5Op}t z(0M6(GLsoPK<+$;=bGNDk21!-Z^Khm2_DnCz42oj3e} zpi1JA%$ji_9~h8KiDqfNmai2B(R=l5cOMI~J5{@nff`gyb2N`!^m1IB;Geu7@~*Qz zO)d`#Bct7E^HyTmm_{hLcjo) z*e6TUWsGaj;SKkt8bB> zeoQ|`4HD~_eG}F>>}=BT;&BmBmKzBj#}o^+e3sI&$CZ~Kz6*tT4U+$?rpxjfgubki z;@iD1*YF+0Oo})Zn)j-3z1+N(l2FXXuWmh_g`0yOi;+yHOTXI9&PMEa8P%p*IWuv! z9K=D*rBk6k_4e*`I&&n~Z+-!6DlYnb+XwFyU(mjl^VlQ7SHJ7`TwbK)em(qyZpFa3 zK^t|X7(Fxo3s5%;C-`5qA7+OCTl-P*us0>3mp8IfcD8|{mm^?g_^0%USvop76L4~H z{73&GU}0kCB>11_{~dtd);_nz7DxK2EyXYh5JiKQ-(4LTA805Xy034!8ghg&6jU^h zU=*DoIleq^?YTIyr%)sl^{59YK=bOTjO(cMq$o}*Hz{^*NLlGn5}TN~&`95-zC`9y zIZhT4R-d6{kvT>>+?c=)3a#|6vv{MTXcfK;q2hR=&?Oz`Zj+Kko_he{p~s{)^#Q=z z_nbwBpq~pQLlVn_l`XpI!=~Ec!|E2M%}B?YDK#xF9>mg8q!!%^k)n=TW@&D6u}GK7 z^k1}6w?wb&l}mx^>djXr!W|uwfX%dk7M>giVJW0OK$al4ieSMRUW?QSI)=$sWvU)g zroJ~@QhJi`X)1BG(9}evyQ4ugO&r9_XIe3Wm!>tZ2hZMgH|qvP$B8SMkcv`0o1X|L zMz*dBEDL8U0)vY*l%Z_ zEU}GRpQu$LHN_KUJO>$$pb$&8gg2U8G-ZKg{I2j%`zO;&wosn3in?wZgh}MU#w(#* zA(aRjg0Em0q|zHH%kFe@8=9Owl_kB25j$LuUBi`+%^b&;1fz!^o;}EKlZSU+8N{vM zeKKFh)+$UTAY6g0ur;kFluf_Lcbc^4!i{Ev}e-o3_?Th=%gMrJQ+r$1c zJJ|31^4&OkI>=2vyUUC3&B5{gj{m)zp+7wzr7tJ+wcvI4ai;0{m4@%TmfuTTMx3iW z7!B=WlGjXLTZif*vZdJj`f}aIhdCujAA8ZS@idB_WXX)W@cTGi#ZK%9_Pukre`dpM zDfTUv6+4>|cR0@N@G-wq+Ss_Cn|!74=>JwLIcw6a9}%?Ln?z2Ek98sO!qI^N2VX0` zR5!zw7+5P#LMG6@N-U)b?nq&cI6nG67(0jH%%ZMsCmkmp+qUhbV?Rm9wr#6p+qP}n z?AW&LLU(BrXaG&V#`W!*oq92-JkVqQ5ZAj8_(@EWOAPX8TbylfR!jyr6DTtZJ*QQq{osSStsv< z9?I#vyjEVIay^e%gGYp4QQoX)kreAa#GHxf@)xzu)E)ZC-t7du5;XVby0&+9wUBi= zY4&oE0%?w-r-Ic{Pc|pkzkBpI3vuMzz3nypX73JeoJ=kG81-R%U=5g9UiH|A`1LOv zc|cq<`M6Ss64W8>`f`|QQfXZKP8LpzlFuTv$NS@3DcgB=SF2wyNn-!?ntX0AU(VW`3kw5NR6S?To7!(mGOi#*4ZGQ+Zi3emWOhKzl#EM5W`A*#- zJC%&%Fj(IyBGBOxiYrzFahB_ZdS+iU^_Tl0#C7UT=!K4`<{rtwwDu4R>3&JTDG^Pz zr5tudOAjC^q55P4_S*iQp_ksxxoZOds?{}*3aMd-PFSETn_Q7aYO}eT+_Tp5=JPq_ z_3dm@?W?g+jZin++8}%bS%{iD*m?=YAi7B1u9ed8W$tjC5X!*Ljct25`?M|}gONA^ zB%z8rUpF>QsR$kko>Z|U8MmtJG-?cfnN+dCu6bjO zF=Hqd>}~Iz&x$(p{7^&dMAkG|gbkNR)aVI4)SSa5Oh*)5FKZKA1T>vGdl#{a6l`jJ2vyEqp_*u(Zytj9BTkTRzoRTsYI`XYC;m+N^&rKX?MUDu zfK90PXlBU7j}wkoP(+eWV+@(=N4x*(k9OQz2L#qXe}frnYuy#)|2ajZUt!G>{I`mwT z_*L{n?tFS8w)X6koP76V>((ZjsH{?Yci~wY=`cjPE=ZCVbyCs!0+~M9pZ4hxMb}l^ z5=``@ykv7XEaINSsuuP9DOEl@)=oCHZ;dP2QO>{@tu~JK-d9CHbbHI5 zvn#l87`D;gyych?^$X^8Sm2Fq^XA@jI6EGYGdf>s-iLR-{N8nHd!PWXD7sy)=SXfn z9In(?l+yq!Z;gEh)D5~S+ zmpX1M_%$gf?p?{l^QL~jhjxG^4=7Lf-XZh4?h-#_<`qLY)Q%~{Q>G947uN`53aNx_ z`3^4focxj_63ZAwV~R9j@|FydtUJJSUA`RHJ)D=2N-BG(6(%0Mrn5^EQ`1l4+E6KD zBW&^(D?W9GNVw>SdM8-!!}5rQ*Y9VvYW2E}!4^Lz=l0`6Yhg8#Np*a?u#!)pB->A- zl{r+Y4V+-H+#9%e)q9hsHI?d;o}JXW(wHxTMD$rWn z2H)>C<-ICNGt6Trs%MyOBt}d5;fSLs(LY@rf2;6TIvL(Yv zbAaWIDjki1L1=!gaXvsYYn(ugik)YTaV*}qGN zsH8?EPW`437M8C-8RmHvhW&+MiJ}}rje0qun}-Y}RunHuIn@E>#d8iKRO^p1uUes> zT3^AZzp&?xrP$~0s(L^YYPu{o_V05-bRc*poL~(fFt+4#m;G{AGlX-7%sOO;^R;EX zrE_pam_&$VFEBA-x>~i-f6>OC2jG1WC0!yqZp5$Y+S>eB+tzur7O)8v!>+o{zHuAB z(iXq)yhtv?5p2?*vN8m(suq5>0@Be)mo@J=R~AY`6;@N^2If z=xQBg%_RK@#20<0{ok*~K38?oo^)@LZWF%^_%ohGl2q5IZ^X;%fT8lUs!|_?%$eWoXJMs@waooWT18^}r{cO%xrD|4*uaR`V1nbn;xMECTT`5cd(*NFL{X zPS9|O$%{jQ94*yZZ0rjIjo#@t=g#g-C%^u>6k{xbkL%0y?!zf%asY~tg#^HYO}_=-+=5ohXBDo~j2%q4kB1UTV_joUxL<0c^?N@oqA16yc(j zdt}AT-M@pnOqQZORY43$lnNhpid?g@g-OIyIx(V(2JqV1a<@es5*(vmkUOfpIg$gl zP&WOppPl^*&!U6|n#wi|?bdeM71T4KY_*pjY`M%0LJHc?)<0&B3$COW&pbmiUy~<$ zOjPh736+?wMU(P~XcmhWt<MM8$ z)PtLc9bD%v$1=+pF}Jg%?K}{*o!cCP5NeJ-=WN>HOB#soVhE-wN7keO%XqbVJh>YF zF&`5?y)!g!v+)_&UO6L`a7YOi@m; z4t46vQLXNZGWf}wd&R#YAc6AFw*wOE9PB=LXW$@_^h!V{f_bYNx*St?C#Gj>;==-) z)que_jfAzuJm|~N%G`2RI%S%+av?n0)Bczw&EnY5#~Iny7NTk|VAhr<9@YQNwVHK7 zeoH15qtO1G@(GbNRuyodR*cV&xATb;&9n+}9GyI&A}0Vb=xxxR*WUEXhi$aYdg7I0 zY++Rd8&XJVda^wbi>G8lHLHPPq8l0MFv+2hFOo84NaqAY|GtaBW=}GZYb&>m5%9_h zW?x$L(asQ&4L2*RdBS6Z93aQ^z;h(1tf^fc@I5IL2?1(Skf?)UulM51+#0E_8AWbz z2El9iQqnOzgQYPQQe6mL#F_zfx~H#QC{cz%K z*s|}5XI(@XF`R0u=5xeVS2b4Wlew78`t7Zj`XHc&`t#H&52GXk!p^e>BJQL zsdHBu)z{AB$W5pTuB}9d0H|pqrF_xHse#^8raLe&DlS8sn*QnY^QwF+0C$bZMl(u- z!O}Se!|$Zb1YQm5a@Tn>crz2kNj?R3*2>|b&4nA%-s{h>QMyR`^ljaPFQVNP0owx5 z*j@lv{vNlfw3`*jqh7e+wCX6>O2m` zpGE@Rxd@Q?crU_@=Ad;I*=`s;MebC)@c-_Figy)<-VKXpg(*3_zkN1ghyQV>4B3~B z8Zu|H98!;h$%%Y0Wx*nheA)jJV?EMi?(W5!%`kXyP0>Q#)+Z@3T=&j%DG8J}51;^wS=%Anxa3Uwkaw07jsw&T>2C-uuJ!7_TlcTa zXtQ(fonOyF^?p4rkU!1hsZVQ_8oG@B`YxXe1-iQ97G2(_5Z0bt;h*{*+upCRxVeV& zxS^gu>*DkIe1(;H&1YrzNJF)Fwxe@e{>#t zAayFbc&j-@oujzL=>6;~a)qLlLCh{K?7tWBFqAVdozCrul?arPr!<`B9r@1EUoTi| ziG|!c4o}mK%MpHwfm(PlxOpfrTK4|Oc`&6;_JYilJ8`+twd{O>6lW)^O4rvGEXq}kO)8Ew0R17)6wnrQbgyhy0Co7>9IpUfgZe~?JZ zgckWz5lKckxB3@3yGcp@J2>}x#pE#-x2#vrSLxlF_HK^LhLHlcg*dqP~ zZSNDBStjBH84$&a1kD2p2-rmO)belU`~?r*I{)Mf>HwBj?F%NC#+x;`K7pSPDRf1o zfCM)n26?=H1zF!1`W?|f*?&hn&uC`>@NYK>7dx;D5oJl?i>4$!Z z5a=?I7cC_d2qffbz)w(59>;Ya5C|K1emQ&$fT%t`A|__q#16$WD20-uI*}BLkXbuyn1jzmd z{2}=J78YU|eL=Rp*Iz+UUpc5e*m_W*Eur45LxXRsY~*TDF5<&mKU5L6A3)lY1;Jww zH>6X+`BkmE1G#L8Rn#p0`&`58%sCRUW}LBk+9MFU1BdutQ{X@A_L59JCLK@-O4_b+0Mp zA7DzkR5|VVmkOTc#v5M5qcRRx@gXbNgq+}ro69B)w-^7t}J z|0?FG(x(AQGsZ$401oBx+|!}mcTlEGK} z+MitzB9KE^5RbK0r!VO*FDKvdcE50B`nxBGKkXoWs(@=EJ!q_;*MWP7R!@WQ#JeK$#K%hCB>~@HDls7RPb* z;eHr7ln?HP9q9wVosNODsa>Ioi5Myq>k#jV_+|FCxuHn0zZ>l+5Uh65Sgtn-c5@F(mef0B7b zL;&g;AYwP76Zs zm3~3p{b+Lr5($8$UK0j|m4E+x2L~twZ7TjPJ`nbE#1E-}q|%R+fM{TDVtg2Ahdi)> zbJy|dAOvjJ=FG20ZBmzb2q;JZ7vpRsfbXF zi2i}0gs)bg!XQGvc{hIU>z@zyz@9ARm*L*qS~WP3A0B_K>q;o~Ic5X!6)QyGJ`Wv3 zVCS%cxcx8^BXDU(3jw9@>;lAzUvq)Zh=x={f_k{KLb|!$&~vK>q2>N(zB<#r$vzN9vkevk(5dE zoPMu>LSxN^f%kSgw%JYQYaByPO5xga2{bwRl8_97z8l7Hqr$57;t~xB`bH6{k0ip5 zre>sS{;-MUws2!z87Ny4iM7m;?bgWs`s5DBqH=3Pkzh}NAkDPXPpWW3WMY1H{qPf zSOYe7{akFh+U6#^PzO;XtjAfTR4^=WU1~tcGv2 zx$-S4_Tg7-E>>9tCm79$k|SSjDPCv#I@!a-#>{DOFurSVanz)n&R^8OkOJ_XGPCynK%^r+idX6oQGjuC6}0t7_nDjyO(|-R156=iMbn+U4KmAc zD+%71I4;he<>&t3)CT0O&FV$)#@+~?5%t}sl+)P~O&x$Is}1czGU@!mhL2InWhW;# zxOwlTL-H6}738h#5~Eh^+J2Rm&Y_f9pV*w8QwJ-=@Er4VePC5vVeViiKu+o#f{i9f z*s4NzaF44B2qeL2|LOe+Yvy77^H=>3F&7x!HFV6n8gW>?r4Mhl_^P~B58_*uI+YGh zljcRQ{zCOFIlH&GSoJc?{U^jQlF(i z)R)9!B1~NX9^d6@U}rqP^Lp6XyP<#Rw*U0@&7T*gaR*qQ%a@=JC4}02fg(|S(3-Yn z+IFSuJVti&-j=e8Y)rSC?^BuufQm*(uAi*7GACA%p|X|pT?lS^hF$Sn6|bhC$#9cp zLg3PX;2*N6%Tn^}RZS9`sk+rx5_?Go#%yI$N94X1q%+UqyOi+XXev2jXB}~*Ri5ho zk%vKao8KoeGv+YtlZG)ripnSrhzr<$#7fOEa@ubWv@#f;S@$8sDORU&K&!s$ zXKX|UlT5~UK@NpY=i6a+)Yt0o7%E4B^Z1I)NAQ7APXGPc; zT6!wB-AO!i3zMHL=Er_W5#T+Kj4i_Wu1J;2{ZQIIU_)?es=BeUkA+HyH=mtlUA=KqZ>#9KHnq5?hBf z{$q!xkAX1deFF4X%LwoY(&vdeu)yjU!2USqDNhmP_SJED?mp?4+{qu~SaFAa-U_vi z+-BTz*A16!GJF8{ap%Fg)sCsR&uzx7`%oWyAa`)R;|ub}52b`$lgw zMx;jH!#%c%!f(465Iw7Q(E^wWJ}RY-D)=;%xaeZU6gnCQYX|=TgD(S(>25vMZFrc zUMRpsxzcFoNr*YHzBju=oMesXOCTn#pB-5KvljWAnz8sQo@dpjn@dPM#Drm!7LH%3 zuL+e?sdK5f3MR)H+XQLX!oc$f?E`*nkTgB?-4`}C(=t54|piz3BwxN7vW4jk1LKN!@1BVUuW&9mS$E zVtv`;+&}%F1|di3h74%*4becCg{TGCeq+CQ<*W zfJVRHVeg=lEdC81eKJm@wI-&rDP2Gx^{M? zzti^zn|yFG$KEx9Fx0jj0)JyMv2a2(Cs$jI1cEtiUDmboBDtorCp z00YcC?$+N@;V#Anf9OoOHmi( zD!f$MdfpdS@OvEt5pLZ_fqgeouhA-1&YiiIr=xM`uhSeD=;a{g@+95j5uJ82}$P_dKMLZWx{|?pE4$S_5u8H2IQQh=I-o9 z`0PJ(!iUx;4n)`WjRn`#$ zPmbuC&$HhS+ueVpi@T&>72Ve-x9nGGR1&JvwCu@Y(asZ25nz=jPnI62+qc3=8z-yovu`VJe5_U=8qbF8??+mMjK`6YVry^votrgJ%t;thFz*EV$n%^Jf|We_ z;AR;MsExZ}#x~Xr&&KsAV{h#0g}Yw3C)O*aAj?JwBE7q=)U3@4;kvly>nC(*l!UPA zr{61~AU4Oirtu&$!VnaSE!?>gPy!H3`y~II^edZ#Vi)_w;~cg9{W3{m*dyvEO-yx< zuPf)W*~N5)d>UB69*c!pe0LkBCJ9>Fc#gxKK3s}i6a7Fgs;9%Zg|6CS6XY<|%I0;< zc%Ni!bVE3o`W!jDYjCF{<{3w%m`#HBmb^kC?eu3V-PIdupO%^Lbkf>V9^@IOTwJ6AHN4X4n?|_j7zDaawf>iAEomf)IAlDPnpreiP3vwWomz z2JHyd<#kva0C1T)gkIrFl!(elu7BfY(GL<)rqI0#bwWAThf7;V)wFb{=$6Lw@ac12 zx`@OQ8k3Bt{$LmNx!X=3nK*PgL?~(=3_>=)MG@s$mZ#zC31L#b!3FI%rIE;{xV;gf zVv%FtWqINO=3p9p%%3Y6MwI%anZ{TYDyAX1oDUy#2=x5#Jhg> zEmPtbTGt$raKCN27^v$!0v;JxBRM9WxxA^>a{nxy`Vp38SiNgeW4Ejz0|fjrg8n{_L-s zYvRiT<8mRb%gs!gxn1*o;oxx*-AoBs-I5Ql%%rkse1Ly~Li3^M39KhXk2&(}AL1Uc3eAx&)>GYMbrf9P&dN)INGd2!W^+BRHA~D>N)D478>1=QNPi~Mmsvm1tCxWP; z?8jGe&r8-#@B9H1IV*g=fasf>z%J=lt@KBlXzEJRgN-`X&DXycV{goIKq4!OH_%yh zuqSrrRam81pOUw6V2b=l{u}JSlr*tMt)vAq3&(-^iGiiR#p7^dw(B7jXy~132UR%0 zRID9vn63BiqAUB9zz|cy!duy2*`_90yFJScGi);D;q z#}yU$TQq3(H_xgmN^pg9Sh=%}gthMcztCp)hlCo84l%e;$}okI$Y5v?x|`0~s+7i_eIK ztu!|gZ3L#&vyfkHJv9u}>0%=R-zHk6HkE>T8Z6`yqs-0g8=P&o&!C;jnyK@UZH$#w zc5K6>@G0Je%NKNY-Fh5kKp)n5yf$FEl9?q#YblPS zoBRw0mi$;dlle>xoC%N`%}jRZDFeZI3uMcgvyP|Ll&tNCyDAKpzcH)qpqBRe@;Q=D zOP#<@g#<&n?dXNH)likiwEk{a^epF6v}n-0KE0T|pvogcR6q|O zwoIwA`Pn)YZ6IU?sfXzyxkzuKO>OKIEK&>6#)oC@VN9=|ZKiZ>0c&^3nX-$!>T%HL z+JjIZ8s$XH!F|?T+~Ya+TrtI{EsxVQC7bEg>L6E&z@!%)e0$IC#)Pt)FlTw|XEJTB zHUCxlQ0;~KmIhXtHKns!&Xx?%_DSgfyq%&aGxD!k4Nk2%ERsVWU03`Oj&V_G+x~Es zNeBtl9wTs?#tiTlTy=Wb=xiz14d+vuyP*JK@C}cXJN({!$d7LGnLXH0b4NZFTgf^! z&91!_<79H-45oG@(R<%^sLOWh@K|GU`R6$78TpG-`1&PI0(yy3-8-8(m1Dc}!NecX z5B7kf{Bmqz7;$vC^mcBa|Hti)PrUjB4A0UFE|v{3dbz(1|}3DOGV2Wt=5uOUt_3m{fRDJa!qQob{XowZSbpCT_I0s zyneOM1^LiAol;|24AWHnH<{`+Klwd~5cuuA>Lj$~s&0y0=%LR_XMV~=aeaq#e;k)~ z#w{ilebD7w|C;++k!gOGcoPF`iGY^QYGATub7_xe4Y5Dya-`dvTyi)Kec{P6l!?k=^cL7T zU&OQ+$e6s|I@#u>2a^A2m1GI9^&rJ#WmfAki26? zvkc2Zo%2&-qBSyhwN9M;_0*-^444NRH?7_UU7iO8hS&UFny`U)T)at&zYxvW6Mh;L zCaN4khg8)~$KQ=ptj;}&wyczvaWI%_mcFzVD*W1uH_qH_yEi$I)In!8Rp*yf1&MuB zEI=g(HGcD&Ne<`7QCJ#b(*YQsARYMEt_Sv;9kXX=+95hvh4#AOMJ<(N!3t#HW~??x z44WJ?skreu>O2mw{_R!e`%>os>&SHa**-`FNHY<*ks_Ei>JK)f67Tkjsu6dR33$kh zHYID^JFcs+{5q+0Vp&evdtQl2Ye_d z9`;zgn4)Qx8O%9Fu3M3U#pJ?W$#XJ64mkc~6`tsv%i1#2mfT_d)D^pa+H*Ss#Gz!% z;?~gd;+wpywo7EwE1N@~nuJmtnLIyUcjiXgT?e)ynl4yzKAS1w=2Q0cy{gS&`ZAF2vZBNC_-7XI z7Xa-CGY8jLA{6fBb{x9Si?Cdr{USL@X^pdh2jY3ueKXGvO^m1gKS zQh0&IFo{O4U^>d_x(7ox0lmE*^{l+^d-8dJ2d_D>)8@E*oh0M6uCK3aHgW`Ol!iYtasfRwdRrC~V2ZDIenLJQ57#^ZtF0+#(tLtD1i z&8yhe5KwLIZ@OqHC}m`!BF}aes*K-$3fP3X}qjndfiWa zepP`?&ii@is5Xb+kS1pEw|`Oju;@RrPgk%h9$E%S7evxY6jn&-fz?4FuZLL#a0=EcJ#n4?O2x;kROf-yZAz7gfZ^KUSt7&&s>jENOio+jaLX5qW9Xr zWHPJV6lo~VnpE@Q(lhKUGuja9y4k zYEle5rC{Rj8_tMhD*RFD&Ucib{wJ~=II#@=O-m`Pf`UGlOqi3!y5Mrops_`+nxl^W z`=6>{uamlb77;R~Yu}`QBV1xzj@x51-W%S#`AC`|C1(dkBWkJ5=?$iIB!tfqJJvox z#JR|6T~vL3)`d0K^LpI>>4Hc#rF9NJs8u=q!D-UnobPn27Ujnd+M)$-_sJEviu;ul zkxg?>w3jbdYdg<@w>^&!pBEjMlB zbnEVca~AD|bC9l|dbIdY1mVL{P7n?QmQ;?YP!BeGVfHnGPiY`Iw$}D9mM3*FaMOZ{ zpsn6RJvnHBlE55GPU4Gg68*?FqrkUJu?}z}Cfi0mz zX(*hy3pt(hXAZ0r0AmCDLA^&eki(DGYw+}Uh)kcDD{@Ro^UmIC9q(nRAzn}_Lz=!c z1$bW8yaUh0h`!_zc?L!{8q@QquT7w0q<2%Mg_H6sR3JmG+H21-|1S-fREE>!(NZG}`poF%5DG}l8=B_ z4Xq!@$l=u~XMLf5_Rw?uZ5QW>UOseP$NtNE1i{xxcWG;Ma*#JVl(XasCuF@cm5`BW z;u9?+jl;LHb8`W;rJe*fK@<<2kfwl1t2ozyhCaGbrYHxhoV`1Z{)SuUtgA|=2HSim zR_pyx8e18!LhAKCJK#$5C+s$EzC!KqeNU)spWDG>An|aN>7u4bQ(2&j%p7x zEu}VFh_lL+gReNG-EI5C7eV3J!Utolu#3g@gX5W(@7#!ls#Yu`KXv= z)$7Gkd?ofS(+>*k>(;E#Mod?nO6>ua??;ah5r=GQ2+`>Oh`u$slCkO9S}eXyyasEB z>=(F?ydrVuiKxuN!hGuxdyYOJ4onm>dxZq^ZLBEm0!~klw00(B?n%WWL zkOZZmahoP7x7B!2eA;uBUf-dQ2N{t_i_$KImU0Z(P31>8{?ST`eZGt4tY3U0+Tw%Q zmR#gdMZxvWHxd*d8hZ)>_DyT$CbHE7%7$FXg?zJ!9cJP$gTdE^jZhQ@1d{3v=Y1W; z`!Mr-AQE@MCGG%7KK3*sW;9iI9ac<(jP;CB{d)uE(sXJV0xDpAU(N^rg$y5c9JS8!m;7N9Hdby}GWb(84f1(I~zDHVPxof^(H+ z<`ndabv@5$?unM^k$KVhk7kLPW+5gapEV;c^h0`iv?3%V4dh=-^(D)Cud9_>Hj%Gz znhEj)iFW(d#e@hZ5-P*xI50L1g~pMKe}IV}+(H_HzZx>d5Q*_Qu#oG*Hj|uGU(QD* z)OJ^H(vHa}JZHOd_ew3)xSQ<8Wc_F*ykKQ|?O^O>Bo73#eYK#0%boaoXkO3WqmqA_ zQqeSnABEM7KGPG`5A$G9c`$TG=HF&|&M4dwyNk|*7wF$Kr;h90b=a3#hSk4>m}gB4 zM>pxkUjx4G#@l**npHGH9^??absaGkg`DGI* z6-~O}>?_EQK{yohceK(KyI6O}DsV=PxySp6B^OiaRy8GI=*7BIsOf})T$B0Pua7m2 zvM$9}ceh?56ay4H^TA=d8Qvg!aMGn8PTq_Md8vq$JtgZG2*2%o1%+Yvyc|%t-W@0u{3|uV)c-o- zd$LDI4XA&7UZx)_iY1*^m{uVx7x)c47?Tqk%NXIJ?NnQl*&J>0fU^Jj*(6s;UXSdN zbfJIQc3Zve$;_5(Qe`#E;f67o4SA`rfzAO^66|XBrZ8TgFvQ&4q~{o0!L0dsl^n}@{}QmWdJSsz(6LPQY+cd# zV4c6v6saT?DND!zPQJO=P^Yp?vZ!^DVmxH8V#_t@EY^$W9W?BHVE31I8q|r6klr3k z{$f*o0?vWQiYNO6w7#H&fwj5CMn}*h=x=1==6bYISTzD}Br-*@tv$cQ95(It^Z2=?`nTrss0*zJU2!*Pq<3d9d+s;v7_v;P0-msewljzkpq% zj8#okW|wNdWfC-OMtBYm0v+rH(U>)_^Hfz%ei&@$uW6X=ui^u~hMNN4u;pCYrIaMq zKX2eO94TFRHkQ&Qt(q~1!*_4TSG{HA;zF4nPwTx~Qpbi~N$rO+SL{R}@6cC=CX1eq zCPg&atAh(rL-nfgeEY*N*G%|5gYKxm|ILQ}W5jmH!hwU0=Sl*aX_lXl-Yl!ZBtEnV zkKBT@w~qJ=JyEb=Y?sOxuRjt3!c zt%Amx6~Xi?B&6}(@^_YI05o8v#Uyois`lwDuXi)m<1=^l;7?G?%h-~`ltpLZ&^gYXMNLI% zPHAyF&B9qsS7!|4=9c1(cs!m#zTXu(=dKQjig??pk#L&l>lv$5I@b3}BERA`jHz{< z>w&x4md;8X0X{pnhcMhnolgy?kp*-H+WzxUGVNFxPT)=&{eDCul{W(w?o#``lxZ-T zNO&jdQxmJv+Lim~p)ml9QjUMJfP>P=d2q!%zu3C!@$3`>Z3Eus4D`mR<3=@h?kBC@ zPVG1(Lzv9V!Prjg&8tQFTXfQqbCt;Qq2F;g39Uh|d7(2llvA!cs7CfPxGtMQEKUcV z?;5J5htw|$ytz`@k8TgkYf7hrh^#B;O4?JAxjvmPNFQ1~rNq?iQ5S9*O!|mVL#dJb zl1LH8l=Y`;JOn8&k4_}5zP=E-{a;B*el_(OFzfI)!TOu>4?djOiS4h2dCa9kKL^}) z2una3C8m0NHM8JdI!ELb$brPaV8%O}{N(Edb8a#(ySm4^PHIete7 zk{?TE3?^g-WB|u-BwU&9kjzo_KU<_}_sdvxy!JpnR|MdlkR+&u&wjU0i=5mnugszP zFJbxXR5wdbeM!d>>FXDHpr!#kS6;F)G2To_<_V!c2*lw<98} zss8@Wft_IA@V)h&3Q}d`8WjC#(7oNtj8LoJ;j)L8$XtSU`l6>xZ>gy7Y#MhLln00V z?RqMX_PbW;TK8(3V0Db5Br7o9&b2q6P}DE_<5adYm3x6jM#IyS-q%_$QnOvv;fF8< zP4=5PcozzzKNV>Q$2JVPD!DXox2xa0KijS{h8xM03dGBu`5LVlD!;_i4kUaH{HE*T zU~Oja(g)lRYaUDI{~;H7hSaW^k~xPZAUm#@<{`DxDXG6_GMiy~);c%p%dTHSnky9H zNjkf95*$b^@_stiR8f|dT`XniC}nh`j~asGwLa7WQnJV&oO-&BEm2}`9qP~<)wx#U z-EW&<;!3(L4oXYtahSNbvEoo^9t=-N(=x(XzT(8vgv8bI?vAq?3PWP2#?h~lqpsJi zJPbJz3I3F68pSYLwd~y`#PYhD7MfkfzhN?I-M($9g9{Koj@x~}V|pSuXkx^ImnD$% zWa~M-#uGMpSG1P9)LA1Exl0*Rb+pg|`5Qn|rVguef*eu$z>oHu8L7$OD`_&`0?dAO zCY*wPEbGwa%vyXa1_*Qd?aJ7FuQy>FUdnB1!TAOzZ1S@9*v=*8w?KBBF$N`~?hjw> zzfYhkA5aL29cKGFq7u!;@#u9HZOt#NW?bEUq(u3oq-6lHz%jcY{p zm5PtnU%svtcLGEU5MX2nNg#!ZYJc8DWFTcEdg z)6UA%Q{#o3R8>^qL`a%hYq%Dy@d>lcBN;rS|43jhoTEwn{O;kF$zZSuV$NyG$HLCAfe; zgya&rtmzuM9i#t&#+E=-DdgnHLDCKyKWb;cfgk6V*pV)MYh6MdJt&b%$T_F$}p_~ z5PM8)WY5O_u-GGp-1y)fKNPv?qrTeX!S$v_JjU3|tmZMGUapX6t{BII>FaA$N~U7g zfC7xStL{V1m+?%V|A+Z+EfelztcB&*ln@94Ek*US>H`gG+#%GnGq0k-pNNP>*m42r?2@=3W+8G1*oIKk{SrqQTOrV4=r^ zPzr8s3dqCR;#Hg0V8L_v;y(~9y>exfT_7DLW7RquY4-RP`t zUyHQqL3shlKLt^-w~2^-=DOZDG2cHeliA&cr#Ld!f(bTfT_z%M54m0WTI=wTMP|xM z0cBC_oA=tK((fFCm@XzHwc7?RL@YwJ#F6UUhK}dZk38%?M7K5nnvKq3I9(opO^=6+ zN4?)x#2Bp#&@w=1@(qH#i#fV5HV!N0j&~))Ea*nSnW&v>QI~ZO?eK`H&b_G?$35@^X}GKbbfuF7 zXeyb>FRWO@Z}aNh05NGgGz2*Ku>b0P8&@NCSiPS z94y!>mPxmqq#WborwmY+u~c7I>*OEfxBn8|;ZIg?!b_%~^S0}4?+RpN2}yF*cPm;c z#Re0IFYPQt^1Z^b!_}`c=6Fpq2J!kTNS0N+T0qTljN&E!x#?+k_4xWSTV%=zEs>>O zT}O8khH#{*{Y^M$(jD@jxEU6f|G>>Kb8vJ24>!ZY&c^y*tN)G7uyQc}-{rE(9(+kw z-xM+q#?+c&1kzDcX$EADV=Tfy(}{y{SYy)-jy=-;A7kefqY2b?>9Wo4vTfV8ZQHhO z+jf^-UAAr8_FL0mCiydye96q^Sr_MOCntOFb)Mxpl7vDl#Yx+!LN%g;36=7s{5=7V?GqIcD<=Z_EnBjaJ0{;RA0uD14KYEX;OUqAsa9H9mXf;|KWC-)*! z8mQ1=!}*8?djjR&oX5Wu!idpHE9=cW0jz;ALTxONEZ`7VDAT@R8pE{$P7>L@ZV!3=%*tmQ7*QQ{oK(7CtNpLNaU+MvF*z9Gre*X3dC-aAXYX z3zDkV5#Tr$FC?51L?cj^2w;wb97mwv2*SBG0+?Yy{t2RoU_cH)0^q#B?l9xAAU00U zj_#!f4i?5y1RTFkzGH{Ogj{z^5ZAO4D@lN!<{;TRlq-ttsh|qqGCzPaqeeuX=R_1c zm?N!3pPfjN;Y7@Vpadw!4x2=c{rg)XL?OOF;c4Ob`nT4B1L$A{gQnr&1_lNkWPtJe z`+3OdGMJ;|CyX8g88R@z3i@~wKxa^h1AbBDLc!)(6LjMIjg*BX4nqly@Nzm?0Oo@Nu!M;o&aQBEbaDafU||ahL_o~Xi-d$hz5R{T#Q%U~aybDC z3WyCDMV)p4;2nIMPr0A1FnBTH%N^Nit<|%O1%rL%@;f{}{yAqDGJiw{=!hdh88{0d zd-a~ptj*QTu18%Nh!-jRk*jzis9}z21l>Kj#vDwET~=_0 zodF$kYDmeTAQ~XZ;p|%G9GH%nv{;a;EFV;VL~pSISQ$S^L#$Hqzyf{v4BGwJz>1o` zf^d6ak)8hj9hvcl^?cm$$6TBd0b-C>a~7o-h&?A#rx|S!0SPzY&j1}DI8^h_ETBGrGGm4OgLu%F z->jJ(bhk&~K%E`Pcg9~*P`>wPux($F5Lto$;~kKHD-YWIvmS_FRa?7%C&3fI$|5Fj zA;JPK6La6?UsXKYL(>EzX8COwy@CbQ?~p*Zc@I`*zlNWogyT&amGVxYF13mF%4Q!K zT7C#U9>7h2QGj8O_gKspGtbt7Q@(Tb5$zJ zDPB3AT$8oH<8b4xoV!`@=@i(knb{GE_}5R7rRO#6F=X}x)}C7xL;3G z7I*nnx(|Ll(BvC6v$IQY;msv|iaDQIaLZWbH1d{3&1|#eu$w+lp2n!Z6uYug1tCrb z1@D5Rvp?@H=3FyLPG_iO_O~AdFO&loYZ)`2{Jqgp9d#VirE8AozD=RbPe^)bU4Z9y zPm;>racUQD$-X}PlPoi7QF7s%r8qzMK3?t8oo0G$uer9Dd@q*3FhFB7M#T$~3(qyC zd(`Jjqh(ue+cqjYD_s9-bOc))aLsO`qAAX;k(iCYnz|0^=Xs`Anmn-DnUi4@x40H_D56u6yL9RcLpM1)sZ8Be<)58%uB1#_ z$1l}leLi(xc!KND{sPGB?V9O*KaKg7`OSQQ(G{U~Yya3)>tv7p+I|1gP(|ViLC;a2 z29u{>;(5?6>VC{i{IRfB;(O|>M5un)u(MKZB-cDkGTz^=VK_S?Nt+vTKdW~R8(sf) z>RYQ4ms|E;CB<=s!|+`^fy~^SQ|qi^XzBL)Jib5QZ>!bms7J*)zE6vYO4aQ4d+m9- zrMcC!hS3>EW$M-GYz&rzhxb`hh}F$Q@4@l@@)ms(h0dq|J=|?4T4m!Blj_D<^4{P_ zZmRqor+gPB$@F`+vFKx}g$k4>Z@pE#7k_o0G?o2QuB2mEC)SrF^@nYPM)$9T8YpLl0PIMR_TqN#=M#!OopTyivsBl~LEkkLxxjft(g z=uD!ejU(B4F6s6oMgIbKDvuPV63DntE3(mJ70oo-(pJ3%0Ry|li(_(&*u^iseZcGA z5vJ!eNNJetN4;ED zKGCL>@E3AYAIUwjs20=6E5raSXU*C^Ks02P+Jla#o z(IsSM`VM0*nAnS*t#X8N`{pI)D1Vm|fd{n=7v>a+Joy#TRfrda?Z;)M2_Rg~Fevh+JdfdGq zI!E>LL`h!>zZ~z91C%l%QJ>H9#3@r=`Cgfjkr1gS?=*IAG^*RxLku5yR%Oi1t^ht0 zj;tg<#vw5##qWe%-P)V&qzd%3TCWL347U|0v=0e@YA)RU&b27KYe*Vs3(U%2*%h|P z*(B39243?v40xmFmCT#(KHW~+2Pcp9=Vb1rM~yl>8R!noA(EG&5WhmpvyB=x$(gEC zVjEi3ZENUk*Pv(_q3THVwlbY$E;|*o{d%ApvBUHQ*78S2>J72ZSetL7U4E2LU!vPvv;E@$|So9DC{vIW9n&O-kUtm3Y1;{ zn0{Z4r6`f7hCfo-86_^?dan=5Btt?QiAlZ%epGf(K*Ue+^qXj_QxaaE4*9G$PPAz< z;c!v;w9R^iZqAKG{WHD)T{g>BdSGu}|QhgaNR{q{Ye#2M$;ooMRC2>&O zrM;!gO<1e%o;K#KywX4QHSTWxO74X(R@oPqR-%7P;ABxjb1}mvJDfak@XK$E$t+5X zJzh2^qdydLNDvH4jdx%jLqEt0EO~>u0F*NLlbH^Oo6xVSW#y}p#Oz=k$g_&Jqo$1 z0&z>x?jh^VQ2X$2{MUXhnlJBlRo8g`l^n8cjkSZ&{qTffu8X&3B}iG>{Zb}lpjCFh zZ=;Ws@mEJs&%uC{#y?_X2^&=NB#-XYFajFvk+8HqW+Gc3^>`Xk!uL4RGgGO;^FhWS zL0de(SDLLQJ9YQn)b5n$DE&olmFc%NNI$CX7y7zZ#kZv6%r&iCQ_d$j;NRTbG<>`f zk#jQ?Gdn`=2&SCu6AY_sBy_Beg+xxmDGw-Y zJ5&1jE`)l=UeUGrL%M`6ogFv1p?SP09yB$Y?@Dbo%=bxnY|Z>_TFkscW?fzTE&(4d z_xwZMbv8uVj>>MU@lNnaeC9LuFfk+@NjAP#JA5Knj=9;aeT1HR$%n+L=_ecK^@cOO zU3E(QpHD5^7WN6$Pj2#iMPb_kn#;p3Ec%Do4N~Exy-n<$potZ&&p+-Pb-eA2_VBf9xWtf_#i2g8 zHK^tJ8`@o-jDyRQoZOS(*Q{AqXxSI3UhZkH@O;`Lm3k!1S{f?e7iGnfF?9yhO(Cu_ zhl~f1GXuf@=Bc(#o|@X46bv?4D2mn{UY*|3zlr_t4|?UbzQ6VP7M%I`Z{ruddy0uG z)+fI_7c5ra@_D&;E?)S03jIcP`aG;G4ozjNuC?nVB}GnK5-eVf@3}I!&GDbDU3N45 zOWzH(;8cFjj?7+d>~9j|KcskjoQ^yE${`akuiS{u_W3Nq?b4CX4fDB*w<9%2E04e4 zTiTwtE}|-}dU#$ayM91Tkg4PUYgfwh-@4NO19Hhkz{;LYUzJ#ym;ZVqxf!4c&>aGZXdcN=dOOVfses+3ySlhv&7ZSiKz+;zr-eX)Z*OlA zT%LxGMHVs172%umequ8cG<(4#?I3UD|3X*%F-BmI1L((>_sM% zeWWko>PvN_>cxF#|LR3*PtLIGn*yeR0uaYC<9sfvtLy7yI)oD|av*_=6W)b!aLmBM zu6?KM_Z5Kd(gzU~oP_r090h!r4t(XF&|j#}y(MdQfP?k>0=5}BCSWgr#BBkdHegtx zqDGEu`pANRkIZFQCGm&FeK(MlKcjvLD8J2usN_^wmpZmo>=AUn`jJ%MW|wcJ77gTmS5z z$tg4*!|D3)k@mrDK&h}1YhrIfFds^|13>ja|3N=dMN!U?l4S~NFN*yg?7$ixFUzMPj?uRJeegxVT$VqopL-smr&Og5-zZHuD zYy&vb352A?6o>-?^tBjiFg`q_nFWa;F5eS!?)CHw2uTqQE8HEJr`tg=fEY{Ial!bv zCs2_i(MRMF32?-|Jqe`16hQ2u5eN*xo`f_=y>Ev%(bbr3Jck_bauYfd4?(+I^O%xX&{{eb%==}wJ5B#Su zm-OzuL86@hO>!SG!Zq0Vhk%d{RKTIH5R5PI7GD7;;O#r)$0i>zKK_ zAIMSDQ9w>Z0ROr@+Zi!`EVD*r&O!HOTdDn_3GvNX60VO14l zufM@S4GVr5fP<=di2b|4y>Pfz2ZVYqC;S9liTHPHWb9Rd<4;wVkMm|5Hwq~KS*vnK zO%!%H9?~a|7qQ-Ggk$lmulVG`z=vH$DA4O6wy_V*SI0&Bo^@AvqRh7fO8g63*zwg! zywu~_9`N_{+%l&dw>{3={qnUBgzAYImBg(0g#-X6Y3bB>YK~+~!^e9~t5ZF4-Q=+? zsd)nUEgBIV4WZ#18_Ha>UxINV#A1n4?e<-B=2*~OHnyN2MaV1JV>-?kj+`wmV6(&M zlm#^otU&x^Ap8~+Gc37mE1KY;1r<*jN!CJ?XPx-= zYkggBf`{OilX`=#kLxT|P+ zK_!L0$LrCm4DR_<+v4Dp78oS;dz9uzE={m-kmLHph3+m{!Fmzq>_3OQio0M)00h3* z5vA}VX9oGi@(m@z`1OhUZFMLgCWVirGk(iQnNUQt;7y~h0Hm#j5x%G9vmE_r9?>g# zb1qx7)3nWiZnt^s)}$K2a%7%xOiZ)zH$7fyLD%=T)^yhwT@0=iZalu< z7|ZpOJN>JvC%X=%b4`?{F6>?}itk5vk(0L6a;m&QWZ!D0dP?X>iQfd~$tW}mW?M__ zVCF<%8TzcaOcHJV%fi)eHWN8U>Wngz@4vE*pb6dL%5B#%$n<+P8r$pF!HQIrl7qaH*z^3e3Vpn5Im^glzv4QjcGq20uiao)S+xWzIOvAcu4yxLZzhBJRN87{GJ316+K?F#WlT*~Iyqd6T z9W76`t~|k6-{DGp*b_iwMn!??P+}GrEMnuBHU2)D(hfv4Uz_?$@)l${9@f>jQ`zZ&$K?@4FDXH ztv!!r);p46bRZ9eT5vGM76Q3Zzw)c6IxP_&@auB*2bgC z@k|gM!q9VlV)1^Si;hDHVl`F2VnyrZ@isq=I%Cx2-=q^P6_efR*-e7C9Vwk-1jyup zO$@g**GV`PPbf;%RaD$+So;Ucl`(i#q;7e0#W2Z_#r6msw$d%-zuPQ9dXzoa86PH> zX}-G(F0}K0gx?%h!-GSV>(VwI)j`5&@ukVNloxeB(GlAJlFIvOIPUGIf!*DoNuxbd zi4#wfPmvQw3*+%7q3zopc1x{uqN_GM($Y4CuRl`f8XYd+RIW}Fc;0WcI>mz}GL~YV zkLkBlzs9o_fs3GKOuKe9?>fP(CZ9~?k1I-GT_H`CAft$G>cHD2#b(9h{ObchI1RUC zR@LwVDU1Hso@ER77K?*r-(%g9>xoX1n&`Us(C*^C81FQtfj+$|?M=xfu~mUfd$dNe zC?_;$?tA(0HN<#CUZ)~FG~YCCQCH^9ZboB)#8MJ>vQJZN)I$zB&s_YA&oY=dTVi~7 z!O45TDMut$;Wd{3?-nLfuJwFH5(Tc|v?$8&cN$yKW?oJK|EV82>k)_4eH%RdGe_aF zG`ly4_kj3b%dEXCL;12y_*@FAR#G;l-HOL^L^5<>i63g4d3Nyrw$?J^{cD708818i zU{q~(I!%9_#?v(%1i}=m%G9!|Gnd*hJ}hq2GB2u$^ph3CzKdd7w`fFqB)A3;`r_({ z#D@8daTn3g{YsY>JuUwiOLUD@Pp_6{X(=x57f6{ZLdya@klmZS}{~w5=NgdS~z9 zd%zt<@2M5?A<89USa6JfZ5JDxJ?W?@L)BHTXfLlzq+@AV(PzzIX*H2<`q1&^^r=qQ zZ=Mf*8$?GnqW&Z|Gsrq#dtFFO&n+4I!5PRNf0tEJ#CN+{sf>C!lq3K)%yRLRB9Fgk z1+2d*`35Q_V~%BNu4S_xoG3x<`#jPruytI%KU=x&Bf#IUJX775Xe%ptodulN?sB#< zcB>(GyR3pHO^uS=cJl{J$ELG+bJLf3bs+rpNR6HZH;hzcf0$`G*G0Aw2`y_g zk}BLt*qLn&RmYhDdlAvIFLJ2DF`83kKWf_j72^Rbl7YM}9UH!i>`2JQX+ZX=v4EI|HanPDPlG{S8bzg6z zwCesKb=N&{`-%U%{daoyuPiJ7!%AEeczSD;?IkaLPh?w?H6`31S`t)-)s9a9V0&`f zOWGdR@*Aoo`Nr)k)eV_v(!~8Kjt0)MBZIx!i^ev!wAEsjQ51tXaqiGT} z+AgiW7&G#0F-CPUmzQo(EV?d9saFcbllGLOBn1`DS_pFo<4E!*%~25R<26iGVBWNh zb3el=Ow;hT6eK>E>>rkc=UvW(tz}fo;#u6Wu81W9!%5>xYxrAexaR&ge3}Wxi-vkb zmu3+ER%7=gui#G4W)Wba9bX4^$a4gC%;S~T<}{}oNXMC`xHXwt`p5A`hM;)1G1QLt z(&k-ns?dUI7Xv5bifghHZFuz4r1=A|$rvwBF<%Ui7+ruC7S&wb9yiuFx+vR&Mp>-Z zhy{P#Y1EB&UzualWLY>bU({R{KY5k`naE~k{pVf=MTrb`#b(BJ;JZ7uW3#;CK;SH6y+Ef^ zuitd`fzc*S_rY3Z?0Uh6h@8~0f3*$i43VdWc@C1WE1A7pfVxm(P=zA*n-T&~)^#J= zdKlqxZ$>QRs7l%Xl^&;&G z*{my__M6eebfJ!5z?vg#FvzRb<@0FO<%Mu3!}6hU1I-znA?i_vlYlzbb!&p_NkhCz ze`JN@H>oFbq29IM@z54)P|(>mx&2#Li^734tmr+C=I?`io1|Y2Zp2o2k80s1XVvkV zyS06qs{Z~AntsB&^9N+9E~SgTv*|p@lZlnbfLZDNdW!hW>hs1+M5$o{+;@A=*{9ku zO2)`t7plVv=)3-zqwpRnqx6*73gK>~>ASTAn<9IeGE+`Jj(zs-!j1M^=I6Kf(OyT< zKk6Z5JlpUxq94-}>|3?Qk}Dj43{zC+4iO$N5!K!58YH|QRGfQOS7h5MScj%Tv-gZR z)tuN(;V4g<_O|dWL7oSB4k)4LIM`kLzDQB=YTHktPVymmA9d40PQfwpN+& zIgYarxy~WQKw7m5c?u_ppHi7Pqf^{;$?c?Tn&-=59EQy6kyIcwMV|S1sOvcPjJEh; z9i0-N1wDv&vr5XTH|O1YT52N0k)cw=C%}2v<8GUh;2uIDFzd?dz0uICx8<@c8Hv%g zyfW?kZm0RWM@`Z_k{gi}Ype=Fcw7i3UN+^*xw6$2JI{~j-ck&LV+mb-(NIUgQ$V3m z&(R0}9$`&R6}CU)d(n6?%95V}LBOW2HUj-5kx}Artt&})s`NlvYJTE_?_Ra$eU|Vg z+Zy@Px+kl2N#c95F5ioNnx4G&Mavz%C@cSNNrko%Id#YAD`J$5o30& zZ*Fr^i;Jpxfmi#CVR5kfeY3-w?3?}O$E)c+)(-z{o^=Pl ztqwLxozLytT!}slR{$x zN9e;0?^q>m3KuP|M(xTZ@x-2&t3v2A+w$~KC!bp{GGk(Rojt?M)JZpTZN2;0zN;ra zd!Hai57`E@PHM{Uh}X5aYEhKfxoQ_Bi8K`7`)xc5l2T;V24c_uQcX2E!UUFNYl!z&q>ZVo02m+@>KhDHqvqm z;;x6I6FiSb6O=NqtX$q}@dQ;WyGRt=TCt}hQ4ZP9@@fvd9~F0Dfthz!`pdymlp4x&N|Ozw{0X3%!9gtJy2P9<8Yt%UEGZ zj4t#ba1q4O3oR=Su@XKLuD6;HWs;>tp3$@5dd#zEDt=B#Dj##{a-Y4$Fa)TWM8_7f z?rN4zLL^fAtI54)J$fmt!+1(OwL9wRf8RD!*?T`1TAPao z*V+t_{!=i^!(v`#)-WzwMi|a{S4ELW$qwCtX8q}moOaV^92uI|F1xK)zrpyi$b}hF zkNT#EwNp>e!^<4BLn-Q3v?)x7jbWt3OCx6-CD>m4tWZljwi}J)tq80*DV{LKLQGi5 z9Xet|6bPu<_h0BUvW#kbV~5uoxyz{XQkah-y5n2Lr;tG+U6ECy#lr3qrDNQ$wdu~$ z6(8a*<-cY!t}$D&ww&wRe92pGP^cIc_oob3zdBLmRIBKkv(ulYwBAnMQZf&G{|M%B zVz6o2s{d^KTYdrz?R(K+wi(lycpIpSFOG{Y5_WGm2{`CC)w&t?>2pO+K8ImW^ept5 zm2$!V%t`rE%}HGc`W6f^HyYVA$kn_&Z0Q=NOJ7%>ohF()p%$?(<0Qv~Xra|hvt}wv zJ4$krZ@Mgc$T4CJZ5w(@cVnPAhV8S>ziJ|Vq>ROyi3!$MsSJ+_NwM=|*xK?yqN~p% zPXRcLV|@i&1F86B{er^PeJ093l3OV8N4ZSouroP4HkI}}SGztbH}WZKc2^KAltN?H z@mn^duj1D+b(1fRiB#q8k}v5z&h@+}bX3Bav)4%zjsuB2PnHm5yW^*}8+-OQ0Uh~y zWMGk=w}dmflg#f2wA`uY&Tip!zJma5fBDWM54G^PmyS2i!D>@pDmD=KPjBSvz&ke8 z@}n*+&{&(cq7iG66{dx|gQSd^NUyWTEzq)*%#TYPj9b3y0 zhtanOSwgtpjn*b#CgHxN(c(TLuiGm19o`n8#5wb*P&yh9&vdJ|Pw?}9msq!D5q!S8 zNSvv0!%I(OBq*`U%G9sT=u!Fj5fd{hq2`R7G$=6_Fh$tRI3H7Wb`xl?5kT>M68eC$ z#o%_oGh}>Fy33k267;W^f|3|9bH)pkAYVVH-O;5tCslWgR;XtUd%d$ zq)*>KS!j~zn=)UonMy1VDYjOEDH2gaHq~??(LE3ApfzEdd5BlA zomN@4!_uzh4#$~r21p>=3^6Y#3Q%fR86?UHlp;1^1lzTqYMmUjisSQ*@4q_8bm$(J+! z6tRnMgl#JL{4h2Q%AT8V*z&H@_8fTq1iN-j$n=C!PZw(1$V#b(%iO}4V-xG&;l>pO zwGY3UL~E^YG2mr++wQz5IMmAJ4t=wpr#|j*u5w>kXzc}Yh6@LUb zbm4wFSyczx$#8(r5(XGqsN*v80gPI^kihlt1? zgGo?)9j1&g{i}z^6T9QjedKgG%S2NdImk|zg=)6P8#PiroU;0wGSY4SL-O&mIg7@n zag={`KYET!u9&h*X}_BI(1GMs68JNPd%aWXabTh%CeYGM8JEmrZcfHZI6_fJHU3*L zf0|BmmjX_=g{QojshRf39+yr0gie;bHqw!GWh17&oZs10xT=Ca-}kDL<76F{FSsCO zJi@VMeRbCWAkvUY$^@$d^5Yp;8JG~uVRRBGD^98*?C8vVeZt2`8)vBSL1iB!b_bti z0Gc1P9F{H(%ovl~hKpf@aS(TW)W-d>Ou6lAb@I)mjtJK4g$Icb;d}s6&6=BMlX=&K zqGtJ~K$lEngyy2ktk2E#!4B<~M32*c7RzYzeu6XM8=uC(H$#gJl8p_o=g4p>rFX^D zV({|txsfN~Bp1blqF&>JRHBK>1LRpSySDEV{tAoDKBOHSgVa??YG#R_8h7VgMK#W; zO*HY@@e|Z~w*BGkbT7TT{yw|f;J}emCm=+&jRQuP&ZBUvOByh6*qCo|K24d)8)HNK zahEYT{jj%u2%*%`$ZmvPVo&?Js{e9hvp}iF;J3(4WdXOr z6UUdE@TdLB{ma>bFzNoSqy=sGm7~k@0}9UZSoU8Rhiw04amdWb@IUEzOa#oVEKL7% z{J;FH?2JtR+tB*|&dqD?G*!`AUtx{u5_X{6+GcI<;zm>8K)b;sXa^7K>hgAgB@EJb zfR%F)N~mA(wI}z@f4yb@r7%_MT-)=)m(5&IsSCeCdBLrlMF57@Q zmzf-_ifVFr4&?&s@W;{Nu3&CKsv|G}&&|NxvHj=liF?bV0olm>_2tE^>E7kl0JLkpp@Rc3EWf!jgS!Xs<_sjjw=ED?(1&$+ZUb&v9!`69 z3)H6uUSMMm?c@ge?UV%mQ|l7cjB)^$==~)Ow*VEd&xrfO2H74g@FQgm>+&=Fde>YB_9ITvKi}xrE>c{|OTEgncIb80cSMpbiiL zzO>e>J`?~eecw<{-%x@W*hg1@F3^gAxVu6@E-TbS;K41z69_QR7LQ;cz@G{btHALN ztmc-6m!KGKgPGD-M9rZc3>@SiO(+AX{+LB#_-ax8P58M4b9HqF+W?|j+)y3FOF-Oz z_cJT^nXl8B7Z+D0K#nx{bAJAV1(JJ(9YJ)mak znuo^r$L@O92^^8Y1VEZ&izAp|`Yv7aXf4etWQOSkSo`Mc1**y6@%bqZjwvk-W**zd z{rlL44Yd&QV+eN+pzj58(*#F%1-Cr;;b!w+ALq~xs=dmw1=8vE0V>GGp2G|UWrEe| z?gsLg8#B-j^Z{UD0C8tCuYn$bINU!$foTM9EBMh;;L``_9;Bby4?!BFf5pB5Un%t^ z0tM)Pu%m!AOnwEwaT2ij6Uza0ZP^cj8>IilwgX?m^dTb#=-#sjIVgU_66mD(6GLI9 z?m`WSKL1DhM~@e1ulNx|qQmM#4v9Xo2iq@t#ujY1_>aW(AL-^l(kr%LBlcG?7~wj+ zKXC-<9sbZbI5enBTXXPKXfgU3J}2TT{U6A&>6z{QHtG_QXnq~SD;ff8Ks=UGdMLKZ zYyc0A;0j*f@z&4S&yMep$X?)GbHEkm9R3JW;xT^&X3{*$6;$S$f5iP)5(nqEA_!y& zUo}5L$jGe_0qPjShr8O2K7^c_fcqxDNy4v$z#rw2*U716)U&tv5NstV6Fb~MIe(L2 zmWhw}Jb^Rq&F`6+Z&uaF^VHb0OdG;M?UNDFiS-UI>@9Dm#QT`#B~{;I=}Ute;=u#v}boK%20VJ|Si zG{ONKARS97cha7fBm)-df#rME<<0>h{&Bm$7k1i)?vk7%34M?%?oMTFAS@7p~hNcmhBbv)rNtb*sxYxgK|V5^>lRA-W~p$gvK3eg6eY z&xpW5@n?v|SPw0dzc6BtX)e4A6j%M@fiTUGT>A_`J#0Vxa-(quxJa?t8V)`hPpI0l>$qn0ExmK)6^WuK6(cJJ?iRg5_jB!n#68SyPZChgz4{C4Q17%+WW+@sV#RPA9puXqc@73cz zxsXpRknSGxp#%%ne&qRKgdTtNkoy;87J?}`}0FTGH2 z?iB1q<8m+-+Sy=S*dn6GcV3LqmM6NsnXT zT@i^{8#Fq80X;u=GWl5#%QimTA(5g5<}~``Z@eHTPNjc$M4QHbs}wu@i&TYyw}*L4 znFmQw6f}fZKsoZm)CzvjO_4F)FzOw2C|n#ywRHjr;T~GEUME~3dOL>+(U7k&{ZSg~ zujG<;qysm@yiiH4#E-YoZc}F?J%T082v6L0u22(07c5%;Jv_#+bu*g>1E!?5&>l?Q zA@M-^x>fU0^Jg|aS=k7idz@iZ`;i5_w7gB>4q~D#BZ6wRtbt2)tN0t!d9~Zs-e#RB zMnVl2(S>GfVAqt?JWrgrvH8Uest$(hT1qBVh2$Aa8R=AQzGgGi8tb4@?VQ zIV50;>D!~*7QnEmQ4od?K~@;e0lO&?iQaH_3SSG$1~{K;;C$Yp9Q^?66b=u!Jf${+ zf@Le_wVFjj)c+%>MV)oDFN0!S;dEtQ#L9H~0nb_ANDCYdDun?gFu`z0xPC>YKr~ zK4vfKDi{T?6u_+gi_GW>KrQ@-ni8a^;db_SaFPqlAsKdeR}c{$3~z$y`x4uiG& z8Ma~V0_}6%Q@EU1q{I*4RqnbsY`m6oi(x3g~iwv{see zEmt=+y{@BLo`;&AkhB>WVPg-#Q*3JqCl*T>CY;SCoISTXlF}Ztq8K1LUD^1ZqTEBN zsLF}5J(W3oQice*W{%^iZxnIsnWgAqXcpRdGZQr~t@>I)`Z;qc?2dnRR+Aj3W{7xC z91Dr4{V<6Qp2TOMLnWwS)&!&4%keGtQcR-+P{4tFooE!38I>a~{g*KT06H1dR;z)$ zmOS)vE(P@B>_gtPv!8^N?exQ|&x$RAI2KChY15MAjD@}r;#=$PRl4X3{|U$7vGm6LZB`ZPJEHShB#J*bjKcftGcl4x>aUp1UH9fwh~B6? z=YuMBv7Ft$`WqfmC8)CLGIxG{y!e_XH>C83mhQyz0G{pi0#D?U(9 zyt3)i`O>*CQ6jFw-&?$+jSay@mS6R(UJsD0oz{NM70U1jKq4XvSX4@Vqu$;Iiq-a3 z+4#T@G9gF`b{@>GZta#|`wx37J&3&q74USZ7{exOYRpV4H>_WK~R(Wt>v z>RCg!v9tX5bO)>^s#Dw#D7MGXoUBAU%jtg_wSws^JEaw&?>d+EyZP}l?M zkVv?~B^P>f9;7G+>8}HWcFtS-qSk8<7CJ10r$LcUy-h3d&?Ny=)%7->Fu6+xO&=^o-JDiprHl! zngqX&B<>5aL%mF-J>4m)b-YWaQ zU&od1q8%WI-Ow9bk=6%vU)dYF?;hZ!3|^Gqv4q*G*3#po9q|g$=x}@M;#HO@F-D9y zrLKSXNivQPp3E(8>2sW*!@{vPX5(avlfFKWc^9O|BT()p#? z2{J1>WO-;CMcV%MeX~I#>&JBa7RzmFLa#FBahR+lVE<7|^34gF>Fr(7pFcyv=Y!xi z{AkBj{z#OoioY-(S1p&eyokkrV8C~i-PsSsl0MNZ*)ozI>rk>wLnh#gbCWQQeC!?< zm6gXD(f8%0RYKn=817Q8Jz-o;vL+Vf`mT*_V5ZgZ5)6%z7VZug;hPCuT9% zo#R5re3jR{DK8sVcw__lf;itMrWfORNI{z8OzE!VNX>D2TT96OE|MtpI=Sye>e1Np z;{}m>)8l{iLy-$T#d9S*H($mFw3VNHdkV#FOC8`ol;%i|aAbOW?coesdU9J{1>RFZ+$9+(XLFH2kZTT#|@4W9Mx|VeCkNmFcNp zjfuz^ae97x>oez`&e<{zJ+{_oiu^0sY69(QG?~IT^3Wb*Q@E$WSeU3*YLy2563x2C z=IEHmQ|BQD-I19RM_&uxF9yW+SEN)ctMq$t6Yk@DlJ+x{%8O80D6?iTZ>cZ;eJLuV z*S@rgE^@kzf{#QC$eNnw_*y`x3EWm~*}wcd1czRpGN=ql4Ae{W^woF?lDn=3s`mJ% z81H3b8kuIT&-nJwb(AD!Xf<+@hLG=uK&|Gw)*r3z7H}9aZ>B8kM@t`qA7m*VKj{TUM$Hr^RXM6O~B`k zn}W<1w~L2ymkF(NtYEO@)7gG9dUA*+vT(@R33FRaFO#1NdUq=P11j=4UfVc~(+*Zt zdMiDS+@f(fPw@cCh}rx(FI5qHEfG($zE%nB zQIa_hej8lA#tL)XT*do-H957XgEEUAI!)OOM}ifxblL3cm(?jGvB{QTcrK*CS*?X@ zL2*v+7r|)55u|- z?qjI9p9vmjY^ZD3U0lDzH>BC-;a5?FU3jghqn9P%SMk?lp0Bdikwx0H>>}7)<(S+$ z_gK#=$N4%>UVwNC9{7yHr!q5DWE10ZDI}8YFAdH5{EbgC`>|4Z|CV-=VC0`gAG@37 zTB#v)qP9-5(2E~r8*&5aS|F~270a5{pqn6c;Ec7lot zSL%QVg>s`v%xjy!9mCXq&+F#TIQ&H#RdENWZ84)kA-XX2?81joy1;51=Dn_DS%Y{)} zO(S))t62${_80^3lr&Ii=Obwz-X+Pq$<~9NSxlwy4Ht$Kj&F!>So@E*Zj5x^XaLJ)Xk)+^sHPwd>#2HZ&y5~5p+a$> zuRV#jbmhqu^xZ*UrsUyE$C31tgRbxl4e>q3H6HOG{+-yC*v&&j-=)NNr!t%D8{6}1 z2OZ3ED-LLQ(u+cQiLfA7UW~;)8DF$8_kWl3Rn88sh@OqBZxG-I>vtF0iGhybj>PtY z8ztY|PPSIEp^zBpi9$3t%?{kLjas(VS)(eadex1c)oR|1LPNr@N_YHN>YCXV3iUDi z_wh+LX?pBH`uweDD@r+-evL-H7uBhlT0J-5yB=+cj(KMA+OS7hhR$*5u;Poo+<@As zbC5$9TE>SD!bnPnX7x$8GryWMLonAtsy<^X(D>jpkFoZf5E9vwU|%fj@kx-JI19># z$S55a5*i_6qbVp-7J`18g`qi6z^?N?#E|a5-CO@J07O8$zc*DSaXUQhm|>6ZG#Iv0 z*)*ij4^w^WXVc%^EoYKZYLaQ0qJ0n!+zd#QHOxpW8&gmuy4E>t`Es5+qZDWG-pDO5 zw27}c*oXgVP8UY5Wl|!E{VX%-@^n1Vd%0kGc`yrBu!P;7hrPbMF8TR-DWfU^nuOB> z!O+ZWo`^s7U6$ zzZ1-tN&#$QWLC0u-k;=1msv7`#w+q)8%4bcuO96Y>{Tw!62}?i?&Ef*o{Kvr*Whq0 zzN>=!CV(@}ug}9dT3x5s^dG=f#UrzsU-S1RyGBZLLyL}2zsbQrZ4fsez5yEke4Nhx z_O6y7VH4hMI{TLBog?Q5am5^V!3dJQ&I<`-S%PkiA0Zy9$+~Pg!aZZTa^TqV8!9C@$B z%Gm;-<(B2q_d*e`5M6}8zBSI1W>q6T$}-mRz%3*9N8&u~uC>I7kkoZgGZwxwgA*mi zx(SAAQ_|t*VXEXH8WCyC3MDZTc5+2iH7ue5n;gTMk{_pyrnu>W)(w0w7#Fdh;=fRj ziBP!dyKJ6p;kJl$haUTp9+~j<@knXkS2iWry!+9ZUZ7~ro&73PZga~YHdHOt1x0RX z%$vDLHt_A>S&=I4TkeAT3jAYhYmKksk0)h>nY0%#(R(Itat&+6v#ImR(uf|Tt7hk< zS~4VNK9meN?_c~VpIg_z&X)?mS0z=CRuZ)X&eru)N*0n!ALpjtJPFFFZVH`lA&HTt zQ@<--y+>K0Qqh)RDnsJrtxz2afEK9J^qvV1^&w}@7Y}_`_UifAIjk-bZAN$y{?iIk zPz?PFXV z#FmBIYpG&*&74-tH|Ofh#y0C5%yo!Tl`2wOMBi;wE)mBXCSziUsF&-q=~W3R2PCEI zbz^Iq+w@A6vv0#M1SF-s4P!Z|zDzTn4zK9SeD=oeGi zu{W9-6-N@G=hjQ&Pfk?oregzZ9dV_-xOZCa749#SI6a6@ z5KrD8DU8_)yXc-`C2GEXwq2RoeFM`J<#eDf{C->lTACH`X1aT!dB-b4cr$XXfiY`ATW^QFx}O*LDVJmo`XcFVi~!>Ls}@Ay z?3XFh#u5XMf@_;Q&lWdUbVU;!p-SNlp#ql~nLQQ2p8dQ1eO$HRH^mLt&H+MGi_2$p zJxQi#j>O=R`U5UBIu3BMhP?a_;(||z65R(4{B4KO>&{x`dR|dVV}2L`yU)D|CCnEYyT?7@Wa*5^uFw&G1oyvegLb?lVms~$nDI|4_ z8O39c6+;!c!wv3wceam0`bLGi3-X(LSQTd3K^%abx33|$7e5b z$&IqV=dm>fMty$u<7o-j|1cyVzEL}(S2JRt4{A;aX#L2DKiXa9nnZ!9&-NCTvW2_9?9=IJcy)ICms>${Ut<(Q0{4Gc~?+i%}mq zdWN1IzU;j*oiTPw;G9lY70T zHMc$9gKy;#3@`X1o*NDJ_Ir!7)4ZNsWoX}2rX`I`0qsI#bg8@ZQ!}qa{cv|^f$Ti0 zKK~pNi;!a}OTkLe78rNmf{BazZeIuNg2&!;-3G@o`WZu}ttMFFbXmdy*3QrmZmyKH zsj#m@c0Iplxz<^iJJjV_c3b~DoY35id!)X&?j%b;N3VRS+C54M1yUWl5pSW)TaS{| zOP6x4RPVW1DgBhvm4$@vHSGNrBrk2HD5t#$;-v7Mj@vx!rLV3yee+KE|=mr)W<{k!`D2UwMDyzF$pG89#YTlbiXi7PXjIQKX+kjL3qnBSEY zm9)}wP|65HZ<%EK7Zq7JlwIYG+AEwN|^KR~1Jq9xL(If(eHV?x^Nt>9F! zC2d2rPxRq0eO}StcMAHe>4d;Hb;;&`d#`j4C(X@sQ5%(~C`OHrlI2{>^E$Pb?e+^0 zc2{Oc;d>m5IusP^>Da@7P>z%rA{UvMG8TLeR&tmy?22_E@&3|uu2#L+H)0l2dVK@? zSqvAIOQrcpqAAftZStn^Rn^aA#vm8LV+#A|!5@?EyPw3KtetoJHcQui(RxM84TX^Bo6b zP_&|{i)nCV^nY>OT@Q1*yapN_7@U!2B;LNG-4GQ>nybV6%z)(Q#eFyJe0(`#!p6n?EP$PVS~*>*=NNeV^PNgJ=g@tTw;TA_j(T_*0vrzaEHP5 z0g0f6PVhP<6d)A-!97`O&IWHflC3ik>D?oaSG>T3)n02WY5UY!E%{7i#2i5YzDXBR z0*M30!g()k~aV;aczao2!ba!bGQ~+v?@3_U$jMk4XeK zVoj!o7{`)HQS(l)>mF~PB+2(lRx_{MT6gUd$G+B2-2%n9koH%?a5jPDTZH+&;fg-^O0S~#6Cq#qw) zMOiBYgul9amh;Zip}y9r8?KaRlVKB1Nk8EOb*OM_a6T|-fL)f5-`DMq2t%{Nb_8uIz zN|vG02GG#Yu=wo;J^VbJmI|tY4niM11*3poq?od#(~ds2IPRi}Ano`kto@VHVf)^e zIopVWAp`zepO=g4esgw#nphjsHb#odI4N#%g0k)J3{Z1yFEB{*>I%_>Q(Y!C;FUp) ze&no`d5t0@TYkx5_xMovJmav2E5cQZ=@hW{<}^o?jH^f3e+LSRQmH zd5tMM@29d5T5S8x_@h61CzFRgp#{a~0-h1Mgw^r^1{CWq-0r1z zAHvAT^)RsJS=j0iFG(6z zj(-yFcAAH8I81x?MJ;UJJ@boA26FGP{}}RT#qPV2^S*nW)h6r;JY5>9si6oUbrr5a zu#kdc@vL4zc(!mmj>QSqU|CX9B2jC8`P^){@S+?2Q#UJbD*?wu-1vf3)rW3Fg0B*6XedGM-z=ZK((GPs~-6JxY-V&&E_dd z)4r4wn7IJov~(NpS!9`4p|x|BaSI1$XY;#XL-*}R$CAhphKHl>R`+7qMx8qBSV0W&s{h3W>)tqxC`(mo^^ITecG=B| z^#K*fKHs|yL~Vg1W3v8lMB-F~!R`n$FU?-S(p>Yzo_1aL8VSg+z|X{E^T~V}5an?? zQ8kw6VmLu%RSVQ%Mt4fmQGTs}7tL}51+zx1M7ZX9f%%DuwH<}=9+;$%od(#_@W;_O~pK@EN!*FUnix(;%H0k>9QQm5wdpE zY*au_`9gg|FpEx&T#bL?2YsmAl9bsT-6n0Axt1P_9=g+2$rg)4OimG~%>7g(`|5%p zmAJ1%qD567s=R7{Qv$)*982IxR-89(dj6rurMSVq*DzE?ySk?;(w76%g5NE{2>&qP z$g}75actPxlYmlcMfoM(r~V!95vcOD+P9+eT+AmkO^S@C>Sl8O>P3;tZiK<@6B}$H zb(tmL-2?1aLDG1ePcwx6q?8#1(3>#f7`8Px$%10`LhBhKLVf|KML4F!aGp%)k4;BydbsJ-%EnCo! zql|e@;0y1eN=D`I6SpH?SK5T2!K9F%k%%4L3v}LOVRD@4tf1Q}?0pxFn}Kvb zeKbWfn9iAN(coY)8SCLNqp>)1V#4}cbF&;wv%c>fXqX|>ZzN&xMBN62Jv@N8#=T># zVdVqnjC72NS8s%ZO_|-pB_~7Hyb>}IBdzwl}ay{zbBBNv5<*`7JW*CML~Vb zRU48Lu##i9ka+&%d-IF8E#WTcZl8^)uTQg^)e{=2_&0dgPMJ!iEDA9R(l&h1B$&*d zxcl8RaS!bT7#gEGI@GhX0s>MKoSm^0kx;g~vYF#p&Xd!8`wx4Y5yD?mqnl8o+j3nu zB{9Y$oSh~d%^=4Nms(deQJ+dL)O~rw?4Ywz)yw*E%LIg9CN2JzvZ~xQ&&1Nc)y1)d z%12W+s}~8@Uspm%@f#dE!xoHRMqjf0RU>w&uSbk~eA-yW$f54l?8|KD!oXTX>x|N? z!$K}9r;{M>ik6L5LKcdA=uz*}m!7yTqPc7ubS_JGFD_!ukZS~Pn=jKp6$V%hDt>R$ zVM*`nA$5eZkbiCY*x)NTq+%#@cPL^WC#wKvgTgNVe4IGc((MTp-OEcNhhE3ntx?i9 zl;dlJ(-n6oE=-4}d2!om4jlZ(_pZA3k*wi+w}T$c0C`LBytGq_bjM00vWdo~mh_q} zSB4I)(9oDneV?ACcS*?INHNqx(S4kp2c`h0_E166O7m>K@|HzvA5Oje2TFNul$fP; z7&lK^?J|W{@6aAgWEovYx~XNk*w~M<^{b~E`A95a4_x7`y8C0)+x1v z`zq&d+)3Rr%7-|M{Cz_ibq}wEZy?HaRF89}w|kJPy`v24sxB|H(Ug=a^{t@3Yc2MR zxQAj9xSGe3LaeeO`J&TXU7M7Qsx9eC+Q-0IGdZoiQqS~%2&f@K!evspGy{vPt8vF(msxO7nP5c-v_X)_@^q7B%VKou(6_r~dY~w0= zf$awR)#62YmQXopAOT^G!t^kEL_nRD@VMl|rtkPC3#zkf&uivs#85Ne(;K=%x_v0K z58lVk1nfFzRdCfTq2VTb$^yYc+lQ|rD<|}jzr>h%zyBtX)hvwiHblFzR)KBsK;R+> z4;>l!@i~Oc*8S-6mJim<{=Jw2ZF)6)%|e_>QG59N^SmZ>+K1zYpP%7uzBe4@eJ;o0 zu&ON(ZE_`R*q0i|+tT&&UludIqR-t9CPpy?uP@=^D^4?9xfj{QQkNfn_z{vGNr1J$ z-Kv3@bcFN$cELNke}zKPL2n$(N!SR62D!bl?(?FPMC$+r(ZSp5n+}^!o0XXgMx&ID zQQVp|yU&~nlmo>3`qij&?t?pRp|EiD#Qe(7vs$4+_dvc}bz7zsk^)of?k?#;T}`|S zwbo${f1608D4TXQ2 zsj7Eu4?P!T#a2I&^z$h)q)_e6(vn?~!xL5qRnf980+7fBC@<(1XN~eu(q%k?P)Y^d znfxyrUSP6g@Lo+qt;l?`g5Mb)ZO|nQ>iFu?XZ%3ENxURU_J>Z*8Yy zvi7v6k~d8_VY_irIQu$MoUW2_P)}c-92Uy1;Zz^{rQP+%rx zYvb3K2sz4IRPgU_$hQ!pXj-avFXugK>w+)Pi!DBEQ~I2KJDH7A_e~`t!j2+)fF{=S zCL>QRvLg@_SfOD@!jWd)wbCMoGG-$Q3exl99P2v~3^wR(LAg|v8rdqFm;UgYx8%S| zrYq;DjovjeBBmjhh|wh`cqN=c6_jNqsZ6Ct`ErjIzdh29JnS+Nt7b-kV4>^CKh3y0 z$^qEaek*099*5Z`40hJ_ot6T~yz6<&hAA)VZkLs2m7r*+vKF@n7^J2 zK2n=p%Z|mIxH!kY;>(^qHk;w~;Tzlmq`egdT>#GYaP|h?w8^U@ZkRKtu>i!%iPUeU_*O`LUYEWx!lG4Zq^doNfNu$cKE;&X)QjtX~V_ehi1LWV$ zb?xBkOOB?_d5wkV-`mu?ntC+}R4QsQj{0(m8+1zkEE=7!d+X2)zlBXc!sYY{I;7%U z?VXWEYQK!FJ}>eJ`u4(DaEG~zYHu9PRI#5yiJCnZuT=_b&tnbmgjUSf=w=W0iDK1{ zRQMIEXv5;z9)X165;(z$+SG{13ShcrBoTGfCGQ#>N*v8MA@rf=S|dUdr^A5HtgghA zi9=~(<;`GO8Ji2FR4)T<=1R?Syjg7t$4*5kPvqvcErnebTp3WPkTenJG9e8#Ed{ck zBaGM-CW6M=_#y?xcC z_Hm7(sZNjx=)RE>?!4&ZP$B+G)>h!#xPV!MRdO_Zgu1gpdm{w?DSXZCx=HZ;b_=)5 zS0l*Kim%%a@ijxhQh(lZwDXsrnbJo=hihX!&}?Bse8$Jtu{2W1_$T^jeB!FY$;?-M zn_$@Ncpu^?=DgReOWD?hgP-L3Utt91(4z*y#)3!syq%h4ZNFa`$haMDi#+?->P``} zAPk)O?u4*NnILzREF|CRo)vz61+gN9m(Vg4f`&6@l(W!B`(gHB_W$vp1fk~qhh|iA zs%PxyV=0uJbgMokCc+CWPt5p>WL?X}>mqD(NTu2VO0k42y^IM``}|13-w5)O00?wc zYY3_Xl8SsfF}R=c-}&{bdk#sn+suw=*rB7RtR;O8B}tp(!|(4JY&+O%N$VG7Qaiscm|mSctmE&L_f&xv!7}x)b@Gs; z#|_lkn;5^OmoGIq_&B^v|5T6s;Xv2J|Q4#*|m~Cghbnk)04)&p{)G!BN33 z!D;e<8CX9z%=J-bAYP^yWgV(kdu1m;1qwM_=M_;jQ|Z z$7%*yI1V`y?2QwS3@ESOyuWU@Bc1yKW#sz!cvFPE1)tu)Bv$fZ}g`2c5$>Lq8D@lh79Vj3;x_ofOTy2Y+Os-|^ z^=@Pz{Ri=++u3EtYSWC879=yS+!!_uy`N_FFoB>Ql#}c;V0t_$sx;vPR<2hj6cH!q zpfFBx?&|n{Sd_A0Gx?JDU08o62QNhF8NlaH+ahlkSW>%rL=}AK{lt zHwD=lqm6dOBykY)+mj4zaqlHZM8~%C=GIM&;PR16-QJ29r;I9TQkG007_n>~cSux8 zS{~<(u9WXEG#1O1t;vl%(_1i2#&Fj;(i#gqEZ;^hbT1bbn>N0j@=o#(BYs3eH# z{<`n8B~V-&s9kV3oBy--kO}GQ)RjffF!8fD6cojKf&K88Myw{_cjOVP*`8@>1&eN& zb$drvsEIPNT1^%ou_dtZipv;gVC?)QXj`ABC+J64yClA(v$TET8W}0lzg;!!u1(WT z^@ejiB9@0$6sNdbsVeQd+rdN=v#nh5sGOsj!qC*31@`;U=*1QwW-V8}3ACmi9_&;& z8a7O`jqog)ey!M)$(+i)7#*Q<9bD`-_XU*G@ouQd3LA~Zo;NAGn6_q=0H$anuqgy{ za3(RFfAW5&6M5UO-0b3SCXasuJCAMeDaR!FXwLP&FMJ|v$YmdP4qktADJj=r*7VwT z9ztYs8WD$Wm>@?Ztrp2)M4+<1#1|QSb+PD?cIy2MEnb`)i9CX~eacDIO7E@cQFDtB zlHTq!<&RQ0>b3=gmI}s9fGsu%@?Syno2~N- zf)Rm3HLC+hfL%_EVRdW={Ut{AdV=Du*?V~zAn=`clN`+V(Pml^6(j-X`6Yu`&##lb z2zc|Hg;K+7vSuS<>gSo|A|$G#PMnHaichJ;InU;zrCy2_(tQQou^3*n*I4bb0Nt`s zok;KU+UF0jA8#Y97-9qn*9!yG1fwQ6Viuk)(o4MrV^Ka7C};MG=|%^>(Z+~I<4-5| z0)M$(P_~N6?P(v!Tf2F~Tf|BzCFwf3;l!Qk+;FRjmx19%!!LhZkgl1XBb`=w8$9VX-Bc!kI^H-k2Qn|RX+^HS94f=nQW+{oVFWFOP zz3JQ^dGa(?PH-stAzSWGdPu3iZZGb;#H=2)dUArXnlq_%UJ%-Gh)k;#-YoYGPOl-0 z!icHmqiL~OfZ}(-u_$Z+*Dba(;>Gh5$?g_4}0gGu?` z8kCPjrL6b&B2UDH^109+-*7v={d^-e+VHkFh%NwC%<{ z1omu<&4*2FBCA~)K_vh4oV7i zC!=9cSHLJo|1=qsk={dYj>c{Tc{B8S+_#eB;i0|Kc&CTY5acCU;G{iOb2Hi&* zq~R$E9$qaa>YY14!%pFU>Cp%y440U!gH2vbJ-Ks9d^a-Q!EGv9z4h1#nH zd|jHoC{B8lIIgC`DGI`E&;Ud6-57a(>t1Y`^TPI=V3)k8H;-@!$Fu3BGvcFMqjdF( z7{u0~8U!4gE60R=;EyjCv+G~gSfZfGj6nFTC4GZE-d(8 z9rDM@-XWKH(*H0ZR$vq9@O}~)i3!%Wt=D`WAc=$Nm15!TxB(R4=6bPy*0A2bno-qV z_;J^5e|bLC{zl>x!ojQZ&q>JT%+CGff&6g6uN}#!IL=f4IOSd&CJ8MmS$6bGYS(#X zNUL!0(L6O{!4Q1Q#`LA)l}GWaJ9UBl$rG?e_yGMDbCs0?rnAfbi~fEQ;)D)gVsj8q z=nf&&-3+UttM4Rj23f;-75`BXGB$N%%h)t2MM9lIFL;`TyHCF6#mBAUMD7UP(gAEv z0_k?Yqf#5y>xXg|xBN47qQv3`v=pCiIEHl6@zBI^xy5uX=*{~JIeJ8~@L0C~clCY- z9Es{CUAx&zG(=j|LOXOTx;)1{D9{3{BPP+2((O}@j!mzOyj!jG?cnG1i@#K3Qt;3j zw9hWD5K53e$SRD1-E`CUGFwnYDoCU+cCl6#zt4y~dGl1gSmWz_tY&KKkLKPj{K(rT zTQgboTA__L^P8EnHQ(nTpCBu&eaEyq*m{I4xnpjdeGjpbk(XsFA=%b|_b~&FsQ&jK z$?qQpW=>YNm%gcsXb3mH;i-hCDU^R9_9@6sD{{My*S7K|GEC*}>-V$yf*%BzUKq=d zk}}wRPP&EP+gSlZ0cMxbo=EE%l$)$`$;ARF72)3|=!&EksucWP2eu@xDxlr(35r&a zIsJEcCDH{w>%#S|GdPdFZ-B~Bld66m$x0T!EfwdzaqCK}=_FAdX%em~ehtMHq;uF& zd~gSOv@dN&6O`y{Cgjt@C?CrW<1v;j-B#%W?`1>C9irCeDO|03G8}ZE^Kn74sFnfP zmxSw~(V=>*$Z7-REf!rqyF8gt7w}xG5^>8JsO8KEAGkSiaHhy85YoiLN-@#jlu8SR zD3|f01)C8dj8hB($%=!nV5#+hS+@mhEqhh0KE9@uNdYF=FCVctl!>>JsrMa(3r-V; zHJ*u*F@~1!_OL6_eO`P4Z;>&u7R84!7x~x5e$`1^o~u9j&Upu4q({2lKOOlgX`}+} zrkJEPpCSg&DHU3(wWd*f@kmC~|D_ZY>8B-Dy&5X`u3SajYNi)V+zW86$vvYXObeL5r zPvHobbh|wXOo^O{zSwzV&%T!%MA>gmzjW48PlbJP!WtS^d)^C8MYs;0IUfyii5|=P z`DG0yVUm4A$z67z7{iqz++;%pHqhXs^F5|seuDbzQ=vsM3J*X{D+$!wy`66c0q%45 zRi6{Lkq7VZIyp4sic+9HThZ$;C^&bBYbeLJT6{#?Rf|cFd<{MemBN@G$nsE%ix~@y z`Q~L;TetLVv8;^E#_M7BvUq2vOn=^0J-w97P69u{G23?-z5OSYwMUJ@vOiafq@zP) zQv&)7xhv{&&F6HrRdJTPM-ji)%WnBm*eLIpEP8>8=}*-`^g_3~O>uTd!uEzsq4Nu) zYE(T?pD0(yf;4K$-NT@Qd33|Q@#JPObk>VhAcM_C9U#=w!O{*wNkTj1J>6DrXKGP*uydc9*Y|*cO zwenzE9oc&9q4H*C2va~vN)@@3j<0Z9Q%nHfw)ixyGXiRMr%VHu&g~a#R(mLZXB_X^}o_W-UbS zR#l3lL47{0#Hh^6wE*WVTvb@Mj?M!P=N1{HOMQ6tEGS2CHX9C`8w{VRCK^cLI$U~c zp!xH-hPIF*Z#L2bw(a;eaSslh2tpM+Qna_mi8(cHaMBA()FRj783MvYTy;6B77s?D zEegE2b6M~FJ>!C#JiShx=Q_DJcNU*QfTO8X%pq(GKRP01}g6(CbL;S>&PF{ffMr8ae8jcQOVIJtd^KQbjU9TIVAw z9j4&;lS3P)Sz7QxDJ){z?S*&|thp>5%P&^DC|lS(uWWzro+H@nkQkHcVa)}0=~^yO zXBNk$wF1p5K5TTpzR0bI@?b=&aF40X;>yhVLa-q#gNo*z40FSIt>;1X?f+xz9F{}@ z5G2^PZQHhO+qP}nwr$(Cjn}qq?ObMK5Bm!hQBkMNN?f=lsMIseo2_nF<44^Jqz)|a ze17K*%U7G<`09KU>X9#_g@*=tDfzhSulj4B5yDg6fU?g4BB2L1?u{?T2qARR5R&ID zSSapxzt6D@p1+S899ovjzO_lMw$SOBEOqfBSxti64!E-7q;(k{>gl&-d^-~;*45Yi zFjLjT4ffA>yS$Rmxbad=dgs#PqlwdeUrPlPX)pp>xxu3{%gbbs0UjweEzgltz9UNL zb-F6zX_?BbX?Y~Oz9bX$0qmG+cz4>la{?>3ae9|&5S|VM zj?Al2{8XtIX4aYWp~+ws`1oXtbzJ8!d~M-;q;q_pp_onzoJrXY=YnB#^Xq+jj8%$! zg@hKc^%hSCfZkJW3I0Hs&!rby9P0a6#)<97_nR%S z!k3e>le{MYJTqH|{)wXD$t?Tgl?wE86bzRA3kkosfU!H_T>ddEax}0!Rw*jF=GOUO zV5$mjqk>EvfBeDU5@} z*A`PuA&x6)=VC4pX=lR(fuSx*m|%K^My*vW+qWh}p#JibVdI+eAOx`@Nr3k;uEXs5 zjizzOoYOiMS(vc&nSZdk_9C9|ZNS4sf;Za!RC)>7@^8u%Kn5d^cWE!PDpo?Rp&G%p=(NQo{-?>6V?X!Jt+>WbD|jiu&FyTPpN~#-c?z3 z^36!L&w`}3-BMGACHIgu8OYJxi_S}88ug61Yrr>0Z84_cq zVy`xps<76QzZNX{B}*Cm805ZK4T|R@Ut8 zhj?jUoVl2EAEc*lkj1@{J|l;`(}3ipZQ{U8!O_+ViHyJ9olb z(2k%*9W4*W$g^Gub8<<|T6yU|+!LmuesbcWfl@nS_A!CH8%bgZLIzvPce+ebw{&bG zl)1l6`kLd6b@id$wEZ5YWwyebdeSP#p2zdsBkgPy&&MDcyR(KY94lnd&?Cwb=uaAL zW(6RpI3mgqd2TjXd|zmvQ>>e+yNodDfOq0#If1Hd5W|YcBvZS~QHgXi$O2S`3f&Oc z7ls3D=8?GdX*Wj7$>$mAx9s~VM)8?D!QXre-ieux@stetBCH7gb>h@0PPVtFssjr2c>ChpXoaAb%Px(1)v4V2f z1VPfP@9LUAeO7v)Pe2Dc#d!H_W>CcsA4)@tu}lG95FDa}GfCH*zY)(HiXN0Ea%D%^ z{g5D-L6%}eet5;qMqyy^EoJRky`&wx;maLsKUo=;#Z}P@bTzpazSF*qoMe3lfZh~E4z}n{>L0dc(TGE{)g5D(ik*&qZ184;2?0WU*k8P4 zn^$Evabyv~SuB5oP6q#otHRg<0fa-D7OzbgE@yMZaY0twq^u}!xX(|iToJ&xtfFVrxC-TPg1H9qDGWIyUO}x6LEZySIoJxX zm5+sopm-lw8{cyKg4^1I?SzI7>Hw0%ydeY`4X(p_X2b4g3>3C1X2$A17wY{I#EIC@ z!yqY8jF+}VtowzGb4#`2G2G#UrBJ69>e2xV`S1*j&TqJ>MI^@XOLJUL^J>i~=UFWF zlG{9gnT}eI_~uZTjge0{a~qYuw+kR zz1nie{V$4wn=8gj@Qvi7#l`IeGHxtcRNv_#<0bzj+^AbWLbF@_@yMr0p8at4o}kb#Tw83#=19UCwTkd9E^i5UL^De^hPj5+Hmz$TlR> z4Jp*a*{yEV3Z@K6HjiN7H}C7vf~`@beO6mum%Y%?4Sj;3f@@|^9)C0LX0p65-su-q z;iQJ-ZGC{1b@(Jxg7$|(?r~P>$lF5Mal*CikF=@$rqM$kMa~sa=8~v000#$?bLnTd zY-hIHBG?nCj)BWM>#$Cp>q7g^utS@F<{m(ttDlRgDQ^oi!2MLSjJivBz}Z94E|`O- zZ(HXT@h0r4Hfxbc4FZG)u~1ZRj;OBu@F55w;m>QK$!p)(xVNvCp)+}LWlD@sP{K7D z6r=O*0)p&H>ZMxbqH%3U@yw{ST9SbEJSjgL`cth3NSS%QV!U5P0cLcvC?771K-)!I z6awrf5ugNPLfrclu@cIiD}5u?%5+})WBDm7*zM#}+wmVLuwl(tmS@>@JPr6h?I0L% zeC5S+s5|fYI?3U7Y^wNa_swXt#bGFY3dk>QFHd>c{m-|-;~#ooT_&p&cy}M_q@yXn ztrA6ks;B0H#L>M#LXv?vC)*`0P^&Nm>am)io#KE%aMRX2c(&-p$p zLme@yl;GI_Eva}$J+5yV$%whzHUNooZ!~5xUX^&>+L%!B2=Eac!fQ%UmIB4oCGF<+ z$)KWWhqVsj7to(ZmrP&Oa-M{lTba75k8-diLaPQGT0z@cr$1hCC>NC>$5wS~*yc-A z?E@$#bRXr1Q+J6C04_85>6v*~KQ?PxzB{Kal%~pqzl9CXk8{jIZWX^uT=eAeYFgGx zU+{YlnrAE6E|<3mr&--cV}u5&{n*>s*BnV}k?q-rIQvwT3)7m8wKlePLgi-YQ^!hK z^vbd=O}z6V{M?DUqT;rmJrZw`hL=hGl~)`OeBcVOSoPmpW$g)c?D+K z)jH&djlh(Zy96Auqv4q#a5W<#l7`E5f%j{0Y3+yw4BP-*<_NYxmbe7@GuIId>g2261`-jB!w065fwN5vowf4XYnNN!(b9Oj#i@W6Tc5L{mckjrw=Q-pQwCsR?f%1 z?O+MN12L6e0EHTVnQvS+UYM$#FlsGfgSh_E^&7X+n=xTsUK3Ktdg0K^^&|3%sWoiK zqpjq*F0v(e`~$3&|A7$6nn3RnrrMKZlW>`L9Ucz02e4TRZi@d~Ck6h#-#K zLB^1TS{gIBbfJ-D0+b|mx+<5FkKl(P(`{0RK9+~?nej90lKo*6vNZ$Gm!>q~IwF!_ z5q8H=kc<9=>a#+5sr>^b*vQ#PrATIK$cI?GhDfL&b1jeIQOJyTw^c&5S}TamP=_1ml&E5hc9-u8ebC%+n+v?DmKY0O zGQ|`%RzXYY#!zPlsYJf83rg;IP=nt+B?UH~yMyugMvp}ey_!jU--F+_zx}a zUBns#$gC@)LXSCyhFrxbe-llN;)9+P znNsJwaG-eePJ}*E$n|sE3RTaBgu3^KuL!L+iu>c~`>!44QB!;Nd%1_} zEv2#&&X2xTJj_z&K@{44Sbr=&J>?D(D(A0p9aU_(UO@P0jp&P@-IK*npw~^_e(d22 z<;yLO5qpNDgHY3Xn?F6V>-4-iZBxVn!yTGRGN{bJIECI0@C-V~MhK0Za$Puz94mcF zwT~F^0JWb(@%DDjrt4~i0*x?@=^$Tql`{Gyf27dJetDNf)jGepYS^ssbVQP2EP+Kg7yyH~%dbs##kYh$1Wiprs_7 z$xS**pz6Ut!g0X_7HS4qR^TSg$s06y{hTsx{ggFi%b}Tn-Zhu&G2XuXA5RIM#`@6T z5H&%VB3cTtcQ^&tE-92x<*9ZahD9IQJQ#db5zqK1ay!-GL^bWH(@NhA|9)9VRCPV# zhzb>J9CUopdMY;?^s65*PpBpp3`;tJ52Hf9bfn=bLhI_BR)j&&{*N$`M>q}HpOtQc zbuUCLbIB#loG=!+(>n_6mAl3J5|3pXI{n887xwVE%NyaZ8F81RRteJ8(1lB!aLP;m zm>wmT8!=fuED79Fiz18p4qZN>oRNOPtdK;b-sL6};9Bp$4-(Gey6*?F3p3i1Op*!> zIO<8-G{SOV?=kms;0CbM2rIlk3lUkcOAdJ`0Y+=t;3t2MVxCq_tyHNBI+MqI)}k6$ z+Th~YO5FV>pi>bzZHU!4&`;e5lonav9>F$g_%Qq58ao<1^|*Gk>9ksHI^j(@H~ciM z^15SJpbP4#606Ws6DP?C4`U-T9jWyiBximKiwRt(fjq-HV)_aKNo|U>x3xkl>+L&g zm?vetnbuRxU&qo#oD6}PGi&uUvcyFZ@qcgi13ZCeq7@BI?qi(ZY|BTcCfVAB!+F^o zM!@|EwE%hCTb@I}DdxPnl_bNlw+p0@(K2;U!R7sKdkv4X6sTh*L!=m`%c)FiEgAh8 z44X9FH~X3kWTi!$ZAxOBd~D?e5o}0z^*1ampMc&PcZIRO*LlMC1NcTTr2ryD7wOv; zIp_MuNM)7s(7{tivoYOUt+ztZw#fGXQgF^O$Ci45s9@M4VAs0vmY*}Om|rq0LBaR> zilDhK;dxW@=z4zE&x+|fkzDI|DeesN1n+JnH{j|RK`7KOf#~(MbT|H~xuh+e4MOF% zh~TYXBNRNtwqU*ragECXQc2m0G-6NLjPbT(EVEdQ9EEL)}|J^Km&yaeH>6v^l#`oY*{v2mcVt zj}6Hcm1bU6j)*&*Q{Bznhy0q{ic6NC+xy9a#F@^?OFqBQp719Y60hDK)uGa|Ds=>6 z1e4a#?<5514esb`^eupRCh30W+D^@o`B^B)JLom_kf@fz@1JLee%Dl=ZNcz&2-U>w zt;*e=shKji5%A<4pxzO%QN4jhMR!i-q9R~{c5!foH5~BY*zGsR;!#iV-zFhcL|LX~ zH%zw#kOM)Ild;|AhKIwxam&tx@JJq+iK72#d@C7w=_Z$aLvw`0?iahI8TWsrFs>bK z<6zj}-T+}hp1DR&dSc>p zZYAfDCO8^Tb3VS)C!&r}l@|Rf3m4HH8pTw&?#2ZqKf(b{05r=YnyDo}UzCn&(WPjCo~R9j~@ zcL^wxMKz!i=TD=Ui5q}pPrv8>DB|4hQm!EroAFz^l}_ofNkYN0T*CRYEnq`E=8LA2 zp19qvV8dN=Fu(lC^u8~_`8L51*a59b6yiuyh+Y~U3*Fd8hF8?|dtFssoxr6e#p_z9 zrT6MsXSD+_BMIF&UD)B~K~2^?V>2yCJt4$r#B-F84)izz8S#|lD)YdlBgUEp;T#ZH zq+=6cLPk3FA1p7a2i1j^s|em}q_sG^5H{Vtxo4Xcfbx>?XRZ#`*%0*v7%Y;ZcFlw zU;XcTsM-4Z?5a|NC9Cvsl{VgES2EwPCJ5^3UdyaW;vOg&7K^k+R!dG=gUgVN4nea` zQvq}ai(lVRLW12V=~5^u=+F3>GrceXLmhLNBt0@ar~TbNB;6eFu4r?!pBg3sRr0Ch zJ#jczzHfmM4wnVg!BN>4ZV3!=#>!OL3MhN*HKyHjOc$&9P;Oo8P`ultB+W~iOx>H? zmxnz(n6~22L!!%mSjb<0Yfl+ss~yX3^8w*xk_X~v3qno(k~ZVAnGZ^F-luuG)%ro_ z%0~&HwJ52DY2ozI=4kim(4S=~DA2W^f@{GZNxE}FINAgLbyA1i(fi2~`Y0)p&D+5k zLe5FejBbVAJD~|gcT-P>MJ~jB6KPH5p53FTao|9ZF{+v&5)HxAW3CB27en*2Sx4Fx zTf9Hk?>%gtD?bmw!tzgOzQC!DI+3h@&Ftdfhly?hU&x5+Ud=l##%n*B{Usx7+t}bD zOx#*Un7k_M&1kq@^M=N8_egV&z@FNUUPRzIfHj5r_-Vun?~9w;1-i%{Y~+{XpE(Tz z4e_=H0C9WdxoU-YCNIq%lTWQ5B{314@EC<_($Ps`n-sa~nPtJ+BBi9sanSiNe&k!r z6kCKu9dld(H}SDRV6=@rySq-W=79mDRS=Xh)Ld+dt8IPmBIOSJjPxn2szq}Q5UE(9 zdz9+bB5;T1w{zE%VDzphuG0(kL!lI+jk`xbn|3y1bzmn#jGnBw3z#Yi zxtV4NoQ5WUSZUS_H%4)}*cT(vRn_#01Tl;lVl%t>QUYf*ctKXmcAY{_V|B+RE*P8^ zKWA2~^BFUBa!Q~kC2XMhCuX?)^k|2+E{H+}ModV-0)H-AXDok?T? zK;*b%*^H`5Vj0Hj)T8siS8G`l*?b#Q&wp+UEPYHqjz6ji`WVQJ7DN$odh;0<4o9oD z%Va}kS+C@f>!pvciW2u@xsFG<&r!FqMft=_80QZcG%mFDg^O@!|x&xXFrfd!YcUcVeaP$~n!@u)Ft-ES2q+ zOox1%&#gOgr&caE6y8xb}(?8a-KCT-|`-2_jq{6+306v&k;`b=9nBMnYnQ zH#>$&uTnwHsCr7m56&oWTiI({c}%?;Vj>29eB_<}X$s`$3uHLrH2l>J)+hXj)I}H` zXfb^bj9_{viTPB8?qUm3aV8XT5n~`{H7W@*Xp+B24yp?V)fLA>T|;GB&86OVgo-~G zHF@83OapRsnHs!jjZSE;R^vJiWgWl5qag{W0sc~CPzrCYQXnQ@4{P&VNJ-{>joUIx zEsRdx6+h&aVb;(Olss58Vee#t@=+JpL#7zT^6ueR`!lpnRohtpMZsoL$D> zjjI%F9CdR*ydErON08?yjI{u?1h=c@pJ1VgV8V2>ASTaNv) zvM8Fx#gQGX`e!bi`&G+E>YS!6sdcV^-yQEn{ToBG9?9QD!N#s)#Hc%ppSu(`E7KLI zzV1w7^(|{WsG+QdB}F9LxkowXeSLUwHt{#tol&cHeN8iCd`oVdmj{|XqM+Vo%k~J1 zMm&vX|lZwFY^lW5pMmkkWx{Na_s@S6vfuQ!qu#n^$gLi`QUzYt##06PPk zwBZR!!7wMJI`zgY(jD5v_?DVNtLuJL>v_d(^M<}Z35?Fo8WvbZVjw<6$#p%x)B)rW z${@ruUEMSs$*kAsyzBGgjckT8U|mdF)w2@;%SU4ekir{ZDys3AaoowiAC9w9rCsw-|3hO0a@f0k^ot04~F`fS}j4(a%uy;jcWTzgiN}=;apChFo=I={? z4qcwQuv!z^PYm}?eRaW1aqc*8ONSL~2;}8#^~>poO!8u8EcY=BwPeyJOm^m8v!a{W zu}3Cn7Vh3%(Gy2V=&89F2bXHoc(teChsIzDJ*mNEEIWmSl_P3L`d`ug@%ahE%qC&Y zef@f+6Dd4hDIAq*c7E^vqJDF~g3Ipo)^$LsssyCJ=|oczYiv1y=_W!77t7n~BP4uF zmC%jDbZ_g}hd6=mIXymEnqj5Z*p%AYLBkud`bz?;_B~g~{XZce^>%cSrRQH^zy7mh z1wL2>*6vY(+g_PLd+2qAtW29kLdyX*HT30iyo0=3oyohzEr~wNhvR?sakP|_!5gPX z$#}_k3s)Fusl&bMw+Apz{Zhl&!n$A}%5A|ysV)W6jv#DFYT73-wAIKa6l}_EB63#* zl5Lrg(n57-?p?Fy<D#tERqYfd&;&*ywX#$0eaWL2-VOY#CECwmFkJ(ieLwQ&DU+7hX$`oA$)-@GHvyw zOI3+zoo7v6#=J;2Yy5L7afoneIB54!Ky2O@Q+&AGgikzb?{=gXOa$tOHqMZQ$kRO( z@%s&XE!FmM!T4Iko{tM;xHT(kJWIR|Rjcm4BUVE&ut1)0ORBYFhzdtbIY%h_J{;+| zswY!>l~75#Y6l;xa$x5JjD2Y$%aI@T{OJ>*dik?EW3Q8pogjAPd54Nq$arkHT zxRpQWh5?05s=4OGh+p;nr3!d}taKgdKLd%MhPPYWrsIBLhR*$c{DED%<+)yM-H8Qd zYu{eSByOW z?as$a!P|PO;alcK+*ybC(^v!tya#^hixJ}P_*8fn;Uu>&V_jI%5i~bA>8+79?Oc^h z!;#aEG~SM2#n(wH?;cq1JmR9s*BN5Yto&>QWr;G{Wr^;8A&Y{M9lCeNY~I(i!XzwC zy|%Y(k+GAbj9Kw8VX0r@yacRJ<}Dj1Jl8tXTUwC(ocw~V2&q@HIMajE_Xhcu2$dhYuP8rZ-PT3G6YXaFw3eXgj#enYcVkjZsAt_>SxJWOC;=d zP&}g|E`>cKi0LV!7{+;^M1}d=PjTY?4iCo=h;mGQNqAk>vodh7xRin-28xh*mfyox zfHn^!CCKA3z*Kl8S(zsVuDF&#+_q-kQi;i9jD&S(Q~}T2|0G>Drkgtu9onK9)&| z22v*L$?w(KZ3)efAFG?xfTYXI0V=*AeHLBccXazl>_fV=p(K3`QQ~=q@Ot+?Ls4`$(zk-(*0j%h_&X` zOZBD1WwQOz6Sy@em}2&mV*z$`mu){@8`X{+<>bq zb~)c|d%V|g`DDWYw^Dc5vEn&s4oJba|Byu_Vc)FH^hOb!muTQ)L`Q({gSXXv>5#OQ z6gqaydHxTY6y>(SMwa_{W9hD&Snu~c{=7MSUh5UAqzOTdB%lidGZ{HK7$3KO6#FV< z54JLXLuv&BE|bye-=BO?#dr1?*&``&}zd8$KE`K5QnTlA42JUDE>OU))KvC*QDW%Ssa zvL4Dj0gdZVR$k@CJPU7>b3b)#0e59=RO;EHY;hLg;&wpx!Z_cd2fU=hyv}*BP!%m2 zk2j|*WghWdV*u%&s0tajYH}bod34O*O6+vS(7Es5fKcR^BSX?kII3ux=J5Xf&0}aC|&V1S$MAfWLZD*7cLiA{Zhe-Fomva~hMA zFHdq{su~IA4}K^VEp(_=a>gQqJ@OuEvJ<ON7f80iFP~1S_0-A&VeRt!4UdEBB zs>|kr47mSm3({@)Ty-O({5Hq>*_HBUBk`*@NPzk&7Z4Bh7{^W*>wrT*(7%rDCnCEA zbA4EYpKokivMLgGcZ0vQzfF0MI1(zleOblcry7M&?L>_b`0%fk8J6{lvRzi4BYyM# zfZYgly1_acGAaUxtQ~H=Uw`n=X{?*#+P%f-XOG3e?o4>s3VW0jmKtz7tW?d;4p+L0 zQy+y?qP35RMn0iQhOGKjTT>PgJt688s>UF{+kQLrQlB%Vq^dOyYA!Q;UWt0Q#tk=U z+J5k%q;(-KnBtxH9nj-&B?rS%Kf16--(`|dRz;;*it!%qPHyPmk4VrxqR)`{D4~7E z8nyYk(GVvKuwRl{P2?9AT4+!TK0Rxsb9#^}<3@5N)I%m45rvlejQWDFi;G9jWRSfp z?-e@K{^%%S#E7V$PKJ&l4Ia^r0F{n!O#c>M|Bf_LhRR^KR*HYruoI(xkp?GZ(iR=pVd@= zJfWE@R_s?(`0UO|HFKr8AvSzKBiy%81V4Xm8jp|3X8F+OHXj;2Wnlx@4bn?(kL)s zFVA5iIk7(1;F8>tqlJ#cY)shA+E{9EK(s>93rx$3ma0u187J58>W;xqw77S_2;M2) z>JaE;F5xlxpd=lF`1Gie14xazG9g!tp4uq}netaFVz#l2qn5!x+yzz}2YTVSg1A`w z=yme2(2*IHKNuWgrU!kgM>&I50@N{{)bmIV62l7_F<9vEHuz^q^%Faw5XPk}a9M$* zNw}X868yTM6>tTzC-2ue3r8PT4-n8p2U0Q&lBEej^;O+ci2y5^6+TdlbW;phgUWm6 z+WCC-E)Y_r-UTw#@KTmvBNGGoB<*N|l*nDs7;<^6(?Z2)n9F2=Qpg!Qou@=ZYjOB~ zuOf4p&q-MsrIDL+>v|*?v-Uy1ZhoguqciVBHc`PQ*5O`{{!}P`@i`5%w!l(u%S4cg zg)toAKz9*~Em5o-zRVx%t*E{EPk>F|$|aZ2X@HHEq@kJ@?-m?Q@Z+I2VVXND$lnSq zf?+fvVxQ&D!|?gbH?@qdtQeK{Qp^#vL{uW@oq3&VKjKA}L#Eju#cRs~1K+mD53ozY zwO)O5aadwB?i?`;S)#wRFR}A-gh2hj!*MZRyCpo(XDwG~W^=qdqDZUoStd~#RG*Kr z2CFF^w(`T#91?~}Tj1L)rVqKno>6w) zWU!FS=0@l{@NXSu=gTPO<3z-*WuCcW?DQ&;8zf5cKqkmrXD!uU+V!5*_;b+Aim5^!;??sUZBhkFXNC}N$)1x4uaR}g&jmUjmjp}SZP6F$OUSyD~kqH6JFURC5T3|diOw!OS zhZ|>Xi3p0HzHlRoDOu|4x+j#poc+-6CqdM2IdA)gLfRRcdD`d3=5Da?A@vh)qq=7m zMjj_SkuP9X)sSqi)QpUB%?^)BnP-Y=Gmk*#ZknIEO?s6f3WTqht_~zE!ZB{a)4oB{ zHc}Sjd{Q3g6w*B}DhX@dokBz~Q?Vku-^dvZXJO_1czpa09)K44)?Hk0Uo`1+CT5bI zl#ohvZorAFhDGRB!(5If2g32}X%-xpA?btankI`YNHzK`%lGG5tvRZ8rmM_=lLL%# zw@`l?o{zSRL4abgVpL&o9mM9ZX=N;M19n%H zwBh}jKItIDsW|x7EVW17Q5nfK4`kdC})Wypp!h z1)*TCG2E2OD8gRLd6GO4^7d`k=iPYX^IezK@Tu1YciPH!+{)xKqtF<}>E5}7pCR~J zp9x~6($|$uaQGxZ2I8P_Y97W5`{MV4Z>J!)KxCPtx78!4E|(%V{mU3PEc+u?MnB`TB$}UOP2wse{}aE1mq|aU+aW z*9*>?x~fnHp~$-;WLlzyK1u@Rk;4ovS9NCo#XoPVs~_W@A&N3VMWGFR(Akr4HPnv8C5iqNTK zm%JWG_)Iq1RrU@XQHF66K-Q(XY|bh< zfZkg$vusu5yCkDlLspB2`zr1STdu;3Xy(Yk8mlu}qiviC*9HMyKy0V*<=LJ?i?G^sqfZ8d$<2wbFmuWYbeFmIF6~lZfh}{h;giBI;o;-|0Jw{tOIm zJrU_=6|2n_JlW4_7gM(Iz{UJ4`MY9J2-HOSr#L*dN?y_OJM5I?247Gc>geF}N+`Fo z?$9_p5k|{qFG~W;xmSdH`3~o00LT2FFg}xj*`JXk)P4`S{D0JdJ4+SU?+w+vFv8Sz z0#Q9}oG{;5>myKN`nsRa+9#z>BBH$HH!HHpim$f6DtDzVafd6`=jz{zkLVlz@ZXL+ zPuX3Fj=Vsm`wqp|?B%QN8LzoBScb(n<@<_=I_D!cH>k5$A zg{3jY#5SC4*ZIHa(H5O5l9dTQoWy9Zu2!BUrx92k61p3T#M#LXp4PU~V;#Gp6clr= z?eK>91cGQ^T*{i@m+zG>oNQR zKjpGnV3n2d4xDR8Spuv2d-0C{t)^tMuYU!!UMHK~YYG{7jL?=4E=oa5?(S7KYNg?| zQIfn_O<(WfZG4%QbSOA~`ZyH-@S{Tf>&|DCBmwLu+QGIcDDv1R3+V)(d$QOGw!P+h zrUAiZ#u884pjXq_kVzB;Bi;q$WKPHsGh$lIwqF%)!D>Lp8E3=bw?+ z3MaR=NLihq_KPc(Z>+O%m-YYZbg!80px&ulNJ}jHUcUF#;0b|TroN=6?8Sn(=+sY> ze`S<&WUu?|e8J$qF5S!y~7n8ZqR`m?`dc7qlt*VM#s59RLwgM3&0Su zzx2VCtBq9f1G(!oaVu^qQpBV7mo@gl8UxXEC&mq%NdJT!*@KBN{foLE-@hk;3bd`5 z&u$al+_NpJdg6SKAxIk_jzl7|RY#x;2&N8E4Ldr#PUm;oht*n(yTzU1RxhK_$;cSv zsScG3Xo#?eD5u7Cb7|BsxkejLZ^i}y+cdciZ-)@@IdNU=Tje5a8mh4*u;x!m+4!Ds{|>=toi|A=HloV zeky@pbGvA`@q?Fy-(ksm=7$&Y4hd%uB=S=Mi{n)&Lc}i|QK|DI0SXKNf8HAY&;!{7 z1S~0GoB^XckK+v@Y_H(P883pGRqjikazE>uZg@mleCkhpSGFsuh;fwKB#if+}fkTEe4M1FC(|b;9WrgF>?wVYv|8uJcIVa6ixi;?pgSJ zeudb=j3Yvm`}E#kZ(ON&tmtagiZ1AAWJ4<^m6ruh;#bm{mD;bkjj!(XF)QJ2m z37jt+dwhdGuBiM_Tf@S`1A6?`Rx)+xdyZ;cVuhINB}%#xH4CmZnsr=mTYeHa=IHIk zdo+Y;khY5lpf`FvWdKQwhzX~G@jmBq?>Ubvi&WB_;}VW5kaleK4uapY9l2ysxtMsD zPZML55L=s)=4f_hoD}*d0N4w(KSO9>9e~IwZ2T4dD@cC8sR{V`CHtT@*~|g)gevK& z9>8oMo-V;Vc(IZmJ*9HF0fNPNAg8nMDf!MYNiJDLS}(+d;x1gC;xR!-?wYQnm)V?0 zyEKmaLN|EZt*!GFRgmtJ>^w<7K1K)6vjk$Y7x&-M{8n03BzDj9jl8a;8wm5Gf&nv_ zmL5?+mAV;nw2z_%<~@v!nwVfL#A|>4LF`#B)`mDeZ}Iynk_5vo!;uo}F-FB?k}Nc|8sx~_?2)z}WYf2n{u zAp*A3elYaW}({A z9o!|m-dK>ABjDF{DlLXFB#IoS#b~DfM=%XRD4Ly>wi$Pu+@viAC2dWunf0w+Gzmkc zMB!jp6cULM1X*aOjZ=S6-q6?qSE(@v4HnV&B#OJC$pG`H1Pn7y)v`c4`^uxk+oLez zB%ei<@eN}|+MkP>ZwM7VR!NPim1iW(URJ8N&orX;S_F`hb*q1F%*lYBrrgqrAcbE` zC#~F6`oaF`H>Hn^`ZQhfve^PH39G;FUU&`UvVnqpZkn;z8u%&oZpPFRn72?7z(?b& ztzFHa|7GI?I`0$v&%SDj?=+Y*BC;S?B)vzdPxy(;(Ex`Z%s}Tk`X!JbStgcn!AEz9 zsrjJ>9{MCrX0`}esX@B7Rb;p}*}Ft;T~mBy^+yJGS##nMVD z8bB}R@9>0+!(5cA-orzpVRHDOn)2%<ZjR)bizcL{oz&0&_^3vEXgMl<1!H%D!_pjnLhiHv!rYwohG`~;^Y_; z2=HnMWv_t1*l4dpAd_T0F;I;9dHA4J0lWRM(m%_aU>M`3F8kKV(3!N8GRMf8Gc?`o zB^T_~5_l{T)#3J!PexsM#xxoQ3^(97R?xDmPbh9a%EtaNQL+;-n?xV;#Jr~lCgAd_ z)Fysn7E&IN@7tWFdDN0$SXQDyY|}r%@gyxk&@ga&4iKH>wnfd`VoK=zWe_ z%%V%`s+%<$CyN!3-OgP4);{Ni!$nQyO5at8nix@~f{|>p{wcsM_T9FOR#TN=$B4TAemShQH?-+zQ z%y@`Fw+3fmj&Z!5B|*w@z|^HZ{7RDMiRHQ=(WKNo9LzxObfHFt?L=> z!m;g*>OLju2(@6P(3R-yGHk#FCLyOVkoZcZG>cr{hTqJsIw%#A=|~Q&p5@ykyQ0l> z^&^jYF=nZYK-cj>!XONDgy+U!q;;XF4H|6J%_j8uDbZK^y22BsR;C3HDHI#&tZUk` zrKSkz{kShpLQL$%N^0$LcsF78+$0}-{M&%JO|91-We(%Qul2P-VUO~rs2v)w_oBZD zJQ;TmtYg}BI-sD-RNe=x&kRGa2&|)Cf?K0iF?JNqR|@}=g)NQbCp_*EA*3QD`r;kX zPCmBCD}p)d>t}?Yi2jeNIiSpQd7j=Es6;1C;_#FkJM$di1(?7&%(sha*0yBJiHNFB zqMZ^egINy006*;fX2k;ByhtK!4od6&?~d*_1hMfG+j%pZ+eJq$I=u64JaTm1+{Q;V z9-m4G@K*=@wxp58Y;gd`ZYGn8`tDWh~byi^^%I;Qv-9Xu- ze(iPci@;OZgojfY?Ojr8AD@9qX85A_2uqZv2P$V=GSag1Z7bAGEkJv^moLTo-SBn< z-9zHa9T|2sh}T}l)9`;cm!owilx$R~9_g0~zEOhX#e2fp3t&A^(@z4hjYQx(_6w=y zyHlh3>RFKQDy2~g@}<+#=gGH!(T;p&4Fq=aw5@`#YXfe$Ue^yo3inrM%6rO_V9U8* zo{`X zg6j_oj`}%XuKspryPfTf^1cx}%E;Rah8Kx(9n`#Tr+xV1?}zNGd>?&Fz5d zn516AD#+zW)Ok-V*O^n@0>N)Jk?*6{3N~DmX!H~-S{A2{-20LtYo~AetcnW<--tju zn^^j%LjtYdJrZXiH+)^d60eYUTQp%)`~x&dik3A0ZW9#)8Phr%zFY~2yO9gs^lngf zT{tWDR{RQ%D8OZ+w39ZfVS(NAjDcPEv} zLLNBYJo`g>beMs-QzM)7J90yAooiY3lFK_S{@yDQJ=+XaO&9KcZhteWm$ZrP=+hC) zqKZZ6z-D*^YgGYE|5z=Io!kV8nvWmF4DEl4saPhflpxKHW|H$(#O|m~JNXLDB7=8Z z=nZIbvE%533yfEX=U%jORDi!(;xd+4O{4Z3G??E*pq+~{b{9xDSUE5Cn$Zk$3WfyTw#ZQKf{2%;Umw64UhPK~~o zb@DBBkH!H(o-^hBBg*K9gg2o2_k$X>bfYzsG+1TY+Pt8Gc4mAtcH0qqWqH(0U>-aW zGt;fBcoFCu&*8bNgjaN;_lt6ldO8g-wP;jqvR{!}W%~$wEGS^(+;(zO=McpJw88GU z)=0Pe`z}7!P@~LQd|O>UIxl7_Opw|9y$SXr>aAD+I~zqZqN#}Ty51eC+d}^c+ z3cMEpFFwvFZ~Nad5s1EMu|_#eN8eUlv*GVDy zv0%TINv^80|8v98SG_S72d&cq%Kba}%KpgRVG~)-cV1Y17v!?*ty`pQfLS>}uaT@A zepD|`Uuycc!#XW}E4z1TX+#0@Lu+n@vY3tI30Q-ycVxk|BJBshf`k;KbPNhwD=7+3 zPxPI?!oQSsws$97m-r2nm+b0f*9cOZ}O(k^zzw%OLi<0*|0Eig!R^+++>foq2 z^i`i+TI>fnjm$l!NcWJDS%+)!EdEm@4>p}@iRf^h z(nh4ch|hG^Ns)oGDjK^m-F5!q);m4B84MEvKgZ#&H(F`~%FE2)=+e6Rrf?J48Bmx4 z?C-=kSqMzlAjDrKBr1u6@C_|gtqfK!jBk`AHcO%c0^7BWMAU0Bsf7aK55n3HXpxG zIPw#Ok0b?%nwX&aGugwJq?60tw~ufeoR=3_!EVLX2f8_Qwg)(ON@QK zn~FPk@d|a}n}cQIprY&BBULfzt5z3a=7+<-+b1=`i|!=FVq1G|e4@jr<7XMTUKNJjSJ5u2Fos}IwRkR$`wRxJb*ICA8Ti(r)N2J!PTJ;l|$*H5JEa?cqjz!r@F zWz&2B+J}~Sf;FlLN4M#0M9kZ_`3?2F7^UFa+QZFKQB9LP*Mn@x*>LSq!4K&7TT21q zIEP#F5XT?E6P2f6XfYDubar4bJ(LG-&Qx_BG2x5Tc@)K*sC#7QjL`Lr;nu4l4)Qy% zUCLTQALT8%gM=R2#h^bpVA3tf;dLHzA`IX0x?2BMJ*KcdV|(t219h9YlC`P8Q9?Tx zbq<-n2!}|0RpH^3i0=(Ry07@DrW-r4@$3=;BlDU|Y-4 z@R3n}nipQaA}&QxMZn6{ zh4=+y0t+S1QYNagI#i)mijMbwl-W!>|C!Yft5<|T-hQ}5+J=JzPmc?$K?aVIkz~rgxCV!|lVM>+ zr{}`kn)ObA)6QgBge}Jd1+sGktBV1n^kP}jAaklrwkXTF@O+OWzId>b4Y^dk4Wa~} z7F*aqv^QN0F6d&3DKVKD|Tv8y~e4uS4nnuLR}Az54IbTtN2SS#J_cXLhz#g11`gxDpQaRq{GX z-D>x(V-W6Zox0tz(O7*{2-klZ_v8^)GFZoUJ!nza#y7HpyHlU&nn8x16HHAjYH4&5 z!wA{S2o7$QSKl84G4VaM`z460vsD**{Me>$c^pZs;+Z|IgcoHYZ<}~SfrVKE$*Brz z138;6#npds;|=eS2sFfU_J}ycldExEIa!um5RJS%IKkvPQqqtseo;z9(r{d?VaxK| zQV@{y)t{%3j{OL$;5*15W?mp$qRZ{sEaG($Y^jBb z9}b@uv%k4w$G44WBLBgQ#|xNF^f&-rci2z8G_BCe^UE>k5cs=+&zw|j%f~}t0(8OU zUn;Ku1h=0W@h@+nUA%12Pg7IC)C|1m08LVXQ`REB**E$F&L}kMbyaVmVk(f zA~$$Bi9X9CZU7S?G2p>`-DdVdbE+K>Y}8ltJR%h;Oq)OP*j!>2F*nnq9TYU-iy5DB z$+KZ<572^@-X)49FJANq`m@s@)?|w?0OadHb~T)TdYyr-)%iURn2AD*Y`XYRe)gl_ zK$f8)4A@zgoVrC;S@0DhG7;O7*s<}LdMtum;d`fjxC?@}qSFuU`x&qXC2twN{L9gw zUj2be#T+!&4}jvQA^@6_oo*kYka$`Acj6f6lEYB^t+i3qk(G7#<2C+Q6Zx_f0PDNt zK)mzsrn*5zLIs_jo8pjt-~9u91C#Hwc_5$RguYvsbp_cLdf>m!31ry5mi5|s$|S)V zUi2q1)WNbv)AxWx45_H$Um}A;582c?2}Th6AG+EoBt4ukKitt;&`l$iv2;HjWw~3L ztvE2JcGzNlO?b>Qw63npTAi>alLT#$jrMS%ixWFSZ$O#1L?6+ zeJI!s_2wuP1Up?d^hH17>te>UzgN4}tQ#Z4^lo<{pA)0q%m$3Qpy06=ttsSR5Jv;# zBz5v$i;Yz75D{9^LsL30DsO3h3Z9;>@@dJQ>V6_h;+kW87hV_OiV&p^r+ z8>jG+Gm6(MSmH?C#p=j(%YJp|NIo9PWjxsC1rugou%An$Ct$TzI#W}|hBl>l+R^)( znFI6l4uVatg0~>S?|nYMgX@X@r|c;D5%JFJP1@r4iCZ=eg*5Q?5+2XVQ@+fGgaWY5 z)vhf3uR2!xPT*Y@VT!;v!jz5u4 zGAz*S`RzUP-x^bT8j{x(#tWT=X@2f@PoB;1yzi7WO3saHn#Uk$(8PKg&jzw%EZ_B; zLDOB9Hxv6li@AsYf!itMGd3i`^W5=`e4)$b#Av$~$c{{wYK=df==IY1=j)hq$aD$0UJ9$dR*ZeDNT#ZrpRxRo~nc{zAF1U~LwRuCngzvxK;lfw3mIKdhGaS{$ zwn(k%7Y*3m?iyg~yM9ZTSX&1#I@ES=em@O93hDy~r|>qH%cgHo&8nj(nxQ?@k>6ZM zogZT;_!jn3mww62C?vo4kxcIYcA+=AVH1YHcT5QR06(0DHCPic!vJ$__{h-cL;SO8&qy~&*Gj!}P=r8dNBj0YgGHD}(-Us8jD5=mZ} zwjErod4OBw#~O0GsnC%&`Jv%E0CzU+w;+q&G{o4@aq`v>v;R~lz%Q@#2P_Xd zETnkden?$Jml+d1v@On6$9{^*l{z(e?c^=qsi{GwPb$q)Z!E&>$o}mD9al+_GJ|s8 zL69Q$wATo}k4z?354r=qWcK}i$&zkr0LIbusg(Vq0~7awspa9RWfo-s{?v%g8Il)l z32EJwJ||j%n#(B-U{mgLUy)$yV_OI!%4L9d#@SP9P=#XmUy%QmFJ~@*hx%<4OuJMu zCdpXMUH7h(e=m;*Vp9|^KELq{DXWV?el1{13VF>;cE?!a*$?r(*DW=fvT8_T53}@) zyd#rL_qS+&7J<@{t`o;9Hi@qRH1z6rX2> zd=uF(l|caQy-bsI08Tu4_UQgwK)QLi*Ph~%+meE^SL!H0dqNs!wPNOwIKeGgcDH)i zo!Zh-3nGVgL#LzHjS6LMWOHbY*fNFGg%(bY(BcP;M4-QC>_#rf!UKYBm!-*06lXXem7d-lwVoaDp`%Jf1eU?ZS7 z*v65bnSqH1AS0^;wl=h3r&j`+IawLn16UZCnAqUS$wlmehK?Yxji{j`kO#o&XbzA! zc6?8Lk6>crh9?I|0BwNw?@kkdksCl3=xC_oW(#BnP#OL+PyjnP(i<5%yt{!mW*{3N z^?Qm4*w)P+WM=O8w+1Uc{ohJ|vxOM|QijHsU>65V5WvvJ1R%vA%K(rAySy7g04lHz zzzAq=Xk`ijn*vmT8UR&gF(qYygp$0ff-*J3dtqfKTU)UGe_#<&R#BCp1BeRAsfYo9 zYIFbzRb`dGuPQ*Bcl>5_06CR+|KC3Efq&Cw#Z-h;G!?{{8UM}!zzlE(+B<;$w*5D3 z6z|9Y{|)Uu)zltr{Z9h`mARv%Ee|83i;D|`nUjMf1K8e-!Pe>@d@AN32Y?IM-V*SB zwFg=O{~^Z7#^jw&M|0r63jAFZfDFhOXyX9XJ65U;y{O%H?B9BcJ6T!D8CnCW{+-f)k2ACeS-JhM$bXAa z2mZxMB?q>*HnjRTAIL!*UzC4Ls(-hW)w^!r z&mPdi?nzFe)f3i>k}g{vXNw6DDS3 z3^oDTm;qSWIRJ+C_J(foOz%2iVP^+;FuzOM1nByYPymb!Hekp16o9RhqZhywY!Cl; zG1<5Qj9?q!U*kVxb|wI$qYL=I#%x>wMt7k7zYPBypu*q%`H%FN{u}Xs99;j5l^yND zmOyoo$@`)GKSE>;9qmD`+Dz}dVScy2KmPNb?*D2){@;@SAKAjfU{?=%b`A~zJu4F{ zfSHq%1;EVA!r}G5a*h9WhWw-1cWVE0{C5KbfIwHEG5pdz*qAraBB?RB#9J(HvJ{G( zo8hz^i(f-37IG{&!$j=JP_0Bb-g8_V&b=n zqTr69x2!ilvY60FmKuYqf12!~Z)pb!^^sIou4WP2blT6@pM(I_kz*0=w$6u0mhm^V z>qHWr(x0n2P%Z_t%uN{fR-auLQSsXG7bn|4IKDk#M1MEz5n7^|F~|+Z8j*sw?e0Y1 z{qVg&;5mY7H;v~WU5x6%Zhcwx2u3^1p&j811c49Pmr^qc0A;)mJeEspuPpYDPoWdY zA7b+zI-LA+B_~BFGXsus8_=c}AGuj?JT(PXR{WF1PFH#klPt-2QmiH1mC}OES_PV1 z`4C)qf6#~lV$yO>q-q~!xyI7yZk_HoUuqAvto1l7#mw$&2GQJ{2)NeajvJJ^WyfSU ztAjgyX9Jx8HbBY0kO}O=6$tCBmq}PYFqG|rwvoE#KolW`3FIhA!n|nQ?N!LX1r{W~H2y zaur08fN*Q9Vad!2t>BVOX*DtoOu6o{TD?&>rWD`>lh|ac{I&)Am@KEI=a8TUV4$Om zd2e;g3kT|x>RA}7-}_8gL#GP7n!||$D~9R8aHs<2noJ10WPAJcmN~lkSqsTSzqgWeJ~&^M`6aiA8$j2aSJF&?k}XF`YRQyP z>^s2#@`cnVj6m2vjmJ$x;EaQ675YTWJA+$$n zLNh+!mUWH5)n-LdJm0{oG{N{5iBbomPceZ7PE=2s0=JHmOv6RMA1jU+LWcl;y0{T0 zfQl@I@Y@6ma&?#O{{ReJwsNJ>}TcLjzyY;`x&wmfd>xjx} za2!(+*!|_B(x|O?{75y)_XF1z=ybvZz|Z!MMU&m*$-g=rag~BmoqP1;b06})vu4g= zy1%gCF>-9-HT2~5C&h>N07GNyXz{iF*~$rH%8Kevs}ngt~SK zE`61+{PS=uZOA!iw8tyJ@qy^^B^OIc_H;+wq1~BP$zi0an*>pR#|*bYFM4_FPTOq; zPpM><1%vFnjqg6T9Yk<63l)-Fbvy+YKf<;y9tJ8L&K=amo^^3ci5{cfk;%p`$voZy zMDGd{`G4z`|koRGNSMO zD_d`Rr|)d}O;-kjR-a(xI1L-zqjy;@=sw^p_s>H;0Y0?o>Qi#gTU%1d})xFo;kVnfi0_CYE+ zhQYw!WPP!hj0nYdn)18D$2=){g(#@QA=p04c%igZkOOq}$6c%KNbDV@CHc9yW)Q$tE)zEdyXk?O{ zec7w}AQuOW6H}D04?=o&M+SDBsz@tOnWO<3VeGVY^D`xPZLtblwy@ z^plbzZ>Gc{Po1P&QH;p%tB9uf^4cj9Op03|Tg1bqpL%a|w{uI3-$6Q&<3mW2h3W6zhgF9(r&e>#+ z=zWT;Ajaw=1!5Fh_+Pw_m%{Kk^~gUM;DRxFRs;&X1fiIQ!6hy(&ds)(=JE&dNM=iP z))(0P2&&`?KZr^OuvLQZQ_5U{u^^P!`sxn&*Rzr3o05C@2mc440R2TFhgK_Qs42rT zGbv|>361=r^++}76MR=*!VzE$jGaylNAxp3|;QK5;ShgRMflqs;0X#tF>!ZdPRzk;(luz16gf}%8K(_2TcRI zXbQ>`SK-R3yxPH_lB%pO#}PIsHLQp*8Q!r%KP#w}Xt;!oAc zx0RH0Dhz=uAFP0Bg45r_EBpek6u9}Lyp+V+8^TJx;AnqRDe(rPsLoVha5D^uKfu{| ziH7e6{uaS#cN0ClyD2Zdb>~1g`xBauBUVo;Wry}^!UvP$mvp_mpGgavlG`jPdSkzM zX#bGQ%Gow%O$E&@8?>~Y+8kgppZa+<-}!7~OV5}Y zK4Meb!=Gc3R{afjo*pkprkpDHenY0v6YUhrqcS^i`qNg@m;#20!8*6axD-4TeBwtY zI8|j=ve}{t1H-WP;Qf;hCc~$bzE0 zUprowsS~AYJG26xb~qM$rdaeo@mm#D6Cl_Fe5zo#Wpv>Ic|oy=x1&E|nx-2ey(};^ zqnYq6ypuTiN1DcVom!2VY1?}A&s4r}phMdYkvV_its>uAw(t~zefo+qV)YehAn~f+ zC_t3lz};zi@4zZISb;)vmrzpE2sNU>gwk!aaMcv~)#x&##ucaFssgjui@&ZbGsl(| zwR#sBOVE+^IhELYuxk8^){na{g+*Sv%py{7jq=e;#Jm#jwqgvP0@q#5q>i*dBz`8< zR%2{OLPIer79-ZM^E9OA59Ypz#w-;6N#(^sB14v?U&)_62n;8`J_zfMYLGHk%CK{z zD)6jxeSo;<6ALCg*)N)e_U6TnnSwm2Xjz71L30qLf|`hqw#+sFcsMd!%uEz>=Uv-d zyzwx=QE2QQ=+d-&$ECDDl?kz(*{Ab9PSTv=%(*88PKB^iYTGD;*0580M&P zUl(`%0_sl)Zd%#nk&b)R)%Km9t+}kxHy)-d3y`LEJ$oCm-kHo56Gda%&8UgU{l>rH ze`9sI6=WLq>)APD7`zY<;a7~C9DREFA)*>vH`-N~>R63)V7&p13pUH%Ze5ZR}R2$qtv z2{I+82*UEu?;kREr1IM@`NR!=ur>&Lfc?9SEoa#a(XW@`Z!o z@1>%l=YG0^+usW$-LPxO9kR-@I_{M?oSPIfiz<+a(RPV=DP>?5m8e^5TNzPm>IRGh zAa3Ks6{58*x6;ERrqHX7H5dq{-Ig~>nGCRc8|3#YSM8+%lpx&+uhacDcq5QLrUUn<1 zpr=}eu^_=xA%N`F{?H%-@rhDjh5cqN7JrMN&aVqE9nwUNX_)MO|hM04wWv7 z7o$JZzX-2bIEMVP1EEfNmIO*h__D7RK{cL~qsFhHc8sdhMsK0}Sd?W2d!*g&0K#|8 zap*2KtM0ZJ=1Jqo;)S&v%6h#i&ZmIrsS;bila;_t{&+%$jsY(oiT*Kg*j zD4+L370O8sRZpdHcYZ)L6ip1yRp&K8vR~g#`iBwS zL(naZ@$6F!hlR{M-hh*j*}}wd5PD+aBa=V&+Gl~zdkoSR(F7ue zwwE~{H#OON8>8~sYCC7uSy+xmJG2~~UZtP}w&6ODkt-2J&@)=AbH0EisUyVw@IL;j&-CBt%%*d7O3@NAcn3}$R z39df6K;Pjh@mr($x8uX~JM+oSoDDD7>;5>zq#6xrZwN(H?*rEw1Ti3nT7qb?yOJ4LC9 zoDA9w1_r{5B}|jca+L*?sAuCAd57{=nQ1pWuu-n(7&n<q*&| zm6#TKhMNBOy+9x5tD-G%j%SBJk`o*o%01gLF~a5O%*lH7I*b)D|6)LQ{lJ62TBD6Y z_*qS(EsN+;O!{norZ~jP;A^VA7tV1?=Cm8Hf@roJBxP!;3IT-^a?m3j2iXqBhprs% z#_zZsP(RQ*0=9{rURc+C>;v>lC?HDVB1F6LNv8EDy(^M*WiUooK}~Y1-RLJIHY<+? zRU(5k4rq@ctw;LLaBfEQmkm0fU*dR#vnkC93s$j_n&!ddR4b0WL$nON&YUPSs*kXP za@Y31;}vV0Lf{W>UF2r5>7H4U5!OSUzN~*M(3gYyG!qLfU}6^5(Ies%wpfn@LkMG2 zuglh6d&QZs8Y)54K6L--hpri%#-s?Cef1nDmu7WDHC>Fy&uPX~kGK6=VJmj}Mq0J@0pT*=*eIYOjmB-yjF@r%}k8U6WEQdaz_^D zryjb6`u%83RJX&JJ#Sh?Kj9H&r)jww8FCrRx2D|2#bn>&O%tIzzH5*$M_;GhC3<)W z0vhFS+z%cwSzksSBc*8I4~CzOd7VY5HC$yZ|k9UJ)h| z)RXzMNoKO%BB?*LC;5TSA#XT%9?oD?TsMM&@%ppSO}L~frf^VSCaQQd87ITmb~U2% zMp9_KYHSi|Y+NdV*i#9iz#7C`lADm|B^#@G2{yv%q0b!VWxC&n4%l^<>K+=~@hHWR zlD}nlysGWoUmeEO`DuXbo=Mo4Y62YxWdf5p%c7LrWu;x+B#Jv)3yM;)pq3(3P_oLb zYE>$ks#+QjTI98~VQe&@+_!u%jHVyvimT`p|jEM+fYnMM1p7Ved7MNV&5xG#?$Tw;57Gb$r)zy=L?iVht zqO_dbnpu)NV^bYMA>i01P|8Jp9oAPc0uyVFcZ;r_#&~Z2rcmN9+x=rFaxyy~jiwSG>2I&(N#K-yvxxgS$IoKZyLp}!Vu5zvA<=E^9gBrdnHe1pI| zX7ff#M}N$DDwh8Jibpc>@L{|H3!+j`c#TC?0#DZ6L~=nf^yUs??$uV^CLaJ#GE=e( z2dVmP8<@0$B-t#z#X(PfOA3^oA%ywKtx_907RPcQ%v!=4fFR<^ZdWljJv|mjgA3Hu z9CmB^_TyKCbLy0VOtVoF~p98AA9a~h3gAHd} zftCtNha?(br#~fXEM*^?!%|MEPPQP04PB}GgDLOm_<(`td%q}{=V-HgV(sdYY5Vrf z>Q{|((_n*xjf!0&k^C;YO*iIPS(7*IV@5%&jtaJJpzkn6Eycu@o>f7J zlc~e4P%JecrU5oZ@o+`=9z?!M^+QI!cY!M=OP~+hX_uJ{wO*m)0gw5=grG;Um3*JJ z*px)F^3?5B-NAM9;^O$ZQlHT6;m27*g2Kt|m5ZR`NE}{9yO_4Zh5NiF9|?Dqz9iep z^(mI=Dc$-FLjS6g;LhE?$fC^tka`0z@1%G@Lf%d5Q`NHAMH(#DqkMqARoXr^;SIH% zc3gCTAcLq==j;90x%zO=F(r!em9{Q9gFtbcg^U7CknFOB1=^&2t{zkC8dAA1D33bW z%MUp2--Ku>%G(-!Cm2a2A}4Sy`w0ia!zn&?S?lnG4aKHvX2xb6P=kELj%}(zkxiI| zm?o+se;%6_isolRS98}Ze~Q|M|{(Jy~_K8Elwxp_s*C!|50v=aLO5V+mLH|l^>Uo-<(nkqVfgkk?W6g@KK}SME z6^7&mMrqi)QIiak>O^2oBrpajp_ z6B3=%=Qo&Bdiy2E``BWU!jE%dI+}`WDT&?5oi$PRwSQ{eq>*|m&k#Fr*0}2fwW?!7 z0%6rxF|Z>H0TxDgVkfDiSw~)qhp}qqvEZJJ`jA4DC>_2j1!`I<;GsyYnPjUSSpYFVWH={au-j`tmOxBG5uy+&3c zJ=mNl8XM`;3_AF*85lp^nxq0v*;I-j%e#Y&GlP2Cz4W`;*{~v4@&yAF(>1`Bqy04T zc=kM`7H4m2&gJVi2Lwas-RK{+0-=u=wQly7!X(6uo83s~ zNGI;S;6s*ra+_Lne8t@pR5cBSWPM+yuzb8Kh7cwXDc9DdRn+m1cY(luiaKpsT6;LT^t$D zJsxy4F-#bZ=I*v}QY6l#_6@Z5^(44f8!Tbk8n=nV{K-Cg>GN&mV?Ko{yAlKKW+`1e z!}KCCg@gr>CW5avo6iY)_Fa#)fQ7K2#+@%x1%>gPjQRAf(FggJ#r)sA-=#ayx#>*! z67nn=2c?xuj;Oojt4dQ4{zw@UCs95~zD*R1Rnr%;!k34pj3F&ktSYC?%RD<}YLJ=) zAT;?9M>lgGH11PJuwZXf0En?CI{3B+R-O=QwQIO-+)I4?QY~IVX8xn4QQ`Jjy*{~K z0{o#&UejnIU;Uk_1uENI(}=8plc6;*)r|}iWfV5p){CJ#wH%)vqhSWxAZOHQt{; zM>2~~m;U^xInsRE{!=aEZ!Oj1PdO~51=3pGO=83R;b@OM}50;72BHzsj@28vR*^J+-n zp*)d?B=Eh|4+P6xJ=-ab0)k;w=qJ7Fb(crVu>fuRLZonj9v)wGG0m)JF1jl}2`WnJ z&a6trov{Klyo|5S`5!V0Ozb5?U5$+>%GoM%G(BiZg-pW@l(P zL)e}swHZk1_O(LSbt#mY1vAqHTIpZ|Mszf_ptAZUFh$bqfC?}_K9joDXyCWmrPE^Q zUn}Y$Nz+^}^+I~5AoJ+bR&mq-a|8bgl6fctDfur%#ws&07;N~X`wet@VSO^B| z2V$NaF(@9qK}+=wc2uT{*p7@^@ERN@<>tWu{Y7v-ntreB7yQsdJSKliU;Tr)dvYrc z{>K3KUQOP0iIh zk_t`Dx%O#YU&Kl<8+pAAV)1JECUf@qel!Zp$pe zcRa<+D8{4IaiA~XKJXNNBZ}U|K6?#| z5iQ{&@q;?MhfS;30o7L$I8snPrm@={6qPa77r!&V>wgl$4o49SYfz%Qi;6^(R)Rs% z$iFO8Rkys;n?{(SL{cn}$-R51m99>K0dW{+OEkQO*-unRU(Uq5bx?)y0*(>fu_z)C zLK_j1`rVuJMv!f}GLM+p6fred#E;rlCGJ{l++IM(5k;r@?l_cy7H+Og9yNMh9p6mRl-`qtA#>a%MwpF3{m;mv1Lt ziwHEA6(9LWq9aU$&(@Tw_-<#;{Kdffml$g$J}a=0NwAm_;rmnCE{)47O}%52{_Nr% zBGfJ3k5aF9C!*eU?o9FbZ6T*BeVcu>u-YbuY6#0m*c6|2SeiSkXHY$JDmNY2G@UyQ z{M>$ZQl&_+wDo(%Mf?hAJ}|Hn2>2?wszkvt*Ysrb#{|wDd)Z!x=SG!$(k)M$o7(!@ zq3%kfM?3NE3*1sV43&(zRY)ZjKbY6}8_!~HWrME{z3|nU3RU@HMN!>HrB9^|@%xhb zZD)B5k-u2uU{BVvD0|5wEQ<_8TN)?Io&c_Qwp^r|rOZYwNNh+BTEW>-71Kl5(I%yo zPxfw)aES)VPVy{3pSu1soubVA5pP2di4uXzK{a;vP#Y?08etM3m?e24B!`D5$qq^q zsvyF3crt$c=Z>nU-=+O9f{``6PE9kpl-zA7MtWc$ih6)SLlhap){)SJo$X0X?@1j{ zZ?|ZVZ$0aji}%`THKox{9{uyCgUwh0;SYn*v>)Tm`GyQNV? zM~H%@%`4O@hzs=zPU6k(1*5%xYKOe<${|gh5jVN^u`;}h7~?0&ZL{PG*0S?#1h<>AkkwZVsNGiRqB!b8|THVVFxS=n|U2Ew7H8ESw!D zVwEj7*-1@yS4DO5bO-IogbI$VT{wvCE%0bO>y^t=`&%S&!YJ_c_&X~L#H5mok z0>IsiYW+-d+WNCi37?dBu^OG78?F9) ze;{@d8Kqvy&lbj!s~85HJ0UciFxOO!elo4526&5we4SZi@ms)j?BWWcmwely_~XNJuABv9G(6X9)^Z9ZkKuQoJFoscasru5;_g1tJ6Om(=KZx- z$0^{0FLzT%!p+Re#z{&}(8GwSp(a9as;BqRnf|T6yxC)m|Fo7@A+DsEZ>gc%9sUVT zJ^=I!oC&na9vxPe_1T0m>hNm?T@9yuzigFjSLLzNc`K~?h)T}Mk*9UJ6q&1_2+pw` z{J4n0y$_M#PIMCh57FlkX;Gt*Ak*}wVj=$;+S{t~N{~E=OCG3Ho?GACuLrhKa+up| zztLJ=+tu$1)EL8AaPJet_vpcxrfXO8zhJ67`N2Gj7H+U`*yFtXnvyv{MClMTt_UX-!boDrp$7lB>ZBk{Nan_b9xv?UpD;+ zWpXBrQn4HbbYi1A=9+Pi`~gCz_pu_z-y51j6XPjt>YbPlp)MH^z1hJAqoYp&7F-{h31=t;|Y?gg~i1~A4B1bJHT1lzXiLm z4`$v*9e55^o`~E-i8n}OoG>YG>n1A7Ziede79`oo#`Ivg(uG!zvYHi(fIsQF+3A36 z)8@L`m8Sx54SLr+U>w)VE57bWvhbE*b|a;V_&hd1%LkK?#A#25NIerqN?DPYf4esO zRz;*?C|s)~6;)*JM;X;GCV5jrhz)SNApXPC_L(5Eo}2hh5CWLAR$e=#0qU{26H1r) ze2|-f9_trl+QBhMgF4hS%F%G09b)o&Q`_)unlp=tCI|;jbdQNnL&g~dQGTAo#A|6E zpb&s~iuHj{2Wb+YTDmN;0I^=OQPQVyX_v^L@CxS$0S7^B(voc|d(bi-*{Bt%a8y6( zc}DVHD8k~yo*S-cZDMfPjZp}t!4Fxw&2b579u10h$c96d_G@evQ3eCi--Pg-KfYYL zUZy5u?QN#892=jy&}RwywG{@uot|rZG-7ia_w26?4NZ3G*++XXdgligBo-FZ+xROn zDH>bpb2nik=s=`eU-ZmhC|7=~@CIoj043}SA{l2s$^9k`hKOGzdPKYUWz2&7NGH4* z)O(+FIh%oIcs$y-u7O)@^CY@A8R>0^o1CiAWm=`IGKRw}TA8V8opklv_-wMI3oRsO z&lM5A#~rHf>j-zz__Yz~e6g-s^O2bTtAEpGAtjy*e{GG?$J(owrl37tN(y&ao+jNG zj25j`mhZjgl6;3fmn0{T7`aK-QA|qPnY9p0T}nPboQH;GhZ&m&Pdfa{s>jgCd#Zne zG0+zhyI^xq=k%cIHa0e%I7e7N71+gRI~M6IwmQht!$GPq^3`K+KQRwy8Zfokp0}~! z8l>luK4JrCDFi$Qaq2|;@us1inz~PL#`jOYx}xIjdIYkx<1Lho#WK>q1?2YGYLOAa z`1n4ji3Fa5Q!%Oj?1iV-^7F5YuR~lUOioWGHnEL!68GBmpf^@!AMkH-sRi*v(6lKe zL3bfT>h>$u?j1vg4v{J_9LE5pWjX0`n#6P!x6Yzq>9m$b(LR3SK*1Gy?C(}=t9Ez; znYPDMc0w30)VZ&Rbn(znjmwVD3=dHc? zUXljoS^3bWaCvTV)xx3x3eV#lz1>O_k5?9I=fWR_!xc^=9$=*PPPr|J3cZx!1)4%7 zY!1{R#D*jNda__*Lo&PX=*1MGEy;;W5*_Zp>XK+L{?42^{pP{z3ml;1ZC2Tm3K0D%GJA&4D3v(7Jd@;3^&O2N%vOL!=HY{e_ z((X~1Lkl!1>ouJMJf;E(AV80x<1CtPsHvec%r30h*Y*c%zF&PxuAD1mga@x{RQhCT z6f_1(WM_LMCivAut48No#%wLIsq^Q+i#mBxO;!CpcGq=6GTz+1$Z|!3OxhCq6=3nR zFKBk^F8`ZQ%ZDy12r|#B^Z+ffw4tpwA7X7KPPi0S49Fh=xIeENZ{JcE%@Zq@<>`<* z$50_Vq6b`7zw!zfqiXa@_ulWqeJ9z=R^WC*?~c^pQ*&?UOHsU43>Zd;NtW;qj-WORjZg(8u-)cluEu<+yVfkwx&@C;vf&T>vQ&qW?J!xn*}NI*!omQy2-;W9=7R-@@H|L8Qk@R9Nigw0=y}|S*Sv_kw%`i+Rw9)%;!)9BO~Z(vQXdXDpcM?Ab;-C| z46L&-yCf^AY1Mq-L^n5hnGN8pZv?f8+~M&5xYrCSo>r1f zuDOFm(}XU8w_#W99;e+t%{BI>1WTu~g3M%b5#?@xNEEfhIpvKA`FmvUvOnCl~^oG1Z!#eJeF?jtUPXrg^gO3*dh^qrW%MyC%WRPOBW)y|9 zieWfP&L3_NeXi>`>p!!cdiFhd=ea4Xeu-=3_O&j(*}FPrD9d~?Q15(GIe2Q!y@C7; z`h%J0_^1!*I{&nA8K7ZTS~9hv^(Sx{j&o2jz|pFtSk>!UorfF^hJ%yH)r1<4SHOwy zB3amwDTk#){U#E=YO`~F*iZFDTAU<7mOFv+iWE6+gc0dy$tt5M)g4As#c%A=jtWj_ zia7=tlH5pYGA*eWo*mlC$QR8uVk7x!A0U)uXy^tg&>hbPFSA< z?93PdGzG~zv|P8a6o;zc8Hmf7uQ+m$aldHIIU-s9{8b|tDQs=bp1Gm?72*L{(VU5m z#Jf>M z^7jqT%_jY0)s=PUC{cV>NA~`;`tc%&yCgb+3;G*ZwU9VeIGxf`TIW|}+%dj$mmB5q zRy5qJuN&0f?bN+%1ipqNvtroLey(xKAG?wCF~()KmYz8O1atAmM23+H@k$tvh@`Pt zU@>NogI9LNIUUzGdgH<*uQJb}yR-K?urs+(%5#JD{mp(|Ioaq34i4!rzlOJmPGnC^ zc5g@rg%nV%3|s^SiJU=otc49H_3`Mzq@6Q zOWwBZ*TWyVAt3a1TI)C2Wn;UIMD^PTc)hCk#-1k{V#tl@x}$@YV4%-XL0M z0OGPHt&r*aTQY?E5-tmFfztRsZqDFHE<>%dr`UzVX}h zhjp~BYa$bZ?{E6<6TJ{=+z|S^scz#5>1U1Fm~`9|JF__Ctj0LY5Z;o#wE6ywUq5LJ zC?w9f5NA|P?qZ!E;%&MuK4HJq;LgTumL$l))qk=J%dMUig&PSJ--wtJf1`J$G_{D_ zQ%(2#Iw!0~$kT9JhLGz|X^8cpgIR{1stQ~Q!9@N!kyQ@sfra{BkHh*yxX zzIqn<-QBP1r6+v$%w~QiQeh_+p@YEk&zfS@jC$+{d6XLh^7i;m7)fM3?Z--k#L9h< z1GFM~^adg-Nf$e?1x>raJe7L7FegVwr?$qf(Q`D8XCKv~qQl;{q939)xqs9a@rJkA z3%;AI82mvF#PXLD+gHIW=N&w|6+THewKUn4`OHiWp=2pUjQ!~G>q~wE;?D)4*V3_M zH+4`%)QhiKUktMCMrMAo^M#8uL7{W_+ZSK)AjL!Sc;eK$v}?$Z{+9lz=bodNmI6lv zpDRQCWH(CLKXn!$T9uD&QEMCOb6#<4P;O`PV52?Jn&U#PrJ5!FY%VdF8#jW*M1Tp5 z_!Ywd8U4z>a-Lr9?Xr+oWN;y)oHrB@#0%q~Ntq-5T_~coDH!*}d?7b4uN%VOj)t(CXRl1fXa8p=8BfPQOX&IJXjv>*)ZneNx7!ZNEp z*FX1My+c3oOIt3=2B1EDLyOB`T=+%IDOThXDN+Ahuv}~}ADjZGNO@Ec1p|N}EY41T zbXB+m_8_O*Ox{Ep>1Sw2ogq_V5Jr%~qEq_)9q);=8pS~;v1+7kg96vx_D=Va0CMOm zaF0TI$T;yQB>FSQhymroAN3e@BH(0>xlBS^V)jiswddV8hjM8dmoZy=!;ntAn`RQ% zRg3mK?xnDMbrv-B``w>`l1-@^e%l{MwEuugLg5brVF)exi9cL68n0ul;B`j0RzNO; zNQ1A9i*$%==E+;qhk7>!(0Q5TqWiq`w*^B=$3wOE12TPmp58EK7*EGe*KCq6K78|Q z*&2LW;olnCpEw127FU`}QA=eo(WubS3;WGeHh(->ZaDmb^Q zrql!~!JkO*Hm7#ZFNx1be@?o)V*SzuZvu96)Dlu<{P9PO0osoteqcqAfvuvEO!4IU%HXb+uao5*b znX+OfdMlVdp6%WBqQnVH^G+BPpwUD_9gv65w`-o3kZpkOfzwTXa^&JQoH$Z!Fu@J*PH-?#!#iE5obaC}*Q=1?smAv~QWe@ow}FKe)C zcJN_W*&c??qYn184H62t(r~F^F4d=jS5#}pRCe?h0W0wy_&-dV6vn7p%?9P@PUHq# zdk#Bv-Mto0Wu2A77lHDR#1bCr0tjXQyotvC$~7gQLlZoT=fe|E<^Gf;LJ~j2mMGZ% z?IPb#&yNk-C5o_n6w~I77NU3y6&;lry<|uv!Z8O2TGZgaaKzzrS;k;xh5BBPx_<&B8s#5G~3(8p<}8m$5ks46V9CR*Hk< zq|ogRk)OXRl(kXmHF>00(hvwh$}Gdzf`292Sf>+Kov2%#RmJTO2oMbO*&Gt9q)ry# z>)15uCb!mW$8J*Lh;2lNQD)CT<1U_4XWvoy(-fDkPMV8mmQ|+9QFZ&j(S4edq+^PW z(Zs0rkl&jsqaWu?U||zluPlco$%DKs+TILJnv%f=@!Lwzaxnikt@OMS(%wXhBB!dBs zIX=9AF6GGUdvaRKMBboQQlyvfzjp?o6+l2GMao2HengM?%E zaXhgTt82rsXJ@D;^19t$gmT_LJJrxSN}bTeb}y#zN?>JyT-DA0ym%5Fa_O_EV7M^%T(kUi& zQlacu!g{oQYT>HF?q9k1xZc27Q4)l+W+3-!hItE{bm^p!Wpb3tu0NNTtI`15qXV_I zFK9NOmxNo!(zFRnoAXkhv}xyb#L^<%0$kC7%}iCu5>&eDmz1c2oMs)WvmnQQl# z%a4(@w4>kmPefPT>`JisJ%m;%+=h~x{Cx)gqvMe&GSmIouAeI0?im0h>EYQZV!Bj% z-`_{Tr>Z3v_66zD4M3F`6}X4ifTtB--MVSFBT5Gku@)cb9EW)43MyH<&CTHW)`l?5 zFZf2X`vVr70&AJO;(Kd_VlT`;sh|&f3ld8OM*S{jSHEXj{}ohicGaqs9t%;k=(?H; z<3;SN7q(dad{ds%J)I0WeSn4BQd#Tcy6_Ghrw7Go z+CKP^R)mqdf#gLP4;4o%%V3uS^_f*Jd4wgfe0B2;>^;_4#soTygma+hn%%JMM*y7F zxhab=nkui#gaw-oU_X2nyCnVk1= z1_!DAemxI=u9d~V_JfZ`?83$z`g{_fSD(3dFXp+D;pf_r@`@)lH1HEwZ|H-e$j=Zc z&d~^dUFoAqqsNxY?B6Y9gL2L4hC-2i``hTTT4znQyWBRq_IKUn>?K zIf%nDzS~p6r-}d$S8-4Q$pp)5A}<7qpt3u>qF8Y3}h{3{9;$bxM8A-$ISLnRKDB;I5OG7Vt%pF-0HZ zMp{yX@NaP6Pc1mcu+SG=&vn$XK8Oi~Bq`2{m|0@zA08C4ksCck*pqQ#qlFOCHLK?d z`Z1;zEU6?!0NvRs(@;dsC(Qg?V-K_wZ@(iaa9a^V{T*aRGBsyNJ~sesxywlzR3e|if);I|shuB+xR21(ks3@w+CQ+1mUJBSw%D4`T|AS$KA%{0taC1~8`njH z{=KIkgVQ)Y1VZCiij>6`rd1z{W`haN>WJQBi|)`%Y!2DOZ6(#A=s@rBpS8u~yyk#C z%&ERT|5(%Jl~#hKZ{pM{7wdioeP%@$BZvhSZCEn(t3UWIdWbzW1en)yv%aujf4*bZ z833Gqz$tQ0K4^vPszj;36v2>5xzQrlR$ilt7FY1n+a~_*+aJZY!Jlmx;Tc5_M61?B zY~+~iPci`vf(M~$+*OW<5l=5rOg&%S0l4wNg#R!91OWRVcTbG5@O@O6wu)E-W~qwK za-3j3jnQ3%PSTBn2xI9o@t2^Wsu1c&j1~o_3EP^Kya?7D8g7^|9hm${k_MAuS4YZt z1bG}`3$d%R+`;@WN!O)Y4r%dbM_xPup#xWV+24Cs^$+m;a5y7bsxe|R4KxjNT85yG zt)IlrvkU;MLQO9)LU_qCA~_p4F`Zl7~)Tnq+p17yi$u!siM=hVoH^+Ns!q5sVaZRwLI%Z1GfEiA-wHvsa3Y zg8+e~X~9$=F^`2KcgQm_MS7%qzaZ?9#&9KxyCUMKL2hnUhZiSU#+q8Um zoOxcCr8n>syer3sy`wG4$yXX;!0=hd$23li?#cGHPTnk4)m=pT~b3gL7)MQRKRu?6*C_Gh%!aaKqmm? z_EIXzD7y$}s#f~nK4ldS_@QP88|R%x&*TRWmma=?d`crJPlf?K{*?m?3zjmifRcRN zoN3pO@%$BwvY(N4xS<&=gQVu`z%EB-N9bHY_@GWY)ADEbMnw*{+gF+B5)CRyVnyxg z4XPNaZg}ZmEo9$mJ6=)ITANq}O}-(70wB1$E}C8kln!zfEMeGukQh`Ak(MfqzuW2qwJc>a!|E|^OsmJz{cwu^> zJ|7hW4x}3%av!kbH%7Vmsg6lzeEdajT)q)&!{5ex80l6a&_Ax^CeP+J)s++x#`J9S z9|O}-dX_Mr0@tWkYg_^f+@blM98h30eE-ck`76tI+bVkDtRlr=7_&?$#le&|6WSt2 zS%Tj&`NaQpw?bSY82%T3)(K9>23ze%DX}1-*JH$%j2dC%rL-WY60%m2K2>+b!4=PK zKwfO=$10HsYwkZ z6o^ST`#(0r+f=&Nro@d(uOCA_Lm%zv*9U==yANI^*xp=-D3{Y}1qXjy(X9y`v;!i+ zwS*2xyf}5`jvB6CgSqiizlfsH3R_4k1E^hRsoMNr(ZwD=FVx#7i{r&@NqEtI>#R;1 zEZhvSh473P5~6}Vyo}B}Qe&qSj15O;={ZKsa8#MUk5Hk!2_HMHlt;bougDGi*H;c| zwal98zqkfy6oQtgZWxvbdV2V9P4|}g(or=ZnJE?m;L-Yfp8I-$(v2YnjFxY+4!cV84p|J`xyy{5Lz6ICkZsGhZQHhO+qP}nwq0M@wr$(C`%XII4SMi~ z`3ExM?6Y&Nj^J(Ms=5dD#aryhGwxs}=zJ~1w}Mo~9%OOzD4E?{U%>Z#`NH~!fCqMB z-y>oIEotd!?%Ji4S3LJ`0=gLCy@(gK*}0f-hvSB8}XC%_6=f6fVk z9$pWcm|cFGfo*CBUI)6^6CNUf%abRhwoL8e;K<1MGylY*8H0^yW^D;==h_MaD0@Q( z;4~323qT(Jf&~FIBX|d3&TpdFKn7?4u>^1d$hu0hx&}07HI_wj;XY{_xvGPMM=bd} zBC4;=rJ}1#imr{7vWX$AtLSH?5J){fkspP*en1&kVZO4R|ekJ++UcE2FEM1xWV) z*M?sJI@1RM1bA2{C+BZX@!K|W;XnD!?#ej6H4D?f+go<~=V?^-S~sir7tZ$Qy>*J< zZJy8+{nW<<1eiv^a8K`j)@KgZ_y?|xqMDSNq_Sq9he!A4@i#+G zRoD-|Z+Zgu@4*pBV=qf(aaRb($oy?|6Z4|p2dRKPSBF3@kLFKZe``~C`=;m5ejCZe z-ta@+6IvY{P1ZZ1m|8*Yu&;ZMpv2!;O#mH$8sHoPz$-($>6`XVzU@17<9E=W>D`N4 z00$tg%)q>WejMrjHR$-p<`N8~oudQDx93lO=ueNZ;Q@d~;FQij_Lq%D;?E?#r7Z;D z{TH;4@xy+}zXjliKUnIGtrQKx`yY8fkVWKh5s0IwHN@th|LHZqm1IZIfK-{Az~*1d zGyjjwjex=2-^m9)-e!NFj%6x=oLkxc-@m||QUN>xXw-!0j+efWA9T7aqtDf^qE|J8 z{jI`pV9Q@z@_X&Q-W}e8K6Uhf$k@~$ej9sQOFL_L7dJ2mhhJl$-idqskiF(VY4rXg zTKYN~s;PVbmuY^xq}Njf=!72BfZS+ogUe%+%gEu+RfuaH9RT~*w!BsVukR&F05mi2 zh$(G|{(+SdbOT67p-=BDdqJAN>o59m_`#XiFpkcwpg-_~W8i-R*MWa}O+XqrGJ9bt zFwNiYIOJRYg>rw?xa4ElFn-xDe_*>24IIEb0k3~zyF|J*xdDH5{V!?*Be;)uB;5dL z^aV?O0e`f~lY7>O`H%)bZFYaI{=ZFX?v6q0KrMo3_Ws17f4UZX#oIm)`tGvi`29zRX>A;oke! z{O=F}5Y7Qv#I?Rb8POpcB){;~4#+iGUO?lb;y+5D#3jV(0Nog!t8x+KiTtB+D0Fbv zaS}vzVH6V%*q@~#o9R~yp$5SEYFbFt<@K~HqHMB1DL)a5B%z&h(h3RzpreM(;HQh|dXh zGA;w$JbnA4qS4r2Amw`4h0TGHt=Pv_jeyI-6bu(kz-b^LeYZ; z2x?rYQKPo^|7qtEOt^NT@B(2bY`R@n?P!0Az3uANKmDvf%a`Jd91*d?0<_hk7Hns{ z;(+4hFSQ5A1u&HZLqhQsK_3;w6gIp(VHM1Ntx+=VsqtIZ*|y>w+e+Fk4vVuD7Up(Fg%Lzj(KI)&Yy^kynfJfsuEQ9f(_oN{F9WG)k)O zP0HYSb5r_+>NRU5cze4+F4>+H9{0`&6f$A~G@Dnj~<EY)tGQr6k^GSch$FnYQ_`gIjp8ZH1kf$fpISO{d}jwmV4)s@8v97B^UJzJ%Ab?Y0C04a^37?+TWNdxAw z3SBY~EPiLrW71fdXeVZ6@k(R2k9k7vgxH&X|9vo<O#5uLJ(*DVXX(Emv^aP!c*U@ycbp8*6u`#x zb=wMy{5LmD>|2vueYNm>j;;!$dSQ?cIe*GaKJvR#`P8a@REa)lzk4}AVEaybt_o}TKa~=Vx9^(fUz?d!!s?ib*~|D$1_UOL zF~+-H#XTN)*hR7w$-8GlLkhwKksje^VQ3+}8% z5pSlb*}2JI{0e(x^^KBQ!H6nMV~*95@@Hk!TEVBj8Z%&{r!cx;2KO&6H>~yo(+u`m ztTgZ>WTzC-|CqYrk;aC0##k#lEA0z*iBcLwA4nKnsjc?NSIP?r&@qh z)wTnj3=n|EeM=F`xcyQXz-vOWa>s-)r-wF0J1o)9RW2<25q{Ae$O){-9xr4OP*wYu z^t$-aIImgQ3BI&2)*9#O=N_D`Q_E9Y?sCN|H9Q36Fk&ipFl!JcerckUS?4bZ)vK_b0Jp zQ1+z`NEkn4OVY<#cPZzr-=426Tgd*HUuzD!1?TB1Pij#Fn|fEa8pRW)9qPQP4a_8$ zS>_%A6)}*{JhsREW{UIV2$oP~JBnvcf`bM5@*{p!e%|Ib zd_G2XCr#oVi_PHlz0oh-lMs(bxPsItFKvxoM?e3W%IoYHTw2!qkyD zoZ!uIl#MQn1QUl)C9(rK&>3{To$^fWo{bWE1g@*}#cm#bTE3d~M@HzmS;0-NV%M(>wj$ zyZjy>3s8>Vyp&IXcxBsGfMaI23JW=QzS>-@Rsq4I<2D;)s-8UYrZ)4JuPk~Z6#fZO zXq-274pgvzLgu*3`i;wH%rrZq(m*G(@?}ek>-g?2mfk+mjiG1{SqZ-%DAYhNkNM8q z?5j=(y>J9075TMUhbL>S;DBP8$L7@>%_L^m{?zSq&lK}B7kiHHrGISBd)1YjYQ|oZ zWcU`l>=F{{Fb&E~F*TA6Um2}$rpyBMiD{! z30|^7vmX#+903uHJyx1AdUTW$HGKg_?)N>5c`$;XJtiyv$bHgcqOoca^+jwGhALo$ zQgJuh(C}`3C0TeS?O}58u>A2-c4{?&=-wHRp$g?>%Kt97^S1HJV;86LM1JLFa@gGg zX3hH0*KcyXwWINQLMmC4=_Xg_Mlkc}OO!knzldBSRMRxA9a0EvG^SyRCYntBBYLf% z^g#E*#!8>{;bd+fv}PayCH^|)22lr$Q(--lhXIk{Zo-Qkkl<&#&JS>|d$*HyvQXxT zaxJ!WG-a*YuJqK{a5dDvWzqJ{JYK4iRGg9d-j|?9t!~6e(>->d$ww4!JjJ|u3rq(O z?a0W9F?IFvy8oF6w>~1NDgB}&U_hnRHkmujbll0!o>D-}SqOy2fn}4u$Ztfk;I-aH z>PwBjcYZc~?GTenxg7Bt+?@c+{#UkRGeyr6Unq@JsbuKTig&*XvV=eL4!jf!3kI3zHV`tb;reQ~rr<{bTYoLSS!Jv|+=*4~{5eZWiYK<#2pVC#^`z zI2=-0xLat7igr--{a`dF53ven0jv>NYzfT85B0tKrsx@)!THh#XT&B%OgFvPdyBkA+#<`hz% znJONud}{sqj8T5L28@1DY|Tr?cy9%ZPL>w6f3zy1aReM!;ZzUBunx~Tv zw!tipV^_RUXg0(sT>PntU{TT3oH|O|aiH$U^>dP`zL&QLj*W%4^!U6bb6c()jsfh6 z@_7<7_3dcbd`C9^+2kgZ<$0k>R&`iG;^@6rL2}>i&7!48?Uu5yvIpao)iM?1BT{K9 zS1Suu#2@tYmCfL0YkMFKUqcILriw_a4dkjvU1<3RoZ3JLYukt_x5l|l;Z9m=XOs0+ zKYIW3roq7c1b{qlVIq<6`tNC<(-%`!Su3D;2sARxe^17*$Z{{n{2FQPgv?9^&?`{l z(b`N)PYiZ$<4a~lDhjc#jJ9)Y%<|eegbdOrb1IJp$*RUl;j&L)!-)m_p-%Q;2=1lEr@&F781| zZNuAX@M5sq##JYUHSMc*bRg}4n5=8iZ1he1{%D#g1bhotf80jOtk8UlhCt;!6&;{b zU3sGPg20%P`jE?XvltWOFFfh6zy<)T)wujUggqPAv$qD}4QF|A>RT#CX$tT2UMsS& z$&9>!NSEe0pF0=A<8)5sLHiCst^fGNTe9%4@$;}(8ewky|H6yJHPCq++MT-4jL63urBwa+{jeUZ zSzGa0{|=IyQ;Ea0?<+9f-RO+(S$C)1+vA-{8F`s_vB?J@l;t|x0op|N8b8f~gL{63 zyq65)hgbD?ABH**=BSQ5iKzr}NH-E)qDP)cPBDU)#J{@myfMBJY|Sbt%F$pGK{OQx zPEcGXz8yZSd|1eL#a}uOK5^<~YTK94_PMssRe59vIi7(K3zro4G`vEe6nFA-6b;9% zZFI}qsathO22n036Dg!c3EDu3P^XcQq4S2D)uGm&g4|d;aFmf@h4vTGXYVoevu`sK zrxm78E6P5Ft4!hc=f!C)1bLy_cMd}ifu(=VXNS>den_w-bC%B-^^9COpHAUfRnfP^ z7xX&#KjX~*=|@rch1w13U?oQN&>c+wx1BL!F{y1@I~7<{O`Mnu-ZIiDC{Y9|LG!GY z-UA7vvUmyz0%z7o+pRhcYEHG*wRTIRrDM`zEbCDX5oEEl48|@z*7p-G?W(nZG1kPw z&s+l=SD_Ji@j!^P8RpTTzuO7~GbK9)^pVH*|xoimBLO*toR@pN!sTY}z_yn7PQ8WDeL{Tn|Cn z?ewO)^*7+jW_@jov@z6PBBfk}5awL0My_!lp}i4#IB%Qzu5x2#P^^qkgSf@CEkVwE z`)%v=>_W?rs#NK9^vR961V0M@t60Z!Bs+4}Sn#OJlxff^ z$B+~G0cP-Yo>Vw%IY#q(yTs-Q2Hw1ooI>pqF%#B3gTEEYD+ct=xHiKZkdt0tFb6cQ zPS`hO$hCTBrm1ZGX|$UR6d<}H4f;Y5n*jEq*C&*;$BD7Bwg(jkUO1VC03CQLP4Y(S z5+;l!mN?;lyRqYjxD{BGS_$nP zPd;++wGmicU!|;a!{_E6ZT(YpkFtr?g6`sy8A_C%NP46q`c}SfVE+!cGNi~f+B$bR z+zH1#4FZmFAWueDDwX`q3j`KjY99)&Xmg?X^JPnZ~BWRuXtOMlwdT104Sv4vCe%1P-Tr2*2V?k7W_T|EK-`YEudt1N2 zF_gPN#KTe61nnQq(S|A4hmNC5U3`?o$V9pnd=VRx#mp0ZDHw;{`d_GH6|>G8n(HGIcOG<@t$(>-`uUau6rVY+fPl7K9Bs1YQV^d1t0MQwuM2L9 zx$QC5OYp=)tKQu&wX)t#fU<4zc$d>=k+5*}mbK)A9iU%-`rI3{!^reka_As53N5I)vtV^f$z*6mSpnO4-PV#&B2=S71Krd86E!rv$}YCiEui*~lt zPIH~n$_vqI%?qziUHU3zs^1$PC=pb5IA3Mv;drJU-=hpeHDs%`v_YC7ox;ez#p-Sm zUct*9lq0KnCi)KpnSY)m6Z|`~@GG;!3lo3#!!yL&fWrJ}$Q&HlXLEtj^KUpc)QC_H z9EBQ>z*?RS>#&tGjR`Ly1)+ zt?BHmI`WL$Lwxc;@!^Oz|I&P)b>53iBz&u>hXqJ#JO8xYUw;yGEry3j>nODx!)H|K z)OVxXV{%vGG@2Kpd74M2?u)Yj%F@OxxptIPM^%ik#-qDa+1Ey)mAct}_b@EWu7p#j}&?3_cE7*JOvQmT2w*H8oD+ znu`y47V-CUb{s1EQTx?+TB ztRLd)APbEnqpOrZhfLytJe%w$^2N8Dv?}O?rt9kwEv#pJUA%TkJ|J$3af@o5+LrwZ z=$}q1is+!`X1iopBQsZ122Tl>Ee5&_*QTFEe9Au5%c>NTogEb|lB8D*5cGI(ZKY^T97;qPxDiAs zkcLdm%5hte>|W;8v`f6jCWnMH^y<1{V>);-s095@PG0Cfu2sauPgoEi-!u&NR!YD={uX}mWvdw>iQh4%pA;G zd5%M0kqW88{*2=l#cYOfx0HVQg~Zy$EP=#JH%3lOABDBL)ghL$U7)eNJ3&Rs0_5SI zY6a8L+X?eKOs_yQL_2npJ=1;3o~iE9Tt;L)sCnhwG{db0b-mVGr-sZFTz!D?A+b`geLL6t18jarbe2DAese;p)a~p(9eGmh6wC_O z>EPv0ghuB_gn<}_ZgqR&UF)X(!RKo6X2$UV|I%;GqWHG>%+c@+StK^MK{80BM|-m# zDm$9nGr#Em-D2`zE@<^_FBJ%gnyVnogQ!xCeS#5e!3HQKy}%91REMihf2p4{NtI2l zz9`P24ud@*BOT)RXM|SRA|Y-uHj(yp;;kMNf$XKKqSwtY5Cm9?PTwS)L;YZS1Iyx= z9nfC2PK*P|P+{YxxZjM~?W%yu5hB=x+vVon{8Lmd=v=GRGItw~fIx*E^p-Tf~7}jkTxpwfOZqeIH74(`{(k;%nynE@TqVZU{ zUYch9#S2Js#2bt5r6DN zWN(zn0`RkW8pjPMpX#iS5kyE%|6GX z{1&c!v(Qzr5PuEdpF1}-@D;wHu0_tex4Apl8{X(x+8S5Fd}x&+0yBUiI=%&M?yS@x zAF0Wyo1%`#owbM-ic`$yGC%w=3G(p#OgUN{;Q241i6T6C*;a07^1^I(F zGSsdNZ^w#Z z=Y!sj5st4kg($6@f{IO00e$iKF856?3pJdsOhnR&BReakiu>KAAR}z3svKaOvXG~B zuC%S3ho1#BS&9amUh?kz-s)$=Z7_bS2_-8O;BLw8jr*Lvv_3nJ15eu`xckhavb#41j58nQk)z<;EzL8{;|DmTv}!9X-wfK z;!P&FO=?TEU)jX0_fJ1#1^@O=C%Yy5(qp;E&K!#aA$F7%nVM~9_~n=dVnXc7Mh~wl zSYNsMqO^m*Rx*pG8V`?@0e`3m#*dg?{ty_;JDX6%xC4B8E2!21gHWFi09*(PSoB8k z#>yP=WuUKNqcO^|Flye3rIw->DO4KNNUVaqh zBcq~5zTM4B2%^X2Fj1d+N*ge#Nnwx2(nEINdB#lz~h7cDeqx!&Dr>05AyloT?`}nN&=g8Wad&hbkGl?3RN0XyH&)S=)tf9 zaWq`K0agD$OBV{T_j%G8i-K_~g)3_fNQo=`D&PHPF6c@vp8<8Wg#lc)k%9i<3hL}< zi_Fl$J5D}8aMHGE->sH+l^TJJ)f+JQQ-I^ zL8-ia5KK-Aan5sWka8jcwz>xB9`MTCLvQ-Obv_My7Pz327svb~n&Dudtvmc)#TBJq&?=lom@04|Y1hz06x!`dq6= zk_C(_RKDq|iQ>{&7Mc5~_X{(QdD%OXtn`?Ucn8nLmg%@UznD$a){@uZ6}ZT!@=vt| z=Xe9};;483PG$Nq>7E3+>Nxsz^zeF5xx`i*jkQ48nyU6bY9AO#h&!oy z2iQ@nssRu^2|Z}ZXnnyb-C>7)OHS+e6l!NWkuAeS&Sb@g(=S2HGQN@yCJ*&`Hxyjg z4)v&kB$^sr(uWh30$d`fh?L)>W^w{Zxo;%;b^?xsCZRyaqP|A0Pao^h{#lcT_fh{v zeqaF^9SQcF=L%ViTW87=!RbgE(@-|G&B+xJpYNs&bbfqz*d;IzQf{-?-5>*8iw)1B zHqPb}84!_wm@twu4^)P3gYMXOs%}$?*G+wN7`ySjyeWQo)PSXL3`Nl+ap7c^_*ku&!@bbWEqZX{L=``pzr$fTRj+H}==2 z`;sc3lXd8XkG1V?emdp0v3qg8QJ|E%n>$I5S`e?JS^g7gJzD}Bn@hSTG}J7gocOA$ z5x!Z|Ba(e)Kr-tEMZx?q+yivUZ&47_htWxHL(4KLE?kYsd5MDyC9z$_V2{cQc5f*E zat}tL$xh#jKsPNpSAp&IipBbgq2}Jn4OAwdKnBw>^vid{9oGLo(J1ld)nWO3!r!^XXQ?ja;tIkbS+;m{EpSgRWGHh-0jN zBBaiKIy0LhN+jcDW>Ks)YHP(sfue*)XvS9JFF6fKU`ouQ`CWUL70C0@$J|%b%`LcM zay183qRiAQSzI?Z3XX83>g`(hyaz}G^-dKB%YV6qZj-TJM&Xk8ZK}QpiXSiZd)*yV zMVl%8KSrY2H;lt7gn{*ElgGj*Khx@c*%||WE>xm7LY|5OX8K-w9q9*E0oF&PRccM5qV|MSv?wvASQ}&ix~6jPErC~DC7-`U zh&B51c-68}tMn-e@(vb2G)7b)WHss|sag1DLrJ+-S8QfZ-@cU}H8Xa=j zkf9jaiSuL~uMYLI=8xKXvkm${<%`;Q6o)$(vEq=9TlvnOt?!!pEo2ih_OQ zu^N%d2kknO>e9M;(Qk{?Z=s}=C$FfmHqZPpkqk0`dO4_s+_lqX;Dsr1GcUT3C?2bw zquCc0EBH4A=EV0=GgJm_kK`greQyxw6RN)rxR*{j%B~u2%WqWlQSjJ})S6-9V3wz> z*s?D`z?acU+bh)rh@s5ru#f#`y#Vvt3*em}1WSZMx$;ic3TV{y)AT?a(<9Qxiyxy@ zhedc2%_lGRR$T3(yQ{1A>OE36)VfjtL>*4Kr6=6`s^Hb>k&{Ms3E*8}7Uq>Mu(1d( ztBj!aO`ozQcwZf4NetrSfA)%0&pKl5&oWC=QV|Ylhl{Op$Z8=!k411CcgHd}!vDxE zKV2#9a?LDMZV#h^3fg75)4dyWnep}!>_>k?l9dv}9Kq35c)7K3W(v_;y7q_36=b6L zwNDy*sB7`I+E`BOVkR#s?!Xu7rzM%u)qA<~6$Q*NDx_?H5+CI2NZP&1>7rn`u8{KG zL$4&(FgzVQnOSmpej$mtc&#PKFj^q)C`ElpsCjOFB|EAIJ6QZg`q)`kct2OBI;g*w z=2A{7bnyJwC7#aeM7P<-)@>ra7g884C+#IpGNNaIP(@%!MqKaY9JWK)>~@*ea6^Mx z9nuH!ZP2-E*rejf{W6x}NA`7gvA>kJN>!5Gq?yq6?@u0$V0A&x!VKpnawiX4!%29n z%uDg1uFT=>^-k_zyTeGAMt0JDcvFJe(K|#Q3mmPtvQ=T?)4x6>-{(fw=v<&Z&|rUj z9|D1un$&N#HW~~Zvf_l-Y>>V?;HR(N|GoVk^_Y)1Wtsx1{l^Y3m=S5TzK!-N+!4Uz zWlhEVS&(luw<-TlZZ~uFc1G7K;^wYfsb?RU#;<(H_LeK4m_r&LZKX^<(@2#X40;xo z;GFZt)UTPCA0!L8u^3Mh_WMv@-ebbJ2w%?R$o11jvUI{@d7|-#p1T}WU^cy0rfAnjdEr?eNd$JKPiSMN^f8IZtr}7NXA*y ziLy7LpMR1#NF>D;M;E|@#r;hP;7$zB_~y9aKmK7B1-2DE91b&VRv znzVv+0g>MX(G4pE6k+~CfM-*HEpsM}&C}%TDE@^U2+V9A%q5hqN)h_UL7Dw6?C7a3Dm_VUIQs*JK2OvkB z9=|xXza^Z%)^mLj@623-1n%PZPUCq(Le#&^ZH;AX;}u3*OWnSjBvAN$n5NAHNkB*W z2l|rKgf4=>w>Oz1|n+qGHfiCrs^_!nzn(Mf2cVtNDMNN zG$h7rz%(L`Iy?CBaLFeXWg-J#z-e4u4tFfx)?&kM}kege@G*UCFj4j!{*a@_4smV;I$fl1aMJ6*oMS z=D+i`Z;Sq{cE^69)063e+)=lsB2v@ZWztU528!L}41&&?FFTi=Ko&3!hO=*}T+Kv@TP$e>Ll)%g=>AmfH*oSIQuGj>9EW!ePDRx9~@={Udvj z-GEqKyR2!*NO34vV26!bA~(Ak5XIzl{xbK=5$#C*)>>^AnAae1G#@P#7Wrd>__(CY zXwq{`;6{U~|Ei-pSpbxim4Jw&zQ&vHkE1snp|vSldBZr?d1FyKalQ(rRvcq4hxzYG zKGu-I`KW(>2M>U)n#um#l~4FB?8Xr|0&q3^pvti5w1uQUm(PJ%J`X?BS49|prjX`| z)X0!8!ooCj7kGO&c(98*S8N?^Uk=MdeRdUO+M(Op<-YJy`Y2;|fIUORo|h>C{Z21;5CcyckpV3LK+hY*=g!^BP72uew!E(v~LMS$nFru!KOKwo<8! zNqChJ5y}w&b9Q2+7KXDxp5*52!c)hs27IM&MkZ8#h8@YV;{;;+Ccy?8YoK83DVJ#q zySD(buU)quL#Pvy4UkoaFKom)J&|1d{NVz&p)?Sz-%lWR4iJItIBJPFqB}A#6Zt_z zSj>x#8DQHu3l^OW2{>6-v}6xVx7zr`yl_zTtuvr2SS5N@Ncw*Q=MolF!?%;fI&|Bv zX;OVPd(AGaraercMSJ{u{QJ0{PZ$1 zLH;6pVDiuWL}A*NM*|-YI~^^BY0A-hn83kruG59;2iMW*bKGIPCX>|pR5T)>Z@ubJB-avfc{x>$wx`AGBU)_mmWFT;r) zs3_z#ESlW@HH5>!zWYTXI*CE60SlKGSTjHGS0T2zUv6B3rXyw{_|V(rXwoGy#lF!( zO#Q1V!H^SLWQ27T(2*1RcjsW_FtrA+Pnz8E>Edl175w~rj`_`e&KG>sad@Z*yzhod zs!P4G>_V=QCEQ}+@)6fzfN2cd=Da!1QgEP&wqJ>{xKb#Ep-YTR!zZ)AHj|4KL{c9r zifoFKoZ08I>Gz+`v*4Y$GZ8n0X3uHdZ;n|h|8#V5z5aA|Y<8udNU|idF_A=UM?yuH zEBU8W2PKwI0`UgtdLI7+tIMXrlm&495GY4$gpj{{^WzwGo$Td#iNoV}xkxi}OArx= zfwPk|k^}VdrrA16M~z1zJp@Zi;E#|M8{Awyr@d3LcPof1tEVSU<3kjKW48ua+X-xp zpC%pxm+_M288cJ;RNl6YKl(&?7ciqOdn_Upe3)q;4#C#B><;WH$qSs!ddK$2f0pU` zzRxR-8z-cglJWS@*@aCU--j>5B4X7KuN`<*XZw#Rf_@IWI*X{~?>Zr?xI8z4*D>RBuDAECb+V*bu zbU}0c9m88d16cXyenk}AJ;Y7@0=rd}5-A!8-j@P4?^N!6eUD6>o|2u*u*`LGFT~l^W0EQ^3YV&5r53X%{`I+WF$)#W`+*?QScz zVpQ%KRBHU7!Ts8U2@83)C>oD?NV_w;a^0&*PDI#>vDs7~o^~>s{KvR2NGphmMoQ95 zdm}yijPY_NW1<#1+mxrqBi!8!L*5!G?U5a~H9iq=iG|y+=sC9nTe!AWqrLj%P~``= zT?WLRo}-;HMy(yW!z`7237+_RbWofshlhpDEo$lon9HiXPp7mR_d0n;0o{h%vv^kaEI>mM3>q-R(G6d4CWP3Nfe}m>&sRh2X<54ldZND6OKG z8r{aoKWf?@DD;|wpi(3~G|>5_2o}uhiFz&aJ-RbHf0&QN=gUB8C~3*I#Eft`qaRjD za{|0XwB<%)Ca3Ic-B zJIh3!62ubVbkE)SNOU(C?i5*%PxjUJ^uJUxo@nyJ$efn97yiukwYM3=S#NSU4Xet> z#!^cQjW9?=6Vm3F;>ZYIf#Kcl}q<`E6Wf^4HR4*SL zFu1WELK^sQ>BpAbO!57}b4mKLYY^XJaUKDiX>HZvYk-oXC})FEIB zJYBwjJoP$qr(ZDt`BG3YgM;Ore|nG{xYqZ;>vnX-&zy#cg1S8pr<&NMS zX?=R_#=aK(F?Wj5AGVY`abDh<}R9AjfmQ(5N7QMbJmPC(( zU~6%6Y>XEyZ9{M8`QDE?E!Gl^sO5@<_+X;BG8FDl!a{>h>ONoinDBjA)pCiLR;{xT zg;G7P;~@KXX)1AQ71y87$Ox0!DqYE_k4*17^gcEwdM6!>ZMz@usY0m!)yuATuIt9~ zHjZ(A1^*sRy=u%Mj2;-{Z#4;LA^w4vW3oqTMXCB}%p1KUvT#y@2QBA=92hm0zyC)x zbuzIl;4^^@yoOqYM%p12_;^_R&7{23>Jr*i;cM)^0BNJ4gJ*g`;}v9{tYN}18fj*I z48Q}aE%keRB17~!WaC}0gg?C}IPe|3{C1iM<${a(H?E?tXja6dr9k=n#`&(nz74P^G7_+(gv200vM7)lSwsg z32vVXH=6+I5&8KMR~S{q4nxub4CRB+(4x*Z;%TpWq_J-8f(~k!V)gi_`X`pGOky)& z)PC_P`KQ8ajNo3eRlxs<&K%~*KJBA!y2PWxR+W+>Zb1eE6eOpQv4dPrVKtMaKU-8v zvNk;jkE>vQgQJr(T9?;`s_4{l90?f|vaCZG5#fLgg<8uNrm7r~9Kb!TkdRPot=`s- zi31uoqIiRFmqJBbBAgn*PNoJw<~*aVe+W#fjnc7G%B*cC9Ttr4N%z#>rb)MfQ*K}b_r`~gnH|JhNYzcpR5P?{o|oN1%G!7U4bv1<*Tg62A$&@L{%)|eh;zgi_eP< zfBZ>zZB>1Br^}7Z42@fTa%xsyO-pTH1yJdf{`?SVd4O|9>@Iyt6-0iL#p?r ze0`IYoSzw0gBXW)!ivd8l*f>iW-M)P{4k*58~BXDCOS0Af_}zX4}<<_4FAbsp_%bp zUq-^HylM#^z88ndWk6a_zP%?&!=EYnQ~@368OCT4L(E>vy0LsVtA_Pa$kAm30^ekE z)Ks0>!#>GZIg-4OSn|KpSwv0aGcktnXNb!$FaJ=xhDXrmqxly?u#u-~5sa=qI`K)% zh>;}3A#H_lZi(>mD{$^fUBHU0Pm98UUHt&E{BMc<>8MJQfq#@7q*}EBu$08m@840f zcjP)!HLf_8;xMp=`L{K=Bj-#HgzCXs{Jowz8I8R6LA;v^NsAp&$y}X8+2LlY9q@`p zSmmqwrcvW8cQG7seVIxfO1NX717yn(OW94pNY9t@s>eW`=ybW8ZYh7L*;q`G)FHI$ zQ1N>hhWZPtLHRDrY(;_#;8bRnC~$^83sTibmdrwPbh%(VJ3!4wtx0 z9Yw!qOi!ok1=kH0;?{SUn>Kcs6|CtC=p*^GgI{(2>4)ZNKYPQx5Y8k-*+7+ulD@gO z2p@#arnu~E_z9&#p>-?8J{D1$^#Hs0(!obc?7r8Hk{FI_euBDF5RxEL7X5E=c1rP5 zZ{q@Lh#Qm)1qSJE)6t+O80`KrrMJ6o&CnCgzLX=_hLh6mp^&NPe{Tk-+U1;5(uKC$ z$YuvY74YdasFtB9L$iE9w0bd^M^h^M zLU$+ST$ae~jFBzgr;)MeXr^qmR60$=ALM3-MlsFRwDsx?++x4ypr zG&_x{RKQYYF3ijG5{*d;k@_$mYFha)JDb3e6Pu5UD~nYZ)>4SmA-0Mu8XpHqWgyQh ztD~E)##2C%Xftq>P3L zmZu+k;zi0nw+;#N1Z?Dq*o(&KN3F%Pyi!Z}e9IDCQr{7AMxN-qt}uJe2*Qkpa+=+Y z-0V^&?D5{09ln-Dt4+IP6@Rok#<^I<>DWHBu?Ej1@w)OR7U_>^%lyB=v3lvMX0P+` zLY2ou+E_x()Gls`>$8O(7r08D#h-OAwBT))Uz{N_n>?+GL{|= zXp!Bgp)}gvnsfdP5)h`bfhX(XFOMoKd{DV?tFLLa_qfTh&NV`ZO&h{rt{_&8L%Hf# zKd~23h~0%ssr-tKqQinwVLbtCwQf@(_}DPJ;ug{zz~mb(3ma8@Orm2#AQ9&de@GLq zFH6AYbZD?&&ru!!wk|1aw#6LZ#DzX#!&pAYZ|}>5uLV^0lq+>eiOj0!roF3hI0I<@ zhp}@A)1`^FblJAQvdvSrZQHhuQ?B~Uwr$(CZQDkl9^64c&mDA(c4Xu%2a&O3t#=We z{HgRgodkB0{*foB#t1^`e*UDXf+GyUdembT^~yfR|1V_1{$E4m(R?u+Vkn*o_C{CS zZ}CFyv0#n-Rg$1M;_%)RqFyQ{f0>Q8>1cChP`1YIkxv`T7asF_SG-$XApzO`Y=0eAOKz zm}ovqnK+qBmMI9YwXFdg^R(wR;NVgpMKMR;&L6L#)}8;YG{5}fqUn+5%fR~EmmT|x zYNNXhHp5F*OKNw%gS?qK_|)9*&+?_PkVS20d&V4m}Yk%lS4+ z;ni#)Snq5S&z1s7Im8a+%{}Afk3TRT%Hj1Y23n__eOMu?C52z^3d*P(eW;b4Oz>Pa zWrLZF5BA%}y|3tcko;{W4qV16?s(-|b8;=aKO|WX?2EZ%(lK!EdTIre6NmmU^)xXKW)$$BgsvD| zVv+4~uZnZEL4l=~yU4Joh3Z2jR8YS+7wo(RDWRE1$jSR8ZQ`aV+Iv{*!P{*Qlo0Fe zygvUVUsuva-w`eWs23gHca?-w$zb7(R_)OO0I`1XkeOZ5%O{V)e~+HwqdQiDpGbt8 zU86=1orkE4vm(YYl=`mtwbz4(i^~ARS{n4mXTCJeMkp#D@JlW4lN4sv7`m2yUB(i@ zXTIM!qV=Vr_y_B=XBk5booY`XeCdis=gXacKjAL)OOS}BJ>9B3*59=$aY%HBX5GC& zK8{?zDPgxwXt{vxCtvG`i9pcTVHVZm)4m)cfF=~LTobU=5+H8HugY&TLMelU=G#A` zz->&0K)ogV^9nPQ(IEvCryC7oXK0~l=MTuVN{O<#MAT%`f8N5y*gSn;!PPF2M$(LC zP7;7oZyj3?;kjwefmnga)V{y0X7==5J3Vl9lRM6~!!U{!Imk^W6y;(?UESwx*O%iCJ-%q!{>8jtZmE-NtX?esW-P z^@b86nc7e`!%`mwgS5Is(gQ9#UNEt{6n3jRM=uYrmIT)ZSr55c-|A_)jWH4TO`Ouu zAB1*l4h)z1X~{nbuaAUupQ?2ggb2rLY-DnQT~u7_4}~)QA=!!j5|H`DCBq?K_B;S@ zBwYFMuk?1%T`dE#2;}Kj^kx(T2|)OiMG%y4 z*>DLvhGMjbG5%}uDWE)X=Hp-84uy4AJz61iC%Z&k7pY{)ft)m%{f_lDD4x$jYb=9V zoaFHLF&lLqS^`P7s<0kR1vf+xe5k`9F^Ft+04Su8b^1b>0b%@`nZdA=?I=iY4+G%C zKKg-l%UWYUuyyt`#2wnRU03w%3Z*oH1E<8ekleHJ@8pB+j z*S=n$yaE>2R}B*P=pp4LWSeAsmJNy_5d1!znA7XE3%E!t$rDwhR@hRYMN&%WxR6sAT3D8E&6y=di?BCvTHPD0j0&$ znr|;a2-qIw=T2)UgHKm%NQSANb&<}lqC{aU*3^PDu_t+H(Q2ibz|8(kB-aSj4{Bh< zuUqMX@LZfB(Nb4W;Hu*}D|`;+yzJ2si<%V<-z7he%ySbft<9Ja+q`iSb2#|-lFeJ_ zHGKn*N^khxsO1z9Wa>M*9GMr3ey}|r9ojmuH4JR5henq4WmdPc1l4H;)hfmN zaHqQl=9jBaLm5rM`jv3GgoHU7&jM><+Ey zT;+~mL_)8g;1=xPa=Ly=MJnBLI>wo3X6h2+%Y~4%h`U9G=N9-CW56%$Iim&BNO#}; zMcvjE@#vC_8l8u;W4~fenj^Uz*-xwp|Lv5dbCgCsvSYi`OfgH5>$32-KnHqmHNy=I zbIuL`ISB0cc(J^OyaV-_aTdmW$>DW%YQmudq!^o%UzIG(+~l5%p`ydNQ(d!tp-^cJFNjqVU$-bRW8TO7 z@}*YzVWI{Ul&sG6%#Yjc(-?SypW*c9ytG!^^LkNgE%^kvB=pBbvKXjpgajq>mj-3Y zRW}H+SD11>Q_?2@rLS5Qb3-|6(j z7`H(;4i;Pr9AHd}k2VGT!b?$xBzL+55(YoiEMdpZCtG9UGO-RKguwg_j; zmD3scYP#eB4Z5WSkMF?s&0|Nr!FX%iW(R)wwVPj{w3Btb&?dVvWUt-Rr(yMIoLTu8 znOpuy^0&~`v}2IJzoUOR2`%On^i=`L+lPkQx&M^I%2G@^FRWn#Z4m1t$=6pRp4*}3 zlt7iE*G~Tuso?Bv2u=P=aW8Pd7Wvsb1}yh91FHri5xmY+Jwfo}hK?=|WrcvKcUfYg zkfp4MRpE7(b}#B1`x@FffPzRv9a99Y$4QI2g_HVy1BV0q#tOTyszQ`o>JpABzhS;h zl_X}@P}*NWHh`WIyQHW;*d&ZBROKSg%k>9S1-WZVYQ+7g5>|5i65x(XFnF$I$?snd ze3s&c_VNsgMQbz;dj)#`NA!z*;>Ly-!Z-QR|0MD8GLfIu?*c7qVGIu9Sk94r(1E`V zOJeflN5lbPs5cEHT^-thLxU~JCgxlfq$aaj<{EQH=PcVO45OET(W-`XPoCYFI!uM9-PUy1i@9Sil4Gdmz}2T{bn-)U633C9c0-3Q5jlKz zCWmZ~)hY#MdnBMkd%hgkYvZ)XP>jz+0g^}GN`PKj=yD`e)d|x%w?{bhPodEoCredn zeGR7sX&)br!He&kInnSD|4(3WYgEfPD@*K~4>{VR53_{69`d&f!%vg(2W`rOANG9U11uPMuHD{U)ratoRSjIM)|`~nP1+Y*1& z{o~GYzmSTTfgv?UdfHtLVIX*IZy-$@fqHUxQ%WQ2q)Bq*_Bx=!S|=MH@63u3I5Gn` zQyuFX;%TGygd#_A+*VFmQ;RKP;o*JeXMXkAbS(*>V*oAQ*^U*u{uIE!Uz)68OT0Rn z#ZD68uKxWAl&}4aG%UOAIEf$WTCWKoa+5hHEyGGRo_(X5c82NRcmN*kro8V&6~ake zvN0N~=7UXjOipNpc|_%GcueB6UOp0}k3F&RnOO>TV|YKtbFTea%1DYwn$9M8`zNKF z)6<_!BG$uO+0VUGvHo~%D>oA&QLLg| zogDblV5N56>aDJC5$!V>EH4<%9P>B4;Yo-lI@~S^fFojSP>;hs!*#E46UTbc0-e)&geFKUM{$)FB6Zl9D;lC7OuG-nA&*R=@iH1o% zEvN#w{E>_Moj2aHf6Q#WVZ4#<;Y)$Not}M%QgI}-TNR=twk6g59`?}B*85%r$VTnJ z72067`f|0oIN=;ZI^^u&J!&^CshgeIPY-Nqe|S!V=N+QGq2J4COhZ&Z16#V|L6^Y) z_Ej=y&3H0^vxlVDr-U5?1rp#MS4UuT*ntA&PHvSOpf3i|z7Stf#TxGSZr@1QZ4|~N z0#syVA4+|rq`0sbb)gOQh*coLV@+uI|4_}k?jqSx&!vw}!a(2~&%4wY*%o}bW}nkP;TbjzPD$G`;drm@>h5S*P$p&Lq-HW^eOgsC180YClR)N1YG(LW2~`P zY#f`4U4m>9m9ivvl`-yqIueW~X#29kN*{_Hn?7C%V`q+R38OtPeB9||Pq6Y2POJL_ znD#-T=Wt;+n%<5j*L4J14eo4*y2RaEEdJ5k5wGW}{|*(Wfc3L#5i3Ph6!h=&)wJT} zvU5%*-AF%5THpVPZ3Xr8HpJV0Z;qSHo)s|I^cQ;MtV}EwCRJ`xma4NfWY=^94EfCE z)w5D3N0C*&>K^xxiC6>f{od*7!v4BYrOS>GyA;R6M2-s>hSGUQRf%`WqQBCxjM|vb zoxH)d%9RxMkcUix36VfHz`_OW2AR_R<~Js(B;55K_(=N>UORc89&5IkJ`5Bag2X#J zh6a>R$}>2x*y>y>qj_m5%)#;lbh{Jqe}T?E@H+<>6{uClB{| ziHH2k5(Jn7d9?P1FFo1CZz`2>s)`clN$`0!!>+iXfZfR?;Iyg~vUEz(Y+rW|xkLC} z2MDk^evlA&XTYUWUFB0;m_k@yH26LnhnOH^}+zJ0`S>KKmH5r$JqhZX{{EMz9$)3tk>Tu^+>Pl&rllkFsBZ^Sh3(jC) z7z`BK9uSC8*w%=uH%99?D?nfTfOA8r*JA?ZF6(+Sb`^fE0M~z^$RS9xI{yilNktHq zlXYBR#VC6vE}~#l&9OZ^W&inJY?!PZkl|Uj7%z#9gjM?q9qE7KmGazEXKI@dnDF-z z$4iN5z?y7F);9=L45s|DTP30t)?(tip|mS5kaU&yC#Z$2 zckZ;CSM=P=W}d7YOTiP$9NcDRH|T6(gq)!fGMIG*+KE+$2NCn1@L?Aud1&rmh?x$A zxz!sbE>)YIH2T3#zq$I;UYHC_GQa}t4Gy{!e1q9%f-*X6QE}Dig7~clT(iY1I>^sJ z`(uYUrqDeC^;uy-TTGK60ycURUg4{NAfbJLKomB>$uLs77IP=|CRO0{J*e1nOT`Q z{|D~*kM+sK$jJ6T)BlI{$;`^i@&C>GG*!vcUSOATa9@Q8>ijpP9o%scZ*FdC`hz+l zp>J*y_T=VLvshR-S!Q2fwP`JC7Amb)YdXEF$w|~y@==-VIhfH~TR|6?=2@ITO^%bU zZQ-xeu+EDDLfKPBfXuwa#MCsHnUN`!h4YHLv&)F)l5qQgn}8^>KzDVsfu^FPb?Quj zpq06mOWlj1`^G>C&0$zoo*fuXK@@O)z|pR=16{;IXKt;Oxt*oCl`Gsak-Z@iDWsiY z_inLhU0ixezF#m(ASG~dv^P6BwY5Ruu4RB0o}8Y5JiS}p^T`4hb`T6qtPCxTA?i3l z%QCD$mQ;q4RDnm!vro70T067SD&i_q3b0TIZ*hQRfYg;K%qMp(f5UL!xmo%ER<|`~ zaxX3ZWIzg9K!FHbaBy~Zc26htGXh@T6wY3dnZ0l2S{huSTUXgtz_+z>CF~{sY$L-n zN8Q;1u#3H*Uwmi*RmJt?*^TG+6po+0$t}0lKJH$rpufUb-N=2rOtwGlkbg2We$;Vb zg-?5ZOwrLzfm=dT+ys=lg}D#ih4YB!hOrIqQm?*OmqHHggGzvqF)2!SIgfs-oj&9< z{FO0V{A*=aMiNPSR@u3%{$xY~?ga<8jRUl=Uk7`?6io0I)Jzy@X(hc*|smS!J6AeYBBXI3yk zTka6W*i*GZ61br&jb7H^={-2->>U$t**Fv{PKGL_rbrJ+WEMQLDZWdfgLs$XDfsAKu9R{Zp>w%Iv-^@Y^@ z(KGCQzWOcn2FMy)b1R#h1JNAnQ!l={eIeezvb@RS_7Ucj(~?$;KKh%TGO#dl8+)Tl zb2|gE;BxoPjgQZPrri4yrme35>B-zS+8SN_tPTNTon2oA-qHg>V&-x5P3=Ixyp}gK z0%L{vMg53f0YU8ZhlBPJK0rExV2S(^(gOvr`6EG+1PFd18bL5h{0QDMng0Icaya?V zdCR>263)C*M*b_&j`_pE`|=(j>p`$o{;k7Z{0i#aV#EF4Qp|sj+xqaAu*E&QzI*x; z`dfQo=uNKw2oML|XYhgeCO3YCXyW&unb|u9-lYB6hy15|#5?~Uw{_I7z{=~4w#Ofh zv>%r5&Z!5$T{qgLf1i5~yT9CCcE$89^hcrpsqrTyb2si63Hzcq-HU%i@8-tlcQDPO zw=IJI(rphy=}$<4^Sd zzwfX9Ucbo!SF=ICVXbv_-JO{Dm!AO(OwQf<<|(hp{`^Gjx4#c;Kc5lrtD%0ofZvf` zfI5(}vb(bJ+A+AdaSmbSQc0W>pnjkn28x;fBavFhPxjp7?l`&}fyZXP3m-)}koXs) zSLT7(TT{}GH`m0!6GgUx5H52yW~GQBk%L?+s56w$iO(b4NXWST;>ij)-AURHU6)ja zzsMQ3NlMe4dGlZs3fRh+U$l*@!(Ws|zAIm+IPOfXsrp#B4Skl0t&n>?uY-Z;(CvUu z*3}QF=UeE%80ITft&$uzN~nIk#88*BYq8BwBvslwJjwRdaSt$|7zFj%wsf%2b$yKO z7ca7Bi?uS6LcTP1al0@teQ)3h`aCz|lU#R;-k9lYh`uVOmI6bQ?Giy-{u8qM91BF5 zMv(Yhn9?ZoN{vlbR)P= zGACw?){=$tFwwixkCMfT6?6jT*Nr@N%5L-j#3Q#XXH{@Xc`>J2hjRrLf+vBBbE#^U zC4TnZe1iHDQG`Caf2flkSL{JsN@w4!pAP>j-^7-?!;8{Icr7mgD9D`n$S`CP2ijJAb*0>1NccSQ^|0;hhHZ%7X{QMFCIvMcx87Rjr zN3kmS8n33>f&1#HedY5Av!`X~p`$x1LE76(!u3;`e{Q;~(o!6kyJ~ej~6cOs!e@ zSnRsHL8*C~15z5LkC_@;!S|TKw0DqKxs5Vre3v}-#zv!ySKGnK`04o?y?bR z>UVDf6e2Imc@l+rHUF6}>tdm_5LR%#l@{3-Y@^P0#}T$gkZZlY2@xO7Vp{mJ1X4#cxk`k za++Zhjmg>+T)rnse|V&Ed1=WhCwW+2#0XNw1_^lQ{D}s+M(Q~TmAGI5??yv0%Dxy2 zj#}!m1A}K5L)p)C(xL4(803~&+hG$*{;mk2Jy-kB3g*?y_xzctqH#-S^xNul#pICg z=|b_xtRqS}u!`=w0Emqs<^E)cl|Av_VK=aooLll~=QV_haddh-cfK^$`u!UI6~!%i zUSqR~$0CA~KP&3>cH`gP#|Q(Hp*5^FnY?OfvR>ctoFL`PkVawuPIbG{+6KWr8{Caw zjPl*DJP5QBKnXRkHWH)88tA|Z1x0aSmnsRZ8TH}U9s{d=Pr_A)6c_m`V+f8VGoH`TN z+OyN+3e8g!{;dM%1>$tcks0d(;+9}?&)zw%Uzksy2v2C((Z%FQ$chh_m;7^m@BAmb zW7%QMX}8H9<6Hqw)T}EmN!a;wvZIrWKk%zU!TWNys5!9}Uu_H<{hbda9>7U6F2$BDG2u_VTotBzhyve704s2oUN${bIu^Z{t>K+v9pe>Z% z9PAPzUEz1Wrd;;y&@)zo?S-xt$`%EPx7JUZC+H?v1yv1u5Mt@(qFz=Whf+vEot+P( z3VN2f)!n(;cVHaj#)-)ZyVY74lxu>oK3(r}ZJ;;>RaTNZmd+_Jzf$cysAS$JH|A z;v1YUUG{(Z@NJK>M)A(zVK=$CO_6;4Sg&Aplnm6^10KZ7IDMvvl{cBW)Yiti%f> z93>u9M2#LRv^l=WU}3S$MRwYT{~q~_}xkBtYjbeEO|uHyO^ zyJce}C+YpT=%fHYZRH;DRLyemSJu?l(@>fJNhH_*R8*k6-A^ARkpBc6vR5~)o6qEw zxYS}JRg{adL1`u`s}i%_cl&Cu%c3vI`wUXZGn#-izy z&Co1M5>@&_`9DCCp(Llm!XFkNAFndpE}cbTGIWN5uL?6)P}@xj=aD6KyJ zCyQaMmz+C>5(1LmE5q~CBgzWI2{ZI%*3q)1`*dn`zt;uYlb(>z2!+)4qF3HEa0S*N z$W)O=pyaW}AXlPK^4KA;Y}LgR72T<#+4i;)XfS-p1NrJql~@&H=OX z+a+H7&J|L8+$#EY+1q=C!+TF1aVGMBLJCqi3yK*1V^+hrAaNZju;@PUKSJ1}`i>h4 zhdfoD{*?l&>brnc=r}fo_2wd$PDM?+0YL>XN0CbDMa%b?z!+IrsMz+OJs5GGlKboq z()^8qi6+KIHE%rudVI!byMeJA$BOx8L4J(EoNCuOL54A#7ypoe1f@GsyEjH9gV;-M zas~D0vGP>a;(R-cFP{0Z{dc@hp;1l>fi5mI3SaU^LuFuWkmMAKd|raEFbJEGrEJ;e zl|ly@GTgn4Dm0RUwTtcH~R3jlCt$;`9vQR(HC+Y4?^{fu9txcdaAsgOL$b zTKHI;(}Og*N0HgT{$-ennd@S}a9Czt%^d;DnvC4Y>ki#&wgNQx8MVFqf~#l3Ds63# z5LSeX$%O=PxxtC;VT9qtq;Bp1-1+cK0Au4lggxLI4bWA&@B|uvKeQRwl< zpu$@u#qS(UJpH(>IuFc{7a3KGS*gORokz9U&v!Jl8CQcO8R*`e=Bp?~b*)X6G!==O z@fcSs5=hG4ai?|Kr|=Ns&>W`wl7ovbon@G4x4JSW+3 zA6N@lohbS0zAY$?gb<5_eoE@UUigrzj**9Sl^e}nAuW?=?KOF2x9Jo*C1d_d=DiXH z#4^6;DMK*w0qAJ3&;{@m>Z<}Ch(Z|kvQ;eHQ->E4Uw(Zp8(v;9=m#_7nxgyyNxbwj zxMl*R$TtpsP(*MD{>i42rWrr2cZr-Unbf_rfT+Hq3_e2uX0A7l7e~2CF1}he`gh~J z{=2zE6}~$!u3V6uN#oq+GcsejF0|MkAD0bFsGsAGH7WvZe^7uDh*Z1NOfl;89T4`? z>&N$mbJj>EzNKeYQlqAk1XwwCnB=PLa|NlybM4;?$mxWLD35L;>RnoJkq(7x#5+z$ z@$h>wDR)&j4XH-bL=HeuiK$*H37fB2ZL)-P=lV&drM*&uv7ZXP zqqBSjzhA105cTMd^NKZ>fcwH-$Ou67!L(mTn|T95Xnzqy`Jv;JYp1TctFEr6%-299 zbAu51#in9{(3VOySXMzkgW0@XqaNU|j7 zsfW+rmKcXvJ6@>8>$QOiYe|Jmf)AhVv(9SZ&(~fRxGMz_u9AQA{5&S|U|E|C8OcV$ zVbCJxeuVIZQoMon)i`$QWuLsRJvk|ihJ-J;Mw{XThApjA%c~3dercq!c)@lB3}c=qHnt`8@iBL1 z`!69%Okbxi7|4sJPL@6ZVU32!fyV)R4T3w}es^U(Wzn;-|C+Xx{D>}tPx!xuUh%_z zP_#r&Hz9J^!6~Nb`Vw;9Dood2i;s1#D!V;K;H~WI)%B}2RJX>` z4bCr=2ue&jAal}TMRD-1sHn@BsF&h%7RnyayTwB!E*ja6zs*>J&734ZHJeNLinSts z_fL;No){PuIR}XLHG6Q_YrYDUsKkKM(J7V#BF7CAWfz; zvP83h{bIqwqgvMfH0;w(v%&gh%qGZ_w|8mAB&0?~+DS}{Mke{`&uJ6#WcnEn#jdu{ zyog8%xO%g;JdvIux=VSbs>3oo)7g?;1{P5?-noJaUD2)#gBt>oS}Zd{?!VM^WE}Ds zI^8~)czX!CF{?<)H-o{HU}JF&-!t!;h($o)hj+?()lu2sHAt$*8M&}BGv`dv4v%ie zmhQ1cuC5B>yOkZ)3OCWdw7zA^p)}#y`e7|2Lt?JaLbGjHq8aEJV^w70q<(T+6Jklt z6BLX$5|=lTTxFHAncM6sMmKLSl!b|mtsil;1){LUUY+xS+oZ&1twEJptBQ*Dm89|r z(QIef`(;%2D0wTk?TN#tO)Bw~2~QoB6f8V@kkcZ}WJOjHj&dm=<-?H7hlsb%T3c$D zlqS}&04XQ5P{?tbYw6)Z%_0zSl^7y8KtCFiyl{mlY13)QPqhJ~x3TGgbDD>7&60E0 zaC)%z+%`y$6_Ij(u_utX0PqV#T!bD~$^DTPD+Q~1*9qJAOkqxwCE1dqE`RDECs=$G z%G?33NLEaWkfx;ld z2$6sRYdVM){s;0ih`YiZcQ7(f>|o@w^dE9kDzPs#(apy=%0u7Hk*$!P5=^L|gt$<& z^WRC*hs!sSDWWCFTm24xh2!&F?YFX(YP9t|_mAUvZWtm*P7Xt1(A&RW5E0*~Af?L_ z_^cx+y~6EA-ZZ3!-Clxay)Zp~=P+LNW%$x?5QPQmeGN8N_FmQHl%Lc&n<#i%4KQ}r zXUYpB>5M9_)Hg!_R8HTV2-1?J?r`Y@Hs|);p6k9f879VLT_jCLK2gRywUiKSkP&TW z9LLTCL6d?Oy6hBe8=w=?;gA@9WG7stZb>GQ(sbh3b?}nB0y&uF$!;5_4p#qiBc5Sx zz6Wp+YGn$^l;^RLO5OyzI`Mq2mx=Vau5}Z7*M!(}kia>$>WQ?Ar6X<29c{>B4`W)C zD<21+^9&d~rmazpVboHGs+eQCUihR=VZ(0aK|1HN$5kwxcjMUpuC2e*t z-XOK;9x1d-D1{R&S#m~VNGM~arOGQzN#l;oRQ~c{}gKGb6j zL#Q%_`f(*Ya#*%Nzu;Zc^RkMD5;T5L0f%iEFc<+vkM=kK81CSHQo~N%tYmapU6ZTN z4KcD5Dv+6j)EkpF)KuMH7_fqE*XQ@VRdjNf1wK5wEC+Px=f8VY5N0VaEAB#%SW}F? zL2Cg7-bompM-^weyOd&M4e|coH|pI%kTMjpyEf1mLfJF`-7NnV5zA;%1sB*$=92ba zgMjF73-FKVFEqSde{u?NnzQEv_@4*?tGc7<1*tKJJJDiiA=8B@S+WWYm{BU&_D}|6 zLl!JIt=KYmW1)F-!TuK_cKT@5az9iFKxfUpwH; zpk8#x)`q%6nAT+RuBs}M}%i99Z zcnh$Yp>A=*OmBl#UQ?@O zEZPy_EG(1FaT$(5x&0ewE@{A^MNcC`D1XjwUqyYhm=8){yi_(yAAi69ISgjq@|S-% z`&MBv`&P_KEPZTeXMxIKKUSmUwn_h-o_grNIJ51i4a!DJWm8NvAgA^r$@ zmDPhe)?*^u9tsTc7_VJcqWZiZ{CEb{WuS-)J3>$S=&cDl^-P?vQWv^>5+uimVj3lp zQRtw2DoEkpZLf5px7$=TG1we@z5xpBF1{egt0#efR^J8S&1tfnx(4GR;hJW^Y$A5m z?b{ob9$9SHf#yC*DYF$7W!B3`uBKL#+TchiDZf;{Xe+uEF?l;JsHsm$hKcyBg9z=x zQ|d(*yrypPOn^6Etxve}=K<*E{$_jzq!h0q*(rn1EJlVoOc;&9(6fe2OdAd)8yeK7 zxqd+~@jxZ1Moi zrs5)9cIAK2rzJAFp+ILnxh`=;(g%4-Nznr*Y|G;@M!Lf`$?3J9Dhd{x6*fb^Tsnok zYduX%7s6+@J|^h7>CG$hD_6{mc{Bf(piI;x24@9ST9ir0xx1xo=}Fo zOi8`&N<*BPDl&IiX1i8Ddaeqg9ws^~ zMH*E_(AZO{qEWY43j5+sE7J8l2r;<+VUFnoha%tT8VF%>Bdaf}_)4-D)#$f?fGjBq zdl+vPKhO-H#4W1gXiYQ>{<q-I%odHjqcFY;$v;SYF_iM@hEF=8>0F}Dl|(RzVtjxl6hv~ z6a6f1pcn78F105VVPbHi-#;tF(0Bbg+CJ|wtM!y0gc~RR4WYK0s*>ujwg)526 zuAC23;}lU+?g_msET`b*Lh9qE5h-20-&Y5ARU1=+Z&SJf{@s4jrr&QJvuw2FYxXA@ z*YtEUN2$8fb;Ax;#tV3D@2j*^yV!|SFpXG1eMzn|9WXE;LK$J$woNmp({&2Ol`nQ^ zEV&OHbS8DcsisezbVj|m+S~&)7?hCBM$d0@Fz%pA?3i2hy&?`h>O58hXHl`@AKovg=f^PNY1lIv7&-$te^X6ibi(L4y-X!47A16CEKJ`V$<=XIo1^18#!AXH5KJKlFi*&# zM>vPt@OeZ00ds*YIA-}_yr-Z@S@=)Xvnrh4fb5{%$3{X$=f)fw%&I$w8^g_=h9lp+ z#p=b1^19lnoUsaO59_Gq`j1-wkks8g_+ArB8zQ6i0_lh1%b43t&7Drt8P4A}QtGlx z-OetBq%2g9#8OpxXW!(m=BqJ3_8kmsKqEfEXsKc*x_~ezt&w=&{y##+SY&`ewJ}$2 zp?#_Kxs$VeliM3x(t4Q3sbfqfIp#YJiq|1+MM~h2Ae|LcaM-MYV?3;UeOY`RE&4~>c_l33&$g;wYch23W`;m`%6b~sF~_=a2z zV}^3_TS?B`L6xCfLnn|T$E+@v^xq7Z9E*0(UgQR4J+DcyR#(w~U zGM-)M#05y$qEZ=aYB7x3vRd>&WAYM$8`E>gRau{jf=NF0#mwLzKoaYO^dgzl+s3r~#lSd;^e zhaIBjOKG`7h}X`P^cUt+gC2f|C80oB@^0T#sF4+YXl8KrxH z(9{r&gAG*7S}+!Xxxj=4G0U(7%M}wi1r{4eUjvfq10(-(0Py`deR9R;04#oGh^@v^ zmRQm$G>FWl5OR*?@_Ok08ZTB)i@W**9^ElV$|6c0SVX6%0=?@^>M)<~+>qJgJHz9f z(WmO4rm?atn|s3eA~y>RPy*C5j)R0&-8ZPFyjlK?0hQI#Kh7|#k)CY}A zdHV9ZjtrV$3HY*IR%Y2gJyt&jKICIb@0ZIAu+@{iQU@Z2=v9{Ry6eEVGp{)Nwxhvc zJ;ATj`8xz)HU&=W6fK3vqws;qFpP+ z%{C93?K`#fN#ab`>f%bICEBYO=B0#J>n)fssXIui+1rM6J#@;eqM z=7OlBrkrNSox&+UnZV>Z9&$|DwkYK~Nq@34^D_DFmv%eluy)9>wsyZySCm$hAFcO- z;~;@hSz~EoyWOb@Yt*{f)#Py~Cg`nb^&Npg=Ytwu5xFkK41pqy)0s(GTAO`r7+*7$T*hSBr zwRbsc`spcf>14#R;Koa@hbFG?^BkSXj||8YIR*cA+}qo=R`dIbI6cwF>gVqbd=4?W z?+uiF8C-qzfMod?j8G1ZqSgk9*pdXI&{Cd9W(kBd2Vd%iVKH2uZ0-K?hGFF0v}Z!O z{g=?=vV@;2lO;F@s52QJNu5eH8=5MJ?bbZ|&JlsSZYvgVX}V5P{y+FaY}EpJt9%me zG>7*wig;|*l=BGvf@}bsV4i#sl{jz4R|)QJ_aE7sF_#ISmdUC7EvWdZ2HeO;yK7-d zkrn3ZPrAh*u&7wZ0b9>qin&U>x_t<99^7(ED&;O{9saNG{I%e8=Z$*K=yF z94Ak>t=)>C6oX>1L+44}Xb&yh2OoZ1&F=i&#^2cugIAepFm3$;%u$|>VeEK`M(T0fEyPjl`**USIrbF*J1=`jjdrx8S&Jd5OCn zMk8Iy%xCF9Vv!*0IF9k09K*gsgqypOhN$TF>1lH!{tc0~A+chcj3R>wMUs9IE z>ZMSktt$wE!(~s8K*ErwJ!+GJsF7y;eGC1k06^70t{9O$(2A ze>S0N$~T?-fb!H6t}S|Y`mZWO|8d}FP-xapueUO@0!NiIh$~7mKQvcibv00T1AC3WbQJ3Xw&n4=V=_eSQh59RK0W z57N1iIFeIDcbO1UEBeFhXd@Ge4_mj5ZY#S+=no8u-{OeF&@M9);IxFZXR5ZA$twF!=fkdp|qN zFtTy+rSRV8kWhoziv|^J)c7sAAJlIGwOf!kY3LEMII3;_U2=w~`;luil4-;z zF#|6_0p>(s6)nEZDGQ2QCdmp{7`9-K)2aaR00iDaYiE>ZVMMrN9S z;!!ynR_SEvyf91mSIOOTFX1QEQ~nU9^Na~up`0mr3Z7>%Gqblb`)Q0TS$(=AG=({Y z6Q&WRc8TY4rpqlvl01u530fY1EKhVi@g4o4BeS3DF7g!%%~6){3vRCT3_^`y8r+=G;5XE(`&5Ns zA{>H7&VTO6py#!1JS?*m>b_$NY*7?hv~_B3Lnd6N*X$qUGwNO*da=_J23NZLHQ54; z_9Z4$%_azIURhl==1eTC2?p5sq$V-V9#N2;v>rQv`To^fJV54wEQB1p3UJ+A(!em8 z^>e^)z6xhtoF8m`|ES)QMRJj4H-O+O&6H(|0aZ|Eqw%zR5>r`QZD*Rf&`JWi@i^XM9Rfc+OHSeYrisQZVDpgwJwt z74F;!p~5fDA~|_wpXP1*KL)uwZbvSJl?zKalm7rh%J(w&Y#XV3iTBw`N>WOjHc7Vu^bb$U2oYf38tS*#@Gir#N7|2+{Yokm&P5^ zysPa>4gG$ICr3UCtee{`1nM^5ECwJe2EV02w_Q~v?UZ9C8#WZrkid!8rIK#z{yW&j z7&a^Wez0jBIA+_{-j)VJk~M}WcI4VXOWz1_P;MMsA5$)sT+X}W3TOWqwkSk;^3w=9 z%}qb5N{R-2#A7J_#FIw-+8Z|U5DqmXG%=Y{bRtOcHG1mm5TdFu=+vpCz9kk_vHxt7 zTi+Hk_p;T{iua0H!Hd^=??T}zZI(Y9b0HP|&?K7*6ZN`Kmz&7l@XXi7oinNOxxs%D zp*V5tMNTZTT|D5?9UZ)&$e_GAU3wurt+Z$#nh+0JFgd~m?zA5^qiv9AY)L(tiS~<7 zn#!JDr(wpKrh+{jDSGO(lc!0KS~b|GDd}SwvE#5Gnz4`td+W5_Tb;`sVATJNlm^Cu ztj5vzC1-iVd#Bj`aaj!V#aDr98hxnRLmLoQDibA<=r(^txb0o7J3Q5P!I;ST?m@Rt zV7fVgZknt`1~8N5dodRappxewtpZ^n=9isgxF@UBYy-(A$E%iFLo>USF9 zgAu*tWSwQXez`h}AH@#CkTQDQ6DVmW&TA9Hqm5VGFow2a3_&09SggxOLb20t1rWh* z93CB!;iF?r2x$1}-V1iJs(*@~u-1Mq;)*O?i{#lt2xR`KJsiwJTNxnnkE7TAtD;{G z7=`K$2!(Nc9SOSDk;C}3nSY>G*P(sP6_b}`K%zu8^Hv{-fq%!V(OFP6aU6lB0b3XbGOe)UKyj+Mbdj&`*$e1a>H2xaqAWYPrk^1Oyz#s}9_996N z9jK9;2TtsTw*d}&Nvj+cCmjEz7cH(KD%Q7Us?)~ye~UBc^PO(St^ zk9*b=?#_&f^Ow+4lDg4VuC;*f@=M2uC$^7p?E3vfMIU(aEmGm)U+go2%U%>yxj^T) zN}xy2`eYX|&FG+wka+#CwRh{|7G%+@L1iDe9HRU9e{l2(L6lND5@{5ag7#m#`j3y# zbbBF2k?{M4trK{ULohZADvS+))X&3oe43i1x}#KG`w$cNL$F4Dd8^h+#Zms&$YiCA z9{^GVlo9f9oZ*=0cBG zFpj(f*Y*-;H*9&IvGx1l$lB5KT{3EA+%cT$QHEjkBrUFV4PUdqhWlqnRFL`0eV&Al zJt1~zqnK6EaqV$RZ}QInhwhFw6foYx_4}u}0Zr|JMvVFgiB9hcHp>}GTvu-!G7o+0u{_*9~orwGs0l%B3DA@#Qr3#vkwF` zp!F12Y9hy~R;e+ya|xmbb0zBoeTl&bP%LTw&u($*`NxRW53>20;ILA3;)Y>inaLQ% z>c|E#*r}+SO;DsstHOtayh34l1ad|(2w)2@I~Xw&C3<(;8x;4LWMfvgv((|r)Ekt3BG@%dwkJO+^_?k z(AeSd`-XGSvl(UthK3O*#s}CViL4klf6~clW~o219e0rx9m_tSRqz*-G*6RztUWSH zxQkninJ+KIsRm4|BMtg8bT;Ae!UgwAfr=zws{k?~Q<+WWsJNDW6)B(s8 zuM@`-X{#}nh0Zf5e~Co7fF4_;I9S5IyVT2wmAXRWcIuh2)zli5UTcn`Bd!|zDzll^0b<7+_j)gU7m;5E}nv+ge6GpE0xg82+4CWT!; z8)7+c;^)pL0*@}EaMRn(N%$VN^yJG)fI(7 ziJV3sDsK$XFZY#wWaP}RLi!P6csAwXvKObw@TrTxApq_w44?G+slN&$9tCB{4lC|3Q{7h~gNoA*xaeFZv zB|F3C~dnaKqJ&|`0 zXVTqM?ce4uVX4ge{CGNR-bMv3GN~z^af__G?eY~PRcLeywLKLnu6pzsF7Q~}#Uhuc zV*4Vd?IwP|E`kuQNQH-Op`nRi9)n(Zg(Kl0)XkMtn;=g#$)zJ-r8HS>E+=yl5i;2svbdDcSBSz}`&w_E_UsYlgu$u4PJivswVr>Gz z<6cKHUVq8)k8E3sEN*e0ikoSNO4`xj^~znTaJ25z8@ifJ`huJJ!ch zFSon(Mzq#;Y}yH)3gSzO7H`%i`}(+OW1K%24?5B(#Xw?-r~l!l4MA!Bq(Kb zfcCjuRYKZtkbC7G6wu?V1mId>9T(*pH=@F9<+s~Ch<|Kd{-+P!Maq`W z=IGZjuSlZ0iL%v~@eqNJLKk+}`O|m@`|s+wWJmO#v4~Rh$^pJK+|>EzS91FJdl`HS zi?DBs@vI{k<~`Y`ZXloFy-xP`EdNNTOT6T5n1}~8jS?2c73GMMw4iVw!T4R2 zyGbpvz8Ox(k+d)JryAgqWg^e*-%SvlMnv-& zW$cKo&>=LNIQZ87YQd6Vpy@=lEoR$rUZ;aEQM*(A!>1sP`~8!Ke5VCp<83OVz4e92G;#Q zlXM3q&N2r=*&$c|sDE`=a6E%$g*B$%elhPU42F$2i}V9}J=>zEPfoC1E-d_v0YQXGo}nDbfpED4x7{b|t(U}T`KwwS z!MNG<L=`9c}MCZq};>!ImL#_^>Z1Bua2QATf)>47yGhWsx2=riWZ zY|ZYP^P?1Rkba8|4lm?Z`zvAYO>~Tl`N~%8E|ZtQ2U&=}V)=#imwpOA8k7bC_B}HR&hVVX^Xr zpIOl_nu;=~dY=}|q(man(VVFg{C_J5cL)fOQGIiH5)m!l)&REx_EBpyU7^(vhASUG z24v`^QaaB%vybWEKy0mW1;u_u&BZ~BWOe`RQ?igLP=Y$nF3D#Nj|K6JL>V6YK^|$YoI{Z4Y%3HV zen57$59maeu~Om!p>2^l@yk;n(vU+OkbFL{5 zvOMXSF=a$i5dy=|avi(fM18U51Oc=dg+-1-Hx7ANzV~qAT*^3yW|c;6Vh~ zxPF?OlBzx3>E?&|E#}H~s~o+yLY=)>%_LHtAeUr^r`aN4cc)0$_=LlZFY>t=FWRHM z6r+81!Z$UZB{miQVSW!?1eH>GgWJX)v+a@6?m>aIgI0qXsQ8D%y4FF;@X!psFJP|C(v z*?R3h8T|}&lCg<Kr!sa3L=ja5P9nrqnR@x z2AvNIdEtQskJH%iX9BdztR)2iU0j@RZ6eylC0(B+^98;@SbVCLORLI@hN#VDrFcEy)b6pe#`@}^=J43tGfTY#ASF2|BjLLFe$YZZ z+cLABgW;cM30H>)>KJ(tZS&fpn$-miZH+b$?=MA-@DGNyb#?h)xmz$HO?G}+R5Id*7 z0sB1$Fmv06JmmO?7zhBZ)%C>5d!|+?YF@yf#a-Kwy^#x$t7b)W2><-o#Mmj4!XtHX z$ShvSFVJHaMC;WowWFoaVLU3H&BC6T!~<()xPrt!z3R<%eI#S;RtaG2nrQTmJupgo zd(c1|r2InFKxm}!F2ni7o2?5_f&PQ2I0O7=;}54mqN|WqaClf`!bYhmI%3ZQFh1|5 z$Qd4w-ot2A*EW#vz|WWIe+@5-;y`%7Zd(2>0pn)JspyXa^r zqJkBB8;BZ<^ftsK<%cM%vM5dZ0ossNjlYfD1+&kP8uw>=;dpKzJ@+d=Sd7vw*Ua%T z6kIAUAZAuhW=OASecD&-CES)3IkI8}%vg@=~Tg5ES&}duonU@AIz@iB7?c)5f;?SeZZyjksIYZ7|`h}m!ne_ z@^TW@P#?hIL*z&KZ=r~KrfGbt_8UJ(n;-tJv3lNdm9&rWZgj7k*YaJTc;1w@D`nS( zb%w5;C#`;!O`$xRO3nOD{5O(I#4E_9Q^OB!S^cQicv!cJVt?D{&DMHX#oX+YXF6au zLhP}mGT3CsZF4y+o~Wy1%;S1v3R!MUm_WI@vvH*GURuY<)eWgkpB_|O*hifQy%kmw z|2oVs47r7wukfS2&CT!GOf6yJ?I-odw6lol1zg)89n@f>8f34Fo8dcNA)Zs zE!V8x<~&yVDZ$)BtotO3L#TsB(aJ$g*(#~8r+iq=5^IGr>?l*ydkwG&+gsOLzPozI zMxQ4~m{r1XK9fWm8H%B>b8DhDs#kmLY1ga}75NMyqta#~xNZ1HHx`I8Go$L+KJ#zX zDFV*aeceJ`{B8?4^7A0o%*NX>UPV@@KEdgt?urrIsikH%+!UQ0twn(fIA_>U4Q8{W zcX`hh!g&yvirnuIJ~rW2Bk<$R0*`2VEK6m@g^IXWawm7i!9hdVP*#Cu3Eu6JRuyyW zsg5n~l`*FInz5Sh5~MVX4iD3Gu<53c7%dILvbnqyB+Q^zZ92AbNx0wdv3K~@fr#JM zUIVyEacs3vl6$)!?*2G~{AYkLCWE$5e^cF2%mL#PiA}YCTP@|HIKH{Jw@Pt?Jefvs zpf3tPmyR?CwTT@YT2^2|4wK}6m^8UPJm1x5m_|n~ ztokC{Z_M(Lx53DXt1``_&r4B&`2y|_Lfzz;JSHSh6y*R&@H^9mGkwjcFcDr}i&Td; zIpg=XzptM3-e%~}^-@BZ4b8NiF450_LXR^~Qlx|HvGT3ZIAb1IfL0Vbu@L9aMA^=k zVHbXlyr@sw{0JEffNTHt%6jZ$>v$|Ane$3j#eY9?_80?sN?_^&3mi`3=`4tAl?5l) z@Aj)e(zD=XA~7bF74cn%x_;e)QBj)1COdD~t9A%nPOu|+zr39{RgPy3J5R4qiG}mw zX1#_{Z>@fu&G0}G4*@I8il8;aO0?{&64R{XK7y#S8Fdv8^pD1jh_?TA&XA@Vv1Mj# z(|i|&9#_QeM~F^Dn{lZE-C%I&_Ch1*&EORffUa7txZr30(>D{>u&eR~788+Q_A03Y zx%{W+W*2BT&#Uavpnv-pGU1KZ3`)^HRK?GW+-E-lim0Q0fBK}Qp~4}IoI0ST$|z0; z&iB#ev_+H^YY}ch`M4tq)>sLBtTz83$8o;bFa$#d9A&6_#B`#Ch!jHQIue~-w z(VIE4!K>cyDm~3E>lz1d5c+-$M3T((&am?RgBSY67l1*0u zEJ#SL zm5#-k7`DZHKB@n4eWh|O%Iinxq(>HNs*rbFj1b$6LjhXlPAnHY>EYs8UsUEFxrBqx zv?q9$8F_}ys2>Jbe~0PdMgjtFkx84&^}7Se^<7f$7{yyG<_}VCYn32-B76ZW;rKbT z3=N$lK~9Yt?*bIS@d@@v6hi|JlRIvu+gc2L$EYp}oT)t44T;pF4`|9LWf|D#X2&dU z{Gm3ji%r!64TGh>oIUZI!I#)XA_=J@?%0m4u=Wpi-8ro5H+t0@=h;alLnHn599;%D zW3PhM8MjRSB|^=q+Ec+`|1JNH63JMf6F4+*10uP3BWI-3Si3DgCfqkQ0x=-9 znn;-g?pI-6)<0p$Qfc64!uP;T&(o$m#c9A0f-!=eRav(iaEFjR#6Fj>0_-?B4%_6 zBk|?5WnED2QsZUxGX`^~?Ozub#IfH_UK)Yc&8lo>vT?#~xh4aUvU4K6zmU=#0a{i3 zQ`4a73Q085J2$m&fyQcsl@Z1c);n4)Tn~eKjNul%qNwVMWq}z}8Ph+0{-5SSp`ra^ z4$y?uZOP_)s}ZG98^bkTr)?Pp2gMind))cV`9Nu$u-t++vLGkz z?F4p@VJqv9^km0zDt8N9!E52F?!|U})&*=IfF&lo5<+MIgyqfT2DYWoGxw&k`~8Z} zdgA>jRN83Igw&qH-D_3kZib4s7z_yE(jSe z>Gq?OSSAVg&g`P;s(ONc3QZJYk#YiRN^ zB_Z?PzG`W467e)+K`10!lrS5O>({aK64s}kiBMp)nTlSlJ)#(fBO5dr1PXw7S;Ze1 zZ8Yr$ESXtf_^uB^W#TY@f-IEH(?kQ7bufZ^*B$8Unju9`$NaM^PmqBhNC%y_eHoA1iggJvY!rgee;J-S;XuQ!CF{YJi{W$bay@0o8e zBD(YoL?M>+2UwZ|jujNKgyS6W_1tJ5syu>-)dBf%E+GQ?Iq3|4y?n=V%6xKo`Gwd) zPO+;GI@9`&L0Fxo*B}2Lz^|F|%h8_G5J!ttJ%iCR=I&p=zJ^hzaF_$}M&R6FTSjwE zaV6yaB;BeMjF0zI@b-}C7n&#igs1tGdSKPyY-DPEsPfvV`1-0?Pk}DMRLOaLfNkE9 zN*IuK*m(4@x(u>tUZiz?&as(0mOMR1P81c)g)xP;zmi-2Qs;4piu)#~=x3?>BL3QCJIxX0AF=Muuc!J$S}jrNPa;09bS#HO!GzwG#^c|c_DoADiTO(J z-nlYH)mhkaYYTlS#JP>aHL<5#-}~%jK;@@=ag_6MaOJ8NmY^pV&KARyYjmAh%Lsmn zN_h1P0!@+Dp-T^z1%m;r->2O3!tr}?s#-Tkw%@Accaq)GCOsSDbYUX^9Ib}Powj|o zx!iPLJYv};Tg&*BK91muJ`}~;D>(Z^MY$+oh8btQga^v}Hr)10L&7Vc3vOZ}ZaKhB zoiDBhoocx_6uF!F;m`7?_ZTt%o`K8~qxj|#F9ujmLS6s6LGtb^2!e)u&WsmLcby(>`R$Q?*66(t08|{ix3<_8opJ*v8rHBd zcYNqqX$c8Hma$e1@XmFZ-=r=O^=rZ_lu68CCHoPF8AFYzX6h6gK)U7Oq`3iTy%AtH zZQ*SJUiHLaBG3fS1Tp_vf<}`CvVFN;j5{J1!{SpA>b{=Xc}OO|HSrv?0bRq-pCM}R zXivUx-okAv*&e`o$aLP)Vuzb32KAN77#Qf?730l z+40&_9w1>2*bC8uS9nf}Vl#dHay&vbo73!Ha?Wv-H_c)`r(h;tsuNa#UCU+1??#1b z$IqC_ivct{YdS(d?zl9kU2F`w3G6fckNBAC|dd_7lO}Fk9U@?fx)^wX3 zBK{v7OKOg%)h?B0BEwQCA7}2tbS*-R9cK+8mTNOMa4p2l4+9;flS3b^Kiee zJy%b(yE7X!_}XetHS)BfM@2_C&%Y4DGFeIlhPXzmvtRM|J-S0=rJMD=P|<%%ZbY`w zB(W2UnzQ|lcR8Kxw{5Pv|siyEdMX727@V93KHb2!9sG${I`0|>f0+gQX`6p)L zG@`wTMzoWwUV6X~5n&2VfpU5_F-O2@X2g}qeWK(4v`S}wrMct_=ks`tvGzB!XgSI7 zk7ro!j=E0YE-dLEat|qB-nDrP4Zco?0zfJ8WZk0eGWnN2L`^z^6LbN0jC5I_MftC5 z7TGXvTN%v%`!k?{l~HEsCKx)0$)=aAn;1s7qNbQ8(|w^##yf@? zGc@B#G(zbqJk2H&sii8ploS!oWKyo8CX?tCm#McnIsNIET>k!PR4C04K&1b%57ElLW9$M*udM24GB#bOx#jpcKYycsLX; z;K0!ZLS~6q@&rLtijf>4*kZ(QV2m4_EeH^9WbQ2);D)hxP(_Vr0-ylGV*qPdZrB5W zP|!dSflMK=1_OYw2yhCI0nldvo!mvaxdA?MZ*_2UR9K2e=47eTh2K-7+nb{3iTwmjqlB$qzdc!P<}@f}Dr`kpeM>4bI5@T|h2~Uo|oS zVg2}4SBC~r2@If2z$8W!{)xmuLx#(MJ1#I^ft)4xAOeHAlEn?y8@~*TKkYMcS7s;| zCVL|hlCYQaSi=b!uXp<1kDm=c6sBAO^M!tM5I~+lh^EcuO?NSZdSkaT@EV%MIXVb! z32&1ST+k~x&0xeSjW@G%=GF$*|FoE{04UFf*C?`MnaUC*5v;lSyL`|G)^P_Sq~sgaPxiOILT2iq{u;;1*xO`f(Rpj$WP#y0b~GRzy=li07<$&y4*Gx zqhW<*B4P@ZDgg0>OM?-ik|>+2jc-I@Ud`zUT*&- z_kb|`!U@UvW&WUqG^K&FC{QhI98s7UVFm&{{7wL&fRn->|K?YI6sMCUN$>^|1Wx}h zZt`z2kODDY{XxGEFW{OosqqUIBruNf*?@MF5TK55AYl3f?reYKv>6!|bTX2DA_)Qn zE-fsqt}T84liHETK}ce2$%DKG8E?P*HV9U|KobSKa1|7H+5r{reGB4uA+UJ@SRNiY zEIb5k`CnS@V#~f3A3&J+3rMTO4=h+@F!;$zkZ5&v34lN;@x#IBsv`!h5GZ3PKnR#J z5drdqj?MR3rUa4&j3`vHNKs(Sj38{T-GNcyA_2!D#0tO@3BLGiwYUI~pjG`LFN6VL zjlpm6dW_LG-1gtQ2YwHe+f=2-ut4E5i| zyqKGV#Ug=pAv-LXejfov28%(wEQN%jfF3UF$LsU;qbj_2>ib2qF&Br35wySnh7{tF z;GjssFRZV}e$y`9eLd>w!zbwZ^E9)Q0|XTo;Qt*sNA1_W5A0w=sw!SI)mhHBkKLn)aSDzBYzJnWk*}iuEaKGM+r=K{;wi>hJAIRbBo7JnwMF zK{iDvB@JP-rFUC);MFo2O&rDP zaDVxC*vwNCWnPhO#8PB&5zt@~ydt1J3^ul9-uSFCngV(;x#WG`9;}ftcq^Mo&65-s>64A@+|PEWFI5VZY>-(S}w{SvxWZd z1wAAj6{fqFgGGMb|G_o5ZQyp7$Fo)UTzp8r>ch@zk6D`rzxWm;LL^TX!7V)&3fo_F z8*m zEAN?JG@axH?kzPKE=ivc#p&nN_ww_{I|%bf69``Z%Vf7Uh?K7^)v)?*Vp7aiMyxfV zs2V?t*A-pSu8}gCiEuLR)cc;3<%9n(gN%xp-*l~p0g3(BL;E0MofXwKA7w{ps}qs5 zXv}QVG_`R;<_4!Uv51*ZexK#@9mA45$eFg3@hFBc-E9D@XBKYbH|Ox`(JJ;RXaErq zH&ziejlB4ohEUe$%Z`(H`}=IP4}Pv2vw8PclRDZ$YTg0gUT*(6eEPC|HgZ@2gvn$M@CbqeLlc zHHFR?!|_DzrL9y`lKx8jHGaBmyO@o;N1Vd8eSbRnByRSxp3K6mT9P=iQ7OiUeQBdX z!wpcm?Ln&WzDe@0rX*dVX*m;r6g*ge9?InM?}7#^!OsO0=!0a@n;5CO@P!dK>KQb( zS7!F3C5cL#DSZAIEAj|`?t|g{De-WpIr6+Y_vATd8RE%8Vng2ZN+N3KlFi9?vf$xU z;}tk-Nw^2%Mx%2-7y{HcN?VY?BA3t!Y95d~Nc9@$$iUgSt-&0-?%whp;a&hhN)C?bQ5gX63QTclz`l1$nB^{Fw=ll-7 zP7{B}U*wQ>pF@C$m2sK%>jGnqb|~Fh#k6Af0W;-o_}6i`P?B0|ab<|PrBdb^rh%I5 zqFZC=NW>B-{A@e-#RA+PNCHfc83V<$J7zV{b=#b}gRBC_{uUXtcS{DLr!WzSBT z&#Ua7kABcg8-(5m>c03pY?o#fa_X*R!@~9xw=Xc=bthj^$HDt!&*tl|LEyt@ze?u5 z9Y9K4zxixkbO+i$mpNZ9^kwJ7*N#&%zsT#fUfHo+69P=nO3B=Xajk* zaZBj;axEU?C<*G;%4$Ly5QIzw@Z#r}$-W+{G}XC{U5f@cv5`1VMSipX z51Ee>;F*(@6$8{Kq0*b`43zQ4LxYJ|2c&x5DSGy7Aj_vJq4<8DzLskNDun-{(uthB!3 zj#I?y3(M`6&!zQa(J-8u<}Y73F!OE@Fw9xBrz?>HO{>85rC8C|0iaeAp>`~oj}BWu z)8d28$nF_j`~(?|0vWvY-Dgh`rCH)I;);s5JS#>oTW3~%X@>Ku3IT0LBkZCYR(Dzl z5(k4qjm^SiYYxV^1G%?$iu2W-8E(%iMz63Dj8D_$ zY1&x2JJe$uU8{fw(b8L6?EL4rOAuqUBCLgcxcK0qV(pb0ENal4)-*&x((t_^vtoX8 z^?~G2d|$_}6IWOZkD|}IqP*#$J5~SfvB-(Ykl)f$gN8B%`hj)*7^ftP$;npQbOJwd z$}y^U*)ni6J>(E{b!dEe;w=3sx}>4%d&MR2X4H@@NOo$amlZ2~U0-@@fv$ezs=(u$ zRLN^)ef8ly&XP^|eEw6e!R5rhkR9H7&-)_MZ>5Hj>QmcrFMhrl?)J{4v`L|Lpk4Lq z+L`=yHNI~vCND(fU!n6$9yJx$Wi5vYi}zJHC5_0Wyt}1qcxvTz(?{PD5T|W(7?pXR zD#sU!%=deqrti-6snO8ylx51v02}eU(~)YJ;2NWoZY;U`&B(y9W1sx7ma7@0a4Eh1 z2*-xWq-K5Lhz_!=-6KLcZB;(!{>Evdl!ar@&D5{Cj^(i(4!XeW@+f7uY`J{bvA$aW zyXP*ceT$bo`8_!KQ3BD8$XCXLaf>ox5)pP-$=1;a7*c8 zuDQN>Z7Y7!_>n)wHhwbjhPrl%^KrOY^-X9jVT)ULtzG-~qz*9lvfF&Q%0GgK3_k-5 z7gmO(HVtkiV;tlnb#rH!-uBqC4*cvXSJlb52)&toSCU>Fl?ob>{zikSzZ>osf{ES7 z>pSjJ5iwb{U0csOVXH4RdgRkYpl>&`V4Iy1?z-UazNm(VpX`YDA1Qqww${yiTa1`o ztU$XwLPkgJGLl>Q{H9MG8syajc9UUOkT>}~OIN6gg-ioA%FJ>+3^z90LGxJ1vzqml zawdYQS#cfxC|X^kvk-Te=`N2q&zdY~-y9CrOe<|m7ESw;)iG^<10kxKGMbK-0x1MHGCI&7a4;61UwFWLyx0t^tuB-B`Q&5^_p&?`<^ zwQFW7J|`xJt_kTEk(rf?XDYgNmm#C1p(=rAy6v`at76E+_EC{%lU-`IN&-6X{k%0K zk;E({{@HiFS(;4Gdj8mo$9%NhzO_TypmM;p-1gc26Wn+jnOyYee1IK$YTX$Ox_5KF z9cZI2+4|TyTPZ&hh#x7k@yhTPtCwt)HqVbeHQ@JKZ&~QMgYzo=Rt^5e<^1<(Z4N~) zf1%mx)iMQvy8P&I3eEhfT=JC)W2>q%r1G-wV0prz2U6+H|16XNmfD-TbP( zGt;CLh6#^a-m@_ZJR(KzS~E7>1_G|}uiK!h+}G;+3UafJzDg3?GmE=?$$k`-INu6` zipn}p^L6(YHiHemvwNq4hthS-4QfWo0zKjif>Y}0XS`ec#`m*$2z2@JGb}mlg$Qn) z&2l{JOdD&@<<4EnHv4#mN^lO^aBkHOy6tGG=ieGlObwUU&>+aAA6b7NIm__yX20*U zfF$tCf5C2rY&Cuda<4XVj(1DJgc&;^{Xp`o?P=>VSijWJr>MPXfzz0;y}G)utIT9v zCC55B{G^BXrCxI4ym)3qx2Kp43+iscJ$3tV`dzda4kcm|N|!_XLs;tFuoot1JNYsW zRM?8-b+hlJepO2+XIqi;bGaWl8D&# zFShf%$UbcM#|Ep6m*DpUh4tw?W^gx5?VOOo-QlwJnzkH<#XR1H8Qy6s{&=Tk2F1a8 zME3l5k&2D_4hlEJ$#1Pb*CP!Ke=~{r|m3 zkb~dnPTr7LL^-wlS`*uSUN7?b%nTaDO0BEm2`DYa<>O&kk9ir)o{Y0PNagRQ;>A$b zU@t{pv%dDVWe(*_<*eMx(nE*1q?xI*fb#q&{+cj(^;aodAHO{;UcyniVo-10zTmta z>14S+`?1%??yA0|G5K`x>6c;4iJ8I|DJ;yPQtU6k$Q;Cv%!mmiEUj8|-;8 z9xi*RH`gfNOsy&VxCl`GBD7Aj+rjP5t!m-msj~63C>hjc+yh1+>>$NXc>C@7QHYAf zI7J-Zd48-lUmA1EzT2A7KZq+h@y#XfYnoIgzv6eES5?J*1^U0vXTLromXDRl(ED`I ziboTu@h#Uk3yVRzAml!WeB!~#4)t!~zH`EyrVdKkKmW}rPD4M@77rfMBBS`|kmu*| zi=3~8;4HC4jjNz~m&ZgT1#Z?Q_aau%U9JziIDcA}Q=k#ujT-Hm6y>q@%@=Z~_CF>e zd$+I&%`+;meeVp@_M7c-n3392Z6~Jpi7e1>$(*NhslkI;DJsH{mUmY3~0V5mh|CXb%Gcx`E9(t;H>&RT<@7aVB zcB)XE&SL4FYbi&i{)fSmot3egzG87Mlbg+9&gEQY*>Z|JNuq_Ko+1@>YG6?Em-G&l{vByLM|Gx-3$L37HH3`SI z@y3|gw(W^Cv2Al=+qP}nwr$%yvs<;_PSyU1=c%r~yZeIFqv+dZSj#31oCyg9i<83+ zpDj!N+8mk?ROwX3WQuqtY-{}1f3rW5QF^70CP~N z4YBX!hlm=hUgFQ$i$6~vn3}FPC}=ScS8xXf!yQZ1ohlVVat95TG!WF6%cMBRCxZkh znZQuR1r3Zu1==UWhYH0&3r*aMFDwYi1fo0yN&sa9K$8;x*dar~AtphIBFHe+NfapQ zzjD)c6W5W>V!yh>4t@hg+d=$AiA(i}&`&%hxPb#Rn3r23!Yi}u?pf6WkUgH%mhu82 zNc^>D6-vSzw8j~O^U4(q9=De%qX)V~g%Ua%Gca$`H0czRWnkpVE>k*v3qwU>oi%Vn zA?)F(^m~e$%Vnz15&_|f_2a@r@|Vz8hW{-Hyy~YY*&Gi?Y`?E-U_=5%RbtSKSE~P- zzyuwuNq|d?77Rodms0!|0|L4yWdFtgP;q#%r!X~ivzXJN>0k^l9jEAYOa`+e$3%;GZ z4N$+mgt%b>%-+2lVU#e8E=0*W8JM~Z8p$Y9?mW_&K=dBVc`D~MonMj2=v=-X@1XM? z5DkDh35fD>x8yQ#@%FrF^jqL%tEyNx0tz7Qe?r7`FZVMVvNj{`k#~~v2Ugp=jdij9nhCZLwndlN5{`OwC7`Y7hTGgVb(GoB8%rpTlXM1a5;!SsXw`&Q^Q6zPFs5OU4syo{~16?P3 zdHxmy*vrB8md4w83Hn=9vBU)ULHBJ<4FMZLe8yXEyVvqz`m%LJ@V2%oAFu0gg@;YD z5R*nv52@)%ahB@18btUAsqvQXjT%c^>br0)RQ7zvla}9jxVYM}lLWa66(G(+*IRXx z)@s$^)MN{C^txF?r8f02mTbQTRI5G7T{5A~>@`#8+LLrS+ZQrV^muI_-rpPPrLu@U ze(wFYxA8Vw8V}rB!cUX%HkuDBU54i4JlDA}*a%p7n&zA~r*3mV@p538-eth#;bASu zm)DL)0C&#pzr#8&g0W@0o|lz7^OuTC8%TS-jCGb9gIv31Y7N+KY6$+N%%F4uEsO;- zJr50m%NE2`Q5J~hC3rA$D%pCOUS6e6-p`Y3KaDOpw{%)nHTY*(+Qt_48&5-T9fDpL z+U$Y+_+~z!B*zEhi)^Y>{Ix4Sqb+lL;GMaQwuVu%#45aFi1pElURN6C*;f&d;>pK$ z1)AJ=2svV#eyFNQfWgDcTvn^>PHuAyR6t1g_>{S$u?PUOr|Zncw#_-t_sS@_L~)4z zN(&<-GI85O2eOE(WBD^ogpM*3-POHe-2hZe%l47tVTtTtmDW~rtV!*#Su9KYL#vC;IOeUL zTj`ncENFDEyN5{-qhXl9R<|Vm#hZh-(BW6R+IkdjMfc@8f!-etho{U)I>~}_h4%Ub z^%S(}=1tpvd>BweIt+`@d8SfHXQ*@a$H%eYV6cbg@^^*Saqi~Czs z<~i}kc^Teu*F7@Ah?sF(R3kIMV*W)5ccE*b^m~l_{d$-Yl|1ELKdX7oS?95~DT^Ir z$jaHQzGw;RfV7{m{qGjRXYT??&Yzeh)Tsl` zO)Rn{LGsmgtQW0-r(@mNGla{2+xk|sG;d!2M@M`XibnNgt3BJx}jz! zCb~~+c-2&{-Rg{leT|

1?b_j^I5?L##&FH{NIsby%i9hv8GE%g$V~8CBhEQjF26 zhL=4C9!CZT_L>l5z+?`M=`-F!GPmx>4Wn61I3~&kQk~a_g*=l+W7yt)UGAC$owUjh zTH&?i$iVblWsgW={WG1OnE8_m{i z&DI-Y2ZeP@hlM>a10P){pPtoiCeM|gcN#NY*PVt%%dk&!fZhlfHg0^X3!u}!tzptsz9aRJ$3ao@NkB)PG=9YCAF#;LEDL|Bi zDkqAdB>@?PJ5b$AEj|+bO#)b33Ap^IM8w9|hhxzSs0*lSfB`8Nfrpk+Q2uzy}XpM-8Z`tK&)rZYpMGz;WPft?CHlmIt_Ab-vp`&@?#!Z(1bjT zu&W32spi+y0UXQ545qmO`ok6G|6SDhZgv#10C|1OALPx_zh^5m;hP3}`M38cVOV+R zn?Tpx-QC%fuRo?sLXOq#}Qrxq}_hRZtyP0ToeS31d?^_r7FOVU#^zP32)z5WW+7o|j zq+exVX=prC)+6Lm<(ZKH!B$RQ%kP8AvFFQFK&;cZ`Xk}EOfT}8i z%ErK#bx!S?cn(0or5f%>(hSxzwxQ|c2bPVkfsN^V!<9rF7XsaqSxY=O@YWaGSHSKk zA`J)+C=w`_7UguQ*! z$fsxDH{n0jeT;RWk&|$Ed}iN_RCYfiDYa}6z#Ko|zRLGIdX?&+vU`}r#|B@0OmazjNu)V``eW-eH^#LCpNml>Bzz3Sc>8esvxk$-OD zkA9**`2U!E?%;o}aK6S=9_m5X1C^&?bEbZGghh4-WTi!596%zc@!iJE`2H-s@k5XP zxCVbuX(=Eb?#qqzkAvhInj8YvGBvNW}wRcUu;=_UTIUl;`$YeKb1?M^yHMuc0H;?3=&{lZ=a!Ao_f zcUvW-^`Q8HRLH2C=IjY@x1D?XZ+99Dg#caJEy7NoOEsTU{rPGxfd2l96N1*4kMiO< zgZ6XWx+InK)uk#QVeXM0bmeZA7kHP-B_;9l=UCYU1 z?rAs%$#>979I_T%C2j`gMxhk}m#-$!Zuf3oVmAP!soEhKczixj%IpxDVc+0~j)G5d#UiqGLdZ z!z#R33UhMqki%5J#@OwsZB4?nb7w4hq6jrFubV#)HGduqab#KZMTOQ53)Xp%m9~iV zRRW(L3ANpge@|*`$LU9=5Vplzt2{v~JxVMRJK+xX&`RdC;x;2nZh5ELzU^V-w;_+g z75(0x7JK%cBx#UJBV>grJ(DsUf)#h265=GN47m3>8bC>kT)Pm1OY&%v(B);New6`1 zO*x=+%m{Uv3e_M=qQ-O)%|~_!`s3?1yOK=r!ROpld1b@e!wJh;FlShWiK$voSp;vg z(StP&!V-++@!KdF`ei5+_i5CdcM!;6uf65>n9OCC`V`iPJ2pD9N>;;6jLs!S4V@~p zEMsn&R`p22)H3sXa2MB>+NJ`2`f34BR@uIonY$SNuH6rwiT~~(Pd#b8A2OWgTwFS^ zz0*U4sr=+eQ;6#IKe?ltAek|<0FIxqeuv6^NRQn4*FbJo}B$u1gGrf%JvhgR?&KoMfi}>K##0g7y9cFg|+GTEdz7niGAUi?N)% z4rdTgO^>7vh2y9**T{3aWq(iq;*j;UhIpIuP{rBd!4ap4nS#nRg_CTm6&FWjJx(KQ zhJ%9Ul%<8-xB`ntEwTN@?XMJE2j&(^haa;Rh9PW@tcJQ9ND!w9$6{LLe_I#=drN(A{6 z=6CCIg5eu8*ar%95ZLYHN0CSaPlq!C6djiGB?&9wX0n@Jz1#;Tl>cQu#-(=fD7AAC z(`?FC*ugS2ENhcikc*vGmRVF(0|#Gixy|J72rh{y9CFiZ8xDZ;=tYp(ek1aG5AXCX zoh0%*pw-X!?kK;K&IYcKy0ItHEkvP)=5%*XDWh;WRqV)&A5Cf*7Y{2ObrOpFJSfhxZ1j-z2$3Er`LLJXga=hb;*8H!MB@prTN z!Q&J*21B#@A;r08q0?u{&Ylx(d?2xnfM82@}QR5qJ-jl28W zHaL%EC#yb9=>fJC;t8J_}l4NKt2SBI971L;?3C~ zGP3f(6)&yYlo|e3D(}pt912)be^3&o;)ARd!rTbpX0=P)Y8RNcW>`*9$5hew1)p*M z@Q=~y##6~zVO2^c6|)qd&DvaXkt(AaqduS;hR~K-lN=fGKFqHDXC@T$fjL6Y9!uAu zG8$&q6@hpZQ zG7?JI<)7rNyFl{{zt(zY)1pbN;N!vbpz$0z;SoH1+>l;`h-#>}@`-A1W?sd<8(9Vv ztgHJ|!QfS>KJTL|SN2kt*(6yRoi8-4&D8K3tH%fQ2@gwqGpU42W5wqRbH*6_wQ@vF zN1uzrdlvxNqbY*8_0E@#Uo`3DxZW9JnDPCth@Mbgloy)SHmTgIHJe{SUB`vsw5Yx} zx=YliC1a&6n3_oO3>T98RiMD&`K<~Tz_4@_V!@pkAiglgFNL`6M01MDGGA|!ntO0| z!6u-0vzd zBsIYNqk4(eu>YXtF(M7EP9|!YmN){Xi8Lg!QA&E}q+o+`Qn|!0wpWmv15KcV1K>CF z{f#gm&0))Bv>CAS_wJs67nBdAih~1KRXD=F8<-R`aMknCuS-qA%82_6xpS7?X~%Vl z$kvFQYaSRmqU8)_(VCG|1Yr$6nd^A$s6k4^&j!o*<7^(VUa0&Wlz+}+qHaijz{A>^ zTAgo<>d*jsZvOK@8~3Nls{vwToeXKy!uA={T}DzmoC`;D>vGH;>FST3=TW^Z(Yote zNFC}kIeG233k7V2#nn;hh%S$78sh!WtgvZrZ$KlJQ8_h6B<{j^Zx+$wa5jfrWho!+ z;nIaEf z^BN6kc#V4DUi*1Usdc9N0*|A>lggQsTqb%nvA-h4+dDnvatbuRJYF^QHeJT?R0Gbe z1*7ST*6%BA8#6ohO#ciwQZc|6WTuYY!|Fm@zvO9gStV>6+;{BQmxI{3=oqLuTx&>4 zQ1>f{>%&WU|81A_PY+UqYQxO0wH0F`D~W1HZYwnt8JDWK8e+DcZJ z|4_rk44Kg{m%-lm@k+05ssve?cCqV9;S`HMESr8#ggX1@qZSO)A7)uayFBy*SR*-* zlnWjD!@t1@E#B>mk*paIk$*3C*h;Ads#KYKtE^jjw#OlFPDuT(Vnh3@DsNG0O+@>v z)(6kWMKzMzF{_D%=kZr?MW2f@jUlV*BA~ru)Tz$qfjg} zcgjLdTQbS`$^E}>C+WP}g+J)^j!r4toY^HUeVw~Ej{H6iOrrLFmUGDcI?`iW$hEqf zz^~TdckM+E9<|jAv`OwyalWc*lNW*n+1h51w7L3c>D;K^f15I(O|}!#l9^f7l8dks;sh>&ujwMD5UFz&sNr6tP*VD^IwX{M5ml+ZY}t; zzb{i;1fw_qhV}V!GY}E8pXv|AeDFT1c<7c~%4@&+huWS-UTL%FTQnNSXgRP!Ze%lG zPg#g;t<#@_N578tGDU-KDc$#t0dopaQmTV0c&8EYh#bc?&q82;@`YfipH#^{( z9oMq**2eu35Sb`j!VN!p&N$T7#T6w%f!gA8*dP7kfbLWE;;dA0J8*Z}Rq4DWcPekG zZ$MN2z}^R!TFp3>gAnT9q9y(mamsVA?WDvse#Z#~1&l4}d{=h8T&e^r+&ey9Zp)7ISv45)d zNuq#^E*@BDpW3roZ=uB(P@1U$*mF|elfk|BTgmXEsa&#La+~wY|Lq64dU&=>`q;$l z^|)A}Q(9mp8_X$%of+Vn7VJ2e4!hqmywLCzq>lhUoa7c zU4I(uZh9q%?=S|1(9HRFqt_E4Zk%^`211|)-Zt*VY&yr26fi1fr(_^se$ek%k<1#C z^8t`QSkk;;eO24zU+-w@Vvz}(e(Q%_9<^FooR>{XK7QdugbmgTy%eB;AA zTHZ_ZF%*3T2ub!&?}^5Jvcc!ML?}kmvG@@GqyqiHjwDTZ@Jq@nQi}3=Aa58AKyy`3 zJt3G@C-==cWr%1039QhHlC#~t5$bO(Uu|CK_pVvePR{MB>f+D&Pv%PS*DJMmPc%-6 zXK>jvf6SMsIJ!puW%N%GArH5M?M(4CY1okycCo-HEilJnfB2y)it8kI*;>^u&+mxO zbKpGj`jY&O)U2p4u)kY~e0+5@kXG%wi*gZfGZr-luaVDRhfJ}t7z}aVTgJy+(3tOIG+*?m?p2X zY<+s?9l4rL0kMDp=3sgs+Wx51C+<~Guw-#t(d{(Ier`V5?n2n9#5Ajy#$IkXCZ2CW z{9nN@0V=D{gjJNHyVn9L^_S%gOoxxR1bcI^d}#b zeeP32-HZHoCpjkZ^|t-G^{_j$l5nYtJZXovp0SE| zKL+dA-LZ=Vs~Z+{lq|`~Qejj>oBNzgn7XDBp+zY%0z0klVejmIj)`MBXXF~SckM)I zW)8vHlU;3yFpQ7)_hBHg8!_M$N?@$74;=sFlH=E|UiFfP0{7&89BEKv%;N@EH0oKx zl}0lrey^sukIBv38@$aEzzHxCx*m8D?{e)Ko&+{9NIAIkxD%ALCt@vN<>|?YzokSX zrIMvoQ?$0nBL5TzPeR1uNN14e)O4t z@0_>$)+?-Y+HxTytfdI~fETYHEp;Jb26Ut;z_LnJQ)a!>7Bhix!2&X%Vb=xYE$os;Tybf1TbNhkXgCwO~O9aG={=BbEe=1igX)#>>5d|@*=ItRB1jN@( zuV``A+NRGsxSVt4IJ-mIHf(w?=wBgGyiB>Ed=J$@gR`!&sQuKfNa~6pmj?8I z7a5_QCWWl}al%Pyw~YzJ8b84LVW(7W*mkf#IcB6yrcczP0+J_?w-;1aJftJM_jxnO z{QIaLDXH}S;@s8M!zbqR^fKWbbInLT%9aIgY^!RH&Fb=*C%YmhBa1(Li_J0qS`9Py z>LshWhEegi*pbs@2Ynn32WVLnzBy}=<-zDmLczQs_|;iE#+zqjWAKby)VuD;_{}bR zDTMR%FT^_~09xe{w~6n^C~hARL@%@+7~%Y*wde#eIK&QCGw?E3zvceQ5U?K}?Ah5r zqh|%4+<=Zz6D|$4f(CAz|+08nEn>@1wmQFY=68L|LpQf2q%rUto@W;Fp`F14dQl zj@DztT5V$lC&?ziZ$CfOZK4iI#krUF#Z@4D{pc+`jKCQw_Lt+)iA>J$NE~f*W}fE@ z2SI#wx|D~1rTuYl$$!+@eM~hnOUC5MS~<=CwGn2DfN0QUa&SDi-quehzwSlfbSY{O zrPaR6>om(a?}Em4knkqr&g$qp-@Ay&t)kuEQYa9&D_cP} zFWdc)NKZr2lsC>s4oV-PSeJ=7xs)~zTYrz$bkpto!I_Ya*_NzFP2;cDox^vpgUNmz z5!n_q(=>d>yz6?m<$O3_SZu0)i1j<5ujBHXJ5n#vr-{+2!v+-HgP*m!it%@azYXeu z`ROatkk3%=NdjPHXFnM?0@=ycoG_71XB90cWK9V&r-`a#KG2cr7&9)mr;uKb_|6V=?}XK46&v`JBi7oM%hczpRf@(yD>Dnp^@)Y z!O&n#$kjPV^SE?5yPCvNCs}---Ax1A4Zw%qmr4Oi;Cp9~LCA$5|I(#m9FNaL>!6UN zFD-+bz02F;KQdSVbYH}uT5UUhF^NA;lEqAGm8ollbUj{;$1jh&TeEWPaOT(SjxVni zJo*&pUhiPCQX9diQ`~6Tx+WsB4)L3@fWpueRKi8%>b#7LFEJ^&&OVyEu@U%5mvn#R<-@GDjHpyRj){q6aP4(>)`gKLkYmP(JJt_kaQbA%Ql8j{= z=OMe1)k!UwVv7LW?8i{)kQ%_4;U(>?3qs^=YTReZVO2^U3~`f>(U3I}%wq_p$z?Ur znI@@D&;vI9WLS!{u-6hP#@7;pXQ4`mis)27|*>l_by~8v}efRw)oEaFGd~Mt@Z5X;e@HLYhS+y z2;^ESRI(q|En&tM`CKWXXO^|RY$y`RXLX$Tbep>T1de?&cT24Pi5G-A`zPpYAewzI zqrfUSV9s#+S(1I`Ee}3Z(H>f)ouH!&r%rb=i5onuwRm`J?;GvcidSK>e$SJieY3n< z;sGS5Sm#%LdYajJi1r@j#tx%Q5an&NdJN%ga50J>y9>ZGIA42n$iAV zGNUWrTGFpH>rjLd0kkR$+c0h9^_bF$TJGNyrqq0jNA_xzG+tOzi=ExC38ZUYe8FoY zU(r&B!lN7*A@fvXWt3n2*huHkzW?RD3llBVlCLDd{(*|OvFUN`Xw5? zX}zH+%MfxrYkqW&ho8P%>rlI^(x?2)&^wclxd^DYqqb6Ym^XGY7Zh`1-1iUg*1pd$ zB=KxM$TRRzq)2U^!_4ZsGPI_M9Y50#U&0lmBXy6o{iHM@YIoeTY_2%YY|m$gU0I|- zP5GYoMe10pvMnaeO~BwDVzn6;tSoMRVg9w@<9ctz0=Rj}dOiZXnQdBEB~Nbg$&(Nc zA$=F{QFOqZ4#Bz>x^FZE(wCY7%)FB+c7P`u(K}+jg-|&IFJFmsJy9s_1(ew2Jd((? z@)AQ$AGP2`B}gNWDVlWrLEK-9yYL(H`f$=puce1fhz$hT9bEQ>1Q5TyX_L;J3v=0(|AmmWpCJ*!9 zEHBv+6y*sE8$e^9ENmrPOM7np6NCmOMyLiU3ydB-2#MURAIdL5&dP}d{&H({2cgYT z7S-duqN_HAkZ=AN9`vage+Bz7xc;QeoRB)qInOnPOQh`ju}X^g>3zZOFYG&YCSj8y zoY$W@I#}8FI~1aWBgOm^nm z6nj#P2g?L11e0h(r{;~HMlH#$?&0DwH_)40%l zIs3t9jc6>GUF^sBdlMu-xK1>b_)fN_#>J?1DQ+s1u_39{=i&-VKTp1~Y!4rC%+mzT z-dXz0+oJwmBPhE&f8^+w7b}?3#NEJ$W7oO+MB^JwMu8is_Zo5_&WJIP3^I*{Ai^(* z`-6>KD|VCTO-wkkPX>ByCRmcgU9?R6Ofc-*3W=(P2)Q&newU%;*gJY%HGwKi;EQs& z&vq?*W?Xu|X?7QNT{uXQo;K+xjbEDmCHBATy3B6To_eeGp95op*dv~yrk@);3u7Ki z>ATfwK*W*Fj}ELLTAg46S!O0=Uoo@-6V;^4YG{da#Im$+dhdN1_{TaKq^sf z`%5zkHv5`*i-y-+!h@Y}09Fi{rUi6^KF%faI{}#ze!rEox1PqNoeYh!g7IRxBY#Ay zPBAU+A{5^CtEkcmaC4wVo=mC#fWH+QK=NW&w6=MNj2Abzq6qZ>wRJk_t4g7_+{9*; zD`Y#?oDSLZ>jf3rnXU{e-|(nH>*Z7r5xAQ!=Ttld~aSD&sG_JlE&9VC&9%}bM?-EX>< zPMMZL3l0_Ma}c-kuK#M<^Y#zjJUiGLRyB)nL2nASE zhuiL>)@XI&oMyclFb-5oEV$z^muuJTUff>$>-s3hg9FPG za6q%L5Vd+a<1C`d9g<`Lycvk{Mkm9w<0G~bZ`)Sd?HWl$rVd{}LIuHwu$ue(zEIC3 z8rqE;Fp-72fbQ~>CbwrIF>J}JkN6}^8VtRY&Hg@rxa&y0y9;hF+46x&m>c%!taS49 zV&WS?ZJ7Sl7D5biuOrtgu5M5#=SRmp=H&x?lo- z4A>#PgwLGA9KV@Fm%Yq!y*WfyXh-k!INd<_0+wxncwAdO4904JVg*n#57l^hsTt=q zL+Bhn4C5n7XY0X~}K(`p*)O}&; zwwk>ZZgrLC4FgzJtx1;c)sJMcdws@rK@O*IcH%1X`w#Q08V52=R?daiK?llq(7$xD z2~=3Q&C@1cHh8u&vEDs1wz;E~HC=}}lT}c%L~(mSCOpM7jzzAF?*Kh2e>1875eW)e zah`{vRzi4S&}|XllAgTaU~p!meWRZ;aGn;bLG zy&dR%PkWMs_1UVkm$d;>LP^dU0;{HJHBNR0ex-cL7=`ld{Plbd4upS??`L>$pr^4G zG^rS^A(m<7F#xgDtqzy$ST)E4VQS9|AY2aELf05qPdTfw?()q)pSqt`ZTjl>w-*Y( zPV^^Ph6`gZy>}+XF&1yDdyGqCF46lbM-va%R}VN@Ep}P``HE=q4na_`a8sE0m7R(V zU`}Ji*b*h-G$ufGY(`h)P{0SJ2*^lR{9BRHmN@2`(EqdThJbito^aB)T;p-WGzZCf zMC=q8Pk*R>v$=6JKd)STXg0dNz_9-GJ+<1?gK(iK{FY3&g zHB8p4&tBHxS9~`H0zDNr&nw;pICk-^^!kjpCnGZHk(;jl=!F5Q5U_0J^%kgu(8e5d zP^3Hc=nUqSX7O(Z4Jxej6i4JbIQ1^)@xKwC6(fu6J>pk7=af#ac1C5~@RwtvV`(Gq zvzomsmqIet3Dc*|m9g)%ZI3>nhjr`ds8tJ99c^pn11R!TMFyauCic3+v z<07WU(SJ%FQC`9R_Bq}JCsRK=L+R8p>fbrfjipid%uK7^EKDjFBgicy-WVpdigGHl zbi;4m(%j>?+{Ixc6FAX&lay*nJw>nW*;6?%6NrTvUzmc#K>IO{ct2Bnc%WwWWzoVR zgfxZPVv2QEpf&1jlf!y5%>d;i2LbidJNwX`VHWH#@uhBCY}}7M!}ZNok_ex5R9)_- zw&6%kfhhP$K5Z$MMbD}X=i{G-QzQj%r=fs7k#hg$VOce3cqHX+qmnmDgSrO&Ry-G* z82CQNVVSzsBf2!5Q~d*Sz_*FfUc!rouAuhZZI8EonBPQ7WcOA9?cz=vMpA{Xuzax6 zo0_k1mLuQV^*bI5h*@rVkX{&Mf!_2$C~wDroBUzf0xsP-D@=YUi}CdAX{6(K4^GWG zoooE7mX>t}wU#egaH-`_Uj2Y3$ah()5bJ(>%B>o`X@TZswn=HKk2+a*#IR9RdV=Z_ zds$XRY5(5JLY^IyKAg4z45(@0B=e@n*|22DdHt84WR4;fJ3YjW=UBdrTeR(JWMM_j z?jm|c<)FibzbBTuc_wu7hNqxOq zLrMNh!&&wNmul8KrBhGBE;{+Mc%r|qK(F$*5`nREerl~WYv}?TYK6sTjdQ)UcZHEN z$zsq=u|JEgu#67zTiuI^MYc27?!=Z&G-E0^osI00-#S4R&(~r$(e_DwU^3kaCh@4AGx`%iJ+rdDJ2YqNoF!Z_O3Kb@ zWTrB5AE#pogOf$ItKp}%?Ar&X`gfGq3dkJm-{Nc%7`e&EkLew_{JgJK?An|tWH%Sf zMk}ffAY(8OiL+cmfYV>{{5(^FeVirxndJ1lJy^b4*^?LvJCwT@Ext zrhc^-Lx*a*Ev)x51bms%dw-u8!_k98M1`ogl`DD*BPZ+sTV`Jz@Co4N*RZsEi60wl zq|ED1iv4Vmn)_17JQzaGg{8)~vVBAA>JYS(`1XzzUpUxx z)93M1I$EPhSwr~M8-kA!ozN(x*y*a9?@97fZfwCccHv^7UXKIjva-o}8o)D=v#zXF z#Ixfi4Z+py!tZKar#ah)=gpkIlv&hAYpcZEn2!JPB(rbG~$^# zS}deW5>w5Nva9%Xc|E7>M4G-o#(?<`;JSYNGPgB<`3h$-Ep)cR`g@^X#Qvgc9O}5= zd6#QQxrajJFKYvO9E^wX$V)kOXA@RWHW59vt?xp=K8U3Q^oiumtWytlFX2vaDcdPt z(Frz{gU2nL8L>R65=fnP|I&ME&9J(S?fGh(wU5=_N|I7mWuSIi>ReBsNtAPPQ)1f{ zB?>6dbWGRvQ!N3MpZH;p8Tw;;GF9XQ8!vOYW**#lrCEf*q>P0GK)n*mhcf1q$M_d8 zA+!?p=}FP?()LgiA0*~0* z4Ja#~{2ekJS)IRQ9iQ!i&G)8Wx9a^SONqG*l?q)8j~K)pLk{COhWNI|JVO8(J_#YQ z_@!r0t5+^CWI%w`BAUi@DL6krq^n8%v}=UKfC_SdEY_F{PV>o|xm<{D+6ha>l*jiH znohU@3S|eRj7F6+^FxbLBC*@lqeLQ`E%apwZp6pvwX>_8zOuQs?y+TRPzYmXWP=HiN6^0oze zd2WQZiZ#7Hw(zAJ?8*A>5@>e$ihyx74Ts_tyh!u6Lx(HcOPe_VJ3}7>%Tk0 z{3l8lKK2FaXXa~-8!Z$H`c)8{&m(TTFjb-UQS$9nR&W$EMs3dW?Ykfqth(Gbr4$eBa2I{tDq3nV9P{2X$<1QIXkln9<*+|eB)#PFrz{)* zkcCG{bkCdN;B0Hq>NLlDWztIpqR1|xS&|{Q;cv8xJMu|$?i@wMQ_pj>ycUgjk-$-+ zKD*Nsw$k57xS|m;u(vjH8Gj0Ma&qfi!REXzpF5_{_ZhBCRom1g_>O7t`YoF1GMeoq z%5~!WSKMLs$mFjI@nbQ*1J#x;=klg~_onk*4kpAJ65=MfwV-uer-rdwpTt`p`h6M8 zKGN#f;FKn#j7GuNy`E$wnPKpBoNNZQOgdl>K#n_maGGK?WQ!fySI%e+;iElPNED!+ zjRx{R3?P!{L1&cKGK#*Ld0R&`D|bo2^5oG5(`Y>49XbLX?n|)!&n{+WCM<6QKbI`9 zGCx$ju)`&~12ZNn>T?S74Z=i-lF$|e8 z-kAyp7XaQCpt^?LouV!#v&biF6i27I3W~}zD3FTf$iNbRNwx^SRwVnxZpRZc{1F|v zm^{VXm$SaF@^Sx1G8(goT=U58eEzjpRZ<`7NU0H?j3s`2t=FozEYQYTE_0 z`vj>wswed3_@{WTGZIl+^2ZOtjr){J=(!U*Na;vo#-BdY{P#r8^}t=px7iS8Z2rrN z3D)N={=ae$Len!WlUqID7Q0a%V5}`*8p%EsfNDb$r!NIEu!L_hf8LCMqXs9ba|v5@ z>q--4aUv^(07wJ90+PJBZ(3If3gKuGB|*Pr?tf;BTQljpas7L`Imxyz2G{<$mr!)i z>1)QAzZbU>t_=jwt#z83c!@1bTm}9EL(0rdKbk%9^p&FLRpQkBe;B)mHeqxyFvDZp zwr$(CZQHhO+qP}nwmtKW+0#X_=%U%>ALL5j+)wjIVA@q($*226LY_fo@g2J-9CzuA zm-J0FE68An}tem@v z84(L17z~z+(LmLAhQ2KA*hiNB_r<1A^dvAVXI3Bo?8Kp!|w6~5Ig7H(`EG#qLZ-aM3w`u4b zf{ei@oR!6(1NN)|;!%-AeSDguVC;GvX2kXOgym8>TC6OPxvOI+YN$O&neoCckR3b1 zyOg4LO0mA11&z1#9r1d8TI}_vU2~9*@qaw)huVIJS^7mN5Z_xkcLoH;k}y$Jh;;^> zcPssYwa39;F#`EH-1Zch{hi)ic6gh)H)(dtj_uNy4Qm>tuYIESEVKQts|!|z1&&h8LrWo1Yz9gpU#x-IzkI}-%9&xG5@A1ZF`8Fq<%q+2ww!gYPiIVPQ7cq` z(ki}uynZgx3HDm}rzc2t>mJ~93v+=Q9Z=_V+DmRd2q}6Xoy|)4)DlDU$Cg692hh*v zN_E72fDw&qi3*(-KFeK;dhZI>;Ru=UHEW0-UEA=ls)*yFlww(3n@6228zE^hq(ikR zlx$ZbRX~=v%KnSXl7mA@M#=Cv-yVZUb}q@flNwrYrKhrNB>1kTg}U`Ql%XLzkX!$6 zwRW8fS;~30f(8}JK$5n+9Gnk^vOKumE38>j$H^m~7eP-S`q>@2uhv8N%adN|?qT#w zEj#As+$EU(4OHJ&)6`SCp>mEYJBKq^FT>f&31{g+Yp!BEtSiVx(F^D*C~g^F`}%m} zhjrLi`qvex{>K_5x1Y*#TH_!M%C+tNTj+m%i|xgh52uzZmdq=RX2uq=gb?dH1pBc) z#aqoQ`++SrEv;t=RN5}!y5c#4vg!z1reP>~Zl9>k>n6VPlww?DHt2$|#ME{kXBIHE zC78*dw&u%ttmgdJw36D$haH(RDfeGb9kL|Y%Df5{5NIpd#$~4I#bUjnsTB|>6qTh` zPl=R@$K)O=6w-Dm?6`C6lk90*MkuSCc~Em*NPdreQgheo|BtBH1<~VCE^x{p0e+AEo0GjJLD5CR1(!cCl7z= z`i4z4v*c@Q&!9kLLv$WNQE;sI1WJt=LaQ1^XtAK=WMX9g=V5b8vNY0 zqQsA_Q|1o)o;NI95OXtD0reP1f9)kRyw`5W6w9w(luI7|woh`jcxeYcu=kj?Tzy(* zIXOaJD(1fDNiTah1)E>EDT#tDAa8ITiw63pLR`tYB#JHaG(*wFVJr9|5bUtJdN?D7 zg^=zU%K{Pu9@u29yM?#*n7(&E6e21PJ&AtFwR+Q_zTQMolB$6heUt?7Ffflk!Vy{nOJPs1dBWl$NPIT;u zhMNd122iz7}4forRxDF%?5*1}@&U z2hruqKjYBnLMyAN7hm==hF7X)ituZl0+mtNw8^K*%A7%8=RBh&#_xV@GLQC7V;RzE zf^C89?axJQz+AG4mP2F2m?~&PtIE(bl!H-K>G2T3yI!8vUf+*IkUN@tum2^pbH_ue z@1>pjHTIiL!xZ85`AA%2Fj2R>=gyuo&R+8ZwqJG2q^|aS?Lv=hS0dzNU!vtET8>(* zlIvYb9&ecGEfXibQHL;{1WC{5i4f>Z9`n?O6ppVwJ_o%(`b-Ky!(M6PU$Zljo|F{UIBgRAMo0+(;=4s2<%D-m;gFxO zuL!vZ39@TT(j>`wL=@~*&acqd9Aa+=^O8X*cEsqW7)Cl)Njm*Bu8uQk--5Oplc3$z zEfX)=ID7z(BCA^_Ptqs%T;<1mWYRiGQG_YmmLv^bAW*-Um-OL~aWQh`)R7AIC_b53 z@~0l*lU`Z1>LI~D;|~@{8~ijG&jo}bK7HMAPxd|o^R~+_Hr!k2Cup$7r7Y@)@f}J461Ep$9r9nYZ8x?Q4DM0zh6#OyPPjX#lI0THQPwp!@C)NWoUcIq1lky+jHlU`Q#t&t2gHWr44EhOcqs+jW$-18H-z%D5RA$eJPCXyJ?8> zU!0tZh&i`S{`dWapfjLfV%TfD5&&gwM*(3E6ZAbBy*tD3|l(Q_)-z~`6STBKjxKE_^$9(rL!M3%K z5C>CI6RKyNw2QK^BnY{NfHcp#$fQfNG*D{g1kqG|gRiwVR-atkaA$sZ%}a}i1C325 zh*in;77dVe;vH$E4CHCP^x1x@8B#8ByHrKLM&^q_#nP zb~s-?ZXJS!rYeMbKqvEuFBWI^3IvW7yMG`h>f>|FG_9jHnxsE5v^_6Og(o8`ZsP5; z@WDUxrm<~>pSY_l)$6RVJ@Bt>y5>a<|HkdqK zd(0JA;3>i%52y*jO7R;Q#w;wR0byd|kEJ>@D1WCBxpZi}BCB1pO&1JJrKk-_>}owCQLySi>|^9fo(Ndo&@8qvlbK> zYA8TD?dm$t&V+wsj#?=w25L*n+i4!ly)>D>UG0zm7RPwK6i#^Lx?8j3@a)=bw(YiU-K^cPnf-ra zLYu8^+xCrD9(%w0KEKi24ClX0#|zxYR8oXPqp4PZ?{kZ~x(=R4TD)&;`((ncd4%C~;}PhRR{!olamV^)>cc)`iSV9;tG zYObxKDA=Vw)9}w0OQ_qtdb7>lte@`>`xN_o)4x1%W5+UTxgB^!SQj*2x^ zl#rUiqvBR zc`4(y+GSR%tkWVPM&jlg*A(-dgXxJhq%00@wv{>*5zdv@tCrIW zGSPhLQ61qJuS4OP74*xkU=@s2e~y52lf4M@`$osz`*&pyvtuw+s{(#T;pBU@?zEE9m zYnojmNI?zS%P`TzvXq6th1HDhgbrX;b9NcQ=XkwV0tm(^86OR@*ViZRydB%1wJ9sB zld%k*sSCN!dq<8uQedxyKV(RPKksvaQ{l9O<-A>-8%E#}&l3BW!9+3`2*U2Cj2Op8 z{~B&mfE(Faalcn&!k)lzv$GLbD5-S*HDVFsySHO1(k9sKww>Pit z+E+I2HOVD(ZnV+ul&X1gXVF_w=Rnr~;)O>s30sYhb-bu`kzisDdL^o*jsxzxll{i^ zB}KU99?O0^hlvOGUXP9e1ru%X92n|9U|E74BQRkx9;+)i4;xP7UNGp~0qdFhIAzCVJOJzRD=YE+bvNgy_gQMJQ3^%AuT|B{t zUk)AJC?ab>axF4%29R?OV3YF^w#{2^NrRLTdx}%aa{GqL^&sg)+VNx9<>4NmzIu^1 zON)bZq?**~e052PQE^7E!#R-Y@$uOnusV>2O^~WA4uTIkfaBoHyWd3wl%VA=a77zO~yn#)=&?i?w z+zBq8ygD;hPfGhjn04aKoWtQMw@*ha3{jTvFAgNj;edzAglpYDo`60rSH8cVhLEIw z{FkakT)D}EbmPCGzt|FMo*AcvFwJBeS$ma;l#KJ!Ud`XR?CtFn{l$?pF`4lQq#?J& z+`vBa#wF})@>p4^cTh@XdUcGg~LI6-I)+HL1Z8vmsla zw!wq{kyQ(KjyswYn=*43ClIxAb!`YpoF#5kU}-9>Tn9`%xBu`aIvr#(0wax%WJ&Qd z{$#{HWcvamnt={I4Vtsm&j1u5UHs{RHM|S%d}XFTG{FSbW?7l>ax#P zL_I1KK6xu%E2*L^c_YN9pAc>#-H4Xt;ZUPu_P`SQYwEJMLaTHqp?yzKL!5;jpe;Yh z?$f*M79OIg$xJsDxjX9Jl>!c}Mkmr|)KpYdM~)`Dd0rtKAbV|ZmqdD?7_eI$NR-c; z_wX}H9d-?(>d}mbrAMNY>qpY7Gc&8(xtm8GP)c__ScPG8H+^Bjtrkl#-Zn7(GnZn@ zI)In>ZGgQNqnr|2UAalKJc_1CBfi!md1p&i+B!^A)pn6GI*BMnI8~mQJgmDH6-tiq zA5-Aahx@DGk#Tp8WB9wPab2$lS~RNI)Adp+O)TdHjP$eEQKH+$jAA}T+*qmW3RODl zi*8;^A9jfP9Yl~eME++KV1uz}dmQ0ZVbSy})ne#3C= zGkmKiVNck2tL)Md&zAZ$YliIdjyT{>miW5OlUcy>D`f|%L?6S7#a%6mbPHL-Yh@{y zHQ-}RLjvkt6OMZAg-5``BcT!oZr>9x1gw4O+wi(y{@|;uJEm@ z6&VA-9HPk!+*IUCdA)1A`wyK4aYYCG#X(W$!`PiWFyB3?gr7H-PDT;ht)S{b?x3#)|#g{KurkD{bsdG=Hsx=ZG@Pq${XWs2f1|ZtNuE1Us zmVzg$JIK+|ld>!w z1HZ|*KA#gGm7fQJM_DFEyd0aOrhcXYnx(@MNrC&CO5;Ao@r)9}IWO^R`hB|g4DY{$ zjB2*?_JvDR!MRF{KDt-|a;lTUf8$h>`Z~ELyQlh`yHJ2*Fwcqc->|QFKtvh+Pi?|_ zh`txfA};i9&l!zX9Ai+Pz<(dvql7vm`@^`gcKg-&jv8t3+E>jiE~ly1*F6HaaYFUG zFVe{yQM-|68A>L5z=|m|IrOrVo|5-`5#=}36_~#{OcCaB(&)ZYGw7^Rfha2tXKvjP z7qi=Bb#hT;Rnf2^t`)t&yKw1g_ah9|Uvh)HrhZ(8v8ekV*GPi&cVh? z`m4hx4raE!Nf(T9GP5z!sr4bj;W;b@_W3k!)r6`!{-0`D1e%2o- z%6YVvc|dl;^i%>Et`sE@J*s`KL5r)Xu7dwQtORwN*yCT(zv2CyzYMTostdQ87-c{!vM&uE9-vI)k)iO$GL1$0#kekI^X_VeGW-1C})NghmwSbo`l zlj0Vj&1OMfKHqHjiXRUxRQX49L|m~maLVgSnQy~e z$@DLHy{DtMS~0`SOfY1mhzsa*#vV_1SDq`D#AHz6!UoPrZ{@)#DUVhW=hWFA*s+(X$-l?WU2-a0bk|znPVS38e`UzH5K|hlFX5?Q zvWXNj_+r%@peP$|yK7ZO2Zr+ouDHFqw}VA3SWGBK=Ko3G>f?B|yE&djc=R;=>Q`x| z&rW<~WXf=i4s^jIruOf}?QF%kv}tvvC&>@PMhEV7i5bD*eOUzo2?HN2=rUR@q?&8`A39gz$Uu^pxk#@UBI8j6~eFSLffYuLUO} z#LNse7v&9>+R=~Ydt@`TX&*1;?=aLt8@Su!PIHJA>)Ztyhgt5*Rww?t@R77&2LvF# zCaK6ZK+ayMG(x|7lc1K8Js&U5xPYJc7nuM&A&~Rok_74UA3@_eO?kmJ;<(CzBDXJ6 z)jK~GIW$(V7lD_7+;p6q2p2-!Y#x?0Ny=q`1;bYf)MlkMnNUw3@_?81VZO*x<9~1D zy40c%E*ANd#B7TDtYLNL=QJb@QyiE<$ep;FX<8YFR8ke=Q&DRfXSA6s`btL*KD9MiFy`s&)(A_Kxc-v9@DYRzkl<)T7r zNzw&qBbSHP{rX3)f4S+w%Zd#mPml=g%3Se4N@D>iVWdQj32?c2;>4KA@)0PVZ_t=j^av%LGVu zAxodpD^bMhEx&xUG#^H^t%k!OSk0R-^KSZvR#24^4g>I-K7xhUFbyOE z(0c3(KAGZp9*owM!H!pDMmTo*Mvh#uCSQhR>GKD$l%voM{CB%7On3HK5Nptc`)K$z zd#8*7zHqj4$@hB16=QEB7{U^Wj&1%KrKUMg=?n7PYsiA)Is2e(Ya5LTX--kqLNrYI z@T^8;uc(I zR`AlELTucr(++%|PLMJuYcByweQ^MvK9-#GuU>>vz6!x>Nk?GCB4U$3`%N^mX4NJD ziBXi|2s$?1-og9+~%(+)(5n1bL?5_7w>He^A72EoYhp6yE{8_!u z=Yuq@x;g9$50fX`+ej75=)1WP`sIPvKTtFh?>|@V<|J)vS_xP0+lZ?QS}e*Zu?JQ_ zb?vxunTstVIOiKMbaGP*;_^>K6l_yPRS zX)Er*vJn;&j^7n+UvW8kbQ^TOY6x+_?KuL?X7X6-yB!{v1GTm&(H9YA1Fd0i_}2Nf z1Igi^!iV*ZY1{h>vYCM_%Jn+U6W75wVwy-F>(H;5ZOw4!7L_oNrL(l++W(Bcq`s8* zBl6%OY3g5LhQyMP9B{x*pS93qCs7VhzC^Qg$BTp*4gZT9ol<;;(sl%#0>@xSanCEF z%5Y+(bN}k;Y9XS1b2p)ofps}J*pxx}QOXR-g*MUby2S)LDrAbg)u`0xPOABnj~n(m|V&yP1n z#nB-XcNHTB7k}lkv}@@Uz>7%>d0kmOR}Q9N!f~F=Tfd_e>yMYU>GWayYceN!*0g?d ztc=&wleU7Ac#INSu1|{lxpTt{UEXDAk0}`KzII@2Js05oXl6L2`k>W}-;nt-2Zkvg z+K0?|x-Qf*L2{Fi@e_FNLG3nPh^34|tXb>d?DrwXR;|gLp6gMx;B@7mG@&xWb9T05 zav&WA_u*<)cL7@_%54#fD)&P2bx;yjO_tPNLJ1k(ztfHYergKM1j_O7Gx%?wDZ#oF$Lk!0|9Hi#!-P2)9~x{3 zl*0An2yk#xyD@H%Nb9)4r$KaPBp*!E30aJS4Th&n)@#hLA zHguUHD3X*f3dyFtJ;iO@BBNbIQX)2H24GV(tUbaBBfjy;s$sR`N_8Ydgnd?RIi@h1 zt1wUOE3gJXt};k#NtHjSZ~r$S_2>km`ob^6t}Cc5VjIelb0-Um^`DljooFE{NgPjg}aNZ^eBaiS7qoyO$ErkfGM&kQo-!=7_hz?n{$I{#ozt$uj?h zE!c(u^fs&ErMXZ9baIs8Te^s-{N-RXQ@T>ece8e_*F2W7m32VlKDrUl0yzTr)Vn56 z`krF^VGl=yz!6K`g(3m z6b!_SBK>j&rp%IStaYI_%B5C*wBc`bI1xOd^At7^0tnWUZ~GwIK7(2`bxE{`hVlM~ zF1KU^Y6}> zU?sTUi?&jT_#07z8g8%ufH{Pc@hcpiCy0>MP_zzG%ZL^RD0~xwfDI8B6Lf@uX3cnZ z`-)R!y-lRRI%K*rt)KJ!{AWm&G{&lXCorPae7a$AaG?|mr=s7NMyn!VShF%lIW81v zZds=f#c9&<&8wW#wUQq=YQPf_b&^y<7vG`tO$PH(sCIOGu1W$OSV)%a-Oh91kL649 z?T;R$OZE8Z!eGl+pVNox&}Xqwk3kqi!U4Fd^l<0`ye)J>(3#+X_n{IGT+JLXxL$6q z(uuRJfUw>DaAGd2d>nycU+uQ~HhpBZ;^NlOqR*AYT3*j*M!(J%a?j~7hn?H({Tr*1 zp!RmHJYP>VRf=0&JW1W8Q%$*fZ2fpuNOGiJ+{*B5H@_l-XXo(w)wdfHfEB# zoD(P8?a%2@JYyE87>aTxXeZKlc_8Y4Jgy8Ldo+sgXP9>Hs?^5#*YOj(6Pva)S+3(i z-$s1`2(F8YN3Xh=plaqybMDHi)cpvlSY+n3J|Zk5enoj~dCI{cq7Z)3!xxG;ONG14 zITL8^3cZE{u6bG)E^c9d3n9c?9!;2Wbnz?*q$uO<0O=rnC`f<^NQIT!>9Bj z<-|5hV?x2#pE~kt98T~b=6>{aET^-74>?!qu_l$&D?~wvcuV7Q3RJX`@g97$q8S6{ zv+iw}t24>*~LlbQYhz{$)EOsxNV`+xV7nHf2m z82-QEWK#{D%{A62Z3hUtP%!uO10)^}kno}1U0x9P{}zOtgFAT#NYJ(VC0}_m-~Ep_ zJ|0C`Jm;B?_iYh@qCycmQxhXV)V2oq5;HTyBY=sCscnsIM&<=-Aemg+=PLjj7#SHG z2ow~ecR^oqW^8LkA8`Wsz@-IL0=o-HXb?|EMh7AV0Fr=p0q=wb2!jLg1Q6~^>aI=T z3;@+nKL{u`xVRUvFL(l`g{_epYzbIs#KuO~W_D%<*YDJzl>8s{a=nEiSU^f&VQ6f0 za44q0{uTtmD8V#9LbD?vC?YAT07+0zRZ&tZFa;u0b8KvF zZ~lTsMI<#vgWw3LNXluz0IHS%NmNo&{QoDP9Q=^I0CTmlR{t;u)iQsIQGh>=8p=Xkjcfh^~Au`)zQI_!NJMFw6T%R zu%Y!ED!;ih19)I$Y!3+Fy9G=e=#O}8Vhd8VyomDd9;7b}IFW$~RErzXA2ee7%cd1r z0RkHM9QOwRjClA?U4FWM2LMj+pl>Skg8Lhls;H0vsBxLC$qiKdQVS~>7xx(_2M5sG z?^2*&u-M%X2n0wZ_xjpzHSxP{@+Y79CxqSx=vroad02Y?)Q`N}?&jq17o6?iZfjwD zYjAdO@$g?bd0=A(=^eUvxIQye^RF}+HR+3huB=3c9$1UV+Dj7~fqM-mt}f2s^KXir zgbG+oRE`b+8X7tPvZcgmO^k?5O^sj$4x;}8nHBi)#KQP~?A`x$YiMh1bnE!acwd=*o>KahY5rf+d0@Kv^!Mv&S=euPaw%bFU$_PH65to{2uO~G2sO^siz#(g{j zl>TmzS}fuReNj%o&D)%3+CK|+gt;4dL>A*{eh~!henO{!t+sDh2>)%togYRMvxMRMotpxk zwSwWVYzhThYjWyfeR-i|eg=l%iH46@q?F~ecnUK!1@pH4*#fM#&jRdw{bf_*-}u-4 zW>Zi>fZ7Xv_Xqwu9RJH18$bC3*pZUj(q;Tu#$Xn={^bCR7Uw2FT`w?ym>A5GpI~b< z1^0gDhTtv-0Ikd)@AM7-4}jc13;=G5$l%OtdkTUPxU^5WGJg^+t;uD`+{fNmehE`dp1zl1?mT|b0DVn@H& z%oNGo3%~Xte|}Wd-(Hj-*L|j+KN-LC+gtigNyWXfpbE1y5YJ_w;n6jG7b#tSMnJl0 zGX%UB=ly?{f7Jk^_ujwp7KVn{>hNA{2&&ID1g{*Lnf|#FaP|J7Sie7e&+ELfU2?zF ze_ob=0D*M^!Xc%e*ur}P|4POVkn9_KCY3Cp?81_rYTmAt7!{-|lW*TSiZ-`P1Ok2f z&GuQQe6TSk@oeuw_^*XrqXyd3=&xxXQIXZf@Il?5drf;E6GubDnp7}VzL!nVeWkf2 zE8$C5wH`rL@(Z3v{8(GP4{O<4MxyFYs_XwFhE}VU>`Op&W3jEG8m$A-#cx|V%!}95#T-1#+cm|p zA*%lX^w3ll;s61T%UPbuZxSWCm=qdA_o)s~s^~(=GdpDh#=9aJ$(D|IdF1(l#BpP` zdBOgl&=>AKpqqOZ8H}lC0+7=IJ7H; zuSj?Z_BQV!cJL z@`G5B>dLX1Vc;?471AC`{QgM>2FgS}Je5l!!f7wQ{5K+(+HtQi34FTcDCv6C?ct!EH@Nd%Mh{meuz_9U=>4?XLpx$MtC6-a2-uVKIALCYXRk=hU-xA#2TQteu)59h1)m0|&oCZWqZ6FQs2~^m4}!(TM!*x6 ziTI90vo8xE~nX-7Pk*Y~*ADk-PKcCcMtK z-2X5;k*Xrp)C#M`b{(U$mXvw0wNhJFt%KUNDLwH38h)~Oc($@uJw;|l{KLgUu&*zp zmWR(QiyV*HfkZNayvwVOuFm7LLqQ;@qarb5kLuNd%7^n)IqK+|Nr*}1aWwdk4U;)X zgKbnuRwFQY6(lpOlZe0C6hZUSqN2a)3&3t!JI`^GdRfje$nUIpw~Ns!+_0SzsAUf6 zoUgnP&F-&gS*OAeS>&0caKcS73LeD|0w)VIkwki)DtCn z8&TT+2+d5J30{ZhqgyX&$+OiRbM_^nf--ey)lH=OOZ#cLkH-Ip5$chMDYiS*B{#E` z3=`J@!d#VRqN(Eq_$-y|C-WjZWcRH2gnLToW|P5PO2uX-AQHk4!(m=%P(7%K1@=HT z2gh)DiY$lAM?klQ&AHH<3sH(FWTrd#YTM}PHdQ?W&!~^DNH!j`Sjl&P7hHrSUlW`J0z$(-F-G zYL#L_+r*(94UeKaug3VNIGk^YRh&R0KTnSflhd1Tu41F@X;!nS*JUKJMGPlOr-pYAmtPA5L|90@6ZDERfLc+OY&;uC*#qe zKLut!!j}W6*Q^BY?xZ&SOJ1i}k6&yh{MP+xO|El9M^|`dH3<(r6=b(6Dh_hpucUR= zUsCe<&i`~GUE;=KL!c8P2{~=TLy+s=*NM+sB;TF@dad{zY(s%2GTXn0g=6p3^yyhu z@_AoVqkr(2cfJm;m{p&9{n-BHiMk3bC$|*!El%>C&Bs?i@)cBcC7wF& z%Q!EuyHKcLvnK&2T5u3Een#=`bJhBG&ReHz*7(m3Y?6<_p`avv9A8oPY*wv(*{A#@ za{=VIo2S+erUuK>!9#5L?$C_HSx>sr+sj_`qZ26M63Zd4eWI2VaF@KcYJW`l=z%J= z^#4+KK8P0Zl=@vALGNo_!8!o}dO^5B3ipJEC9Bh|m3;}uq|;RXa^7}Vb62=|-4Z?A zhwM?lP^>;YyMBXzKa4%oglX(V?sZQoxdQx1^d#|a8iN@n!^sz$$D!WjzxtL7tEf;| zi+&~?VQt!+j#!lLDOEGLCmJ{B)kBMPuY@o&>6a@VmZ=t037$aqfTc_Ng2bJ%(WYG^ zmKNYwlQX@p7)*{0a>=B(zubLPx2DLc2;Mc`)*4?4cj{PeSf{|Imywf>p%5TCv54aY z9fSD-C7+P_d}nBJ35aV^V6EwxM9`Oy^pg+ucHYshtFAMo6PBi-=hLWy>Qf&rf#pY3 z;`hxC)^_*^MI2{jTYW4@;9u zqaF@ac%rC!st^abLznUSmVTJ z`%R5u0Kbw-?oUP%@mVzzX?g_3A;zU0}*A6 z#lKFw-lbaN6CgD)B`5#O@M1}omurk%e}czuvW?F*L9lU=%S}%go!K3=OORGH$}K~z zLSstF2Wce*n6V@GAb2xvw9bP+4NxbRMF^_Syj-_#{=?2|;~;4y$#Wh7udiHppE+LH z$ePfTBpf>aaz4%yLgiB`xl!I9Tfi(pQ0VKy9}Z_TzjsBwi`MO8pJj%{KZ{R2JMRWF zj=3yd&@BpXjX=S*2a|#=OXH8_(IKQC?)y?AP;pi?t6>RN%1Th|41cIwnufbxyQ^9h z4p-@6VIBzo0y2*DC0%YjSD?FDWRw*N^z6>>~t`x^WhooBb=6&Y#_ls;tLwokX z*6vl~k9usuO`kZ}3P&*=_F7h$U!vJ_j2LZHborAI;0i0Q?R!9R);6~HM-uzcSi;|e z&y+_+2+K(P%gwHo4YgNnNmT4o)W)yZ_Jxz011mrs$F+ID$lr1PU|g$C-Z z3|v!WRh%_6WHbma&OD0!lSI7r7A=0;erlsn;5jI#iD|{hddCIOKV6oBn;K68;iQZOl?2K zq52#Kb=+;K9V_j6mf^^qqlq9<59Bc%w}4AyHb)D92{DQ}P;VKphr-5-w}t`BQ>6Iu#1_;-2rr z%~t@3^u7BL9sOt-eqPqO2yxVO*{St;^LPwme9@HT)_xZ{$A`w}JdD7BPHIo0G^dQe ziQXNr#~GG%^N2l4zAc-RbWhb7w23$1GxSXrmEKv=K~1cq4BKgMUQ^Wu^FxZPJ_)ax z&Sde@*z#@qV1={o01wbIR|tsU6-W2t7U*Ohzb%x~+?szpVW@?gC8>mNxHk{i33^PP z+7q@zC<;10nU1b;d3pdIE0hgr7=^X93hS~Aw`bK;iy;!838!+_O7O#J{e!z5YccG~1JnVxGaDU#vq0ovp%$O5`*kd^(Mc1bgyW z=o~`yA|-&t3$x3*pLDoiW%X)AtT?$#<~}qL)!HLe_{lc!Tcd zlys7O=d}xFf4>XIj0#}*zyV!p#T>q4@FmGk^{ze4PRA~Ii+yk+?zdQ^+V`>7=dDiK zsOsmg%aN;{a&}5dcRRGgY*sF9`u(P~v@sXDpFdw=#^?l{%|4gDc+S*^Zhx*e`>!d4 z<90=wblJs(+e)X$b)?^mLjdl#D@L_25n+(+uI6*YAmZzN3Z2_Nt!%eT#D#p&Fy*tB z*V6s(uB0E!Eb%r2h3~O!e*f=l{)tE+W8|M3?gc zR@KC0b;Ru2q5u{IJ{yMDEWOZq9oibB+LNqqyQ3<`h*TA&$>wn zBEkzptJc9Y&zjFWdEWKFhr*zJSgzjhSIS$z&aR+7-6ubOgQepLc6R{YNuX*s_VH;NK*yI@a|MS?kQfP@Tb2#plNe6~N(2yd!~+=od%wwA1~f2F{Bj1P3r6ITm8L z=b0teN2%mI@h0g$Uv#sZK$!DnvBT6KX5{mIj^D|$5G;Ja-C1S%y{t!s$W z)>*9x{Ju4Bm@}>#^+ywy?uu+QToY>dzY3iP_y~$Z0&FQ_uvMC+cfh0uuceZMhfBXl zT4WR}+oANt+?X16C41@$8X_%t)AkW6ESA+FJ&7;XJD=@`}Gt$piRtP>#LT`WdG!(NxWziHltMl%x zpAJa2o@Occ&L;Kv;<;Z|IkAFTTGiiR#ym*tYag$u>RiPKMWewUDKS4cj*iEOFI36% zuekrw1Zqi#FY|^ercDQh7>rZ-+zvmgEn4R)-JOf=Vy>||519k@NkamkVG)sCh4 z=|thGt0acrVwTRB-fbA&PQEaGfBgWhguY5{-|*v;X_#L_ah;gLrl&g?pe%>^12XL* zr$qmxu)GeNTLkuOeF?+Iw(>NFKr&$x+vLNi6Xp@HbYOUoy7dUaMW^LgC$`YZz1!$1M4%SzCP zZeH2bD8-SadgSOFSS``q^KK>U4XT<){stT7n00l|=4x;6SN7>U<#`o0VIl;Wh+ngn z;RO0guB*MY5QrSv{oJ+cpiZr>+_@HPmr2VMyy@q7m(Vkv3vo{gEaILh}+;Op;SZLwiy&Tr`Y-s@!MtR!uCd!U{sI3kIhfv4nb zoO(iO#9doss}%cO1qj((rZZx9T6$=ZC9c&|rRjS|!^^b5I~Y`WMUx_+)l%FbE$`WM zf>XLOh%U|liZK$#Z?FELL`3pSQ!fKL>^`ORvP#3(sLuYT4Qh3BLhjsLf+L1;x@|uQ z{Lz*k=EhcokZotCdiYHlwil<4lNdxiNZJ$sZIP=bt6;@|P6`=(eO`6I9@W?W{p!JlcQ z&NG@p>49(Nkid1f_V3|*OXC|`DisK3d#N`rn2_V+3Ei7I%)+oC;r2%Hbtniz1g_3@ zKda`R8Curu&cF-G_yRelu^Cm{|HgTv6Q8`9MB0+JE?G5JdB=`WmX*Uj57o@WUDOJk z@v!u&8b?EU1LhkdT_bidJu%NGpiz?;=1}g+u6~YIr~oGbS#}JL|G_{~=r^X+?;?20 z^b}D3##`RuUnwndhOO$;na;53KXb@nFrPX#>D1qrpDH8oF@C<0ZssrO%I`@A?%<)(Uvow3T_l>}V1$M;fWx*9JMk>AF$LUnVjLu5@(c7C0+d<9Uga{kV&ty$*OPku5bQ6Q?yeO_bSG(wAjgvw}VPwG0IYj|`+|Uyh6k3-=L4Zg z&qQIcQIJcC`hnE-BZ=;1BZuWGe^7Rpc;G9}AskJeNxH_UuAK%rI z(Ks5j8NT;>lElrS^$gZS{SN>}K)Ju}OuAbxX0fixt^bHg;jz5VybN)F0B;L{5+Rcj zf6}5%bp#o_k>-e*uB#|KGeW_~Q7kjWxnMz?y;?Og8Xa6F3%{dLPjD-B`pKyqHR$kV z)ax3oe-ZnF{-sJCJMQb0>%LUUTGH(5+d2A5ql*9{#!;Q#x%}&%@nW0QjmmGdpM4_B zIrx#B9FVMzSG$66zBoZNU%Zlzrk0XJ))%$zEp= ze|8V3AEtk zaIY^zpqLKO?~D14(qDV?oo~xl)UPf!bcDOMMyj`3h&7VggY+1n3dDU@}<7 zWGL%yRF>o6{bMCb3@OWfoHw1m@cfQX4udZHMzBk_ibtPd`E~JF&*v~yYn$&Q>YSyG zn11y4CBd>Am|5IJH|%@q@B&r_Y(Mh?8a8&^{gUEILJzTDf5cy)XG`HDhw_w z4O1oHjFm9j$#ywWxBIC$dXBfYv4oQc2Hi`H?TW=};aK%Kg}Jl4dyV*u1B_rc^W46L zra~S0YeeI;n-IK>4;_y{w}16UYz?+9M*5f;9eUM;LQ$HMi$KR~57Gm~YPOGB!v&Un z8ELj-U2TcuLGlglOY{Xxo;UqBd66yt+Os`>-=+8DKG2V0e4W}s(asH}2Z>2rNOc~j zC&fG1{hTpdNeiaoeGhM5a$A=yEdf*L#60o=0=&|2o=4PX32NPj#y z(F$0cY5PGF=Y80e^H+}?`T1-GwQ!&|lH6h4oT7HA7s5V9qeSzW=Qc=lw1r>BV$6I1 zZr+5ey%8UBZyntbyL^d!u8_DFV!q~v9?ju#`7UFQe5_rZ{*yFvXsB9TbNY4(1 zlo+)lLiPbw4upnry{#I4TW}ojW;-f0|BI+CJF=%_>uNfIN>WT#N0djw1~OIes=&N{ z8}0l!^3(Ch-T^j_HSxxoyrasM5=yLPm0lu&ePw+5rs3~Km{hD5%@O76ZQni|3o$Rb zS?Q!&eoLNdEoCuxb>4V81MG(V?-2p=6SOhJL{AzN@R=H zhi9g00}PowS9s#m=pJ2w>Jrv6;>?VKooc9p+N28TBGVdit4~v0S~%VnGbn5Cu)Y&zp4Dq>>9TTiEN8 ztpz-Js;52{Jo-7GvrUhc7MMk_+9~`pEUXD`{aFhlVv$f@=O6W`#GPCBNA2!z*BNHw ztGYh%GE_G6VU3*WP|G0oG>VO()WRC37wf!9Z{w~;6-1?dU<%y8Gh5{C$NrIy;-x3k zEdE7NJsB@-<|s0p`HI4wIvzcb=T2}0s9bcnZLGQ$&SA{AcDazlG`$&m#P`LHBpVKi zA;!CRH(aca3x?x|4gcjtCEv{AbWflS}#PKEFJH{>GKU@={L-S9QKUXXIK!VK&ao2fa5`d+SznhOjOyQL5t=Nyb!OjQPqR z+&v*3V!#MwEc9dQILhTem2uN)(oGu@voPjXu%VErY3sCE z#FIJbp`rXn(Cb4T(ZWZ9(m+(I)#-AwH`JQMSPC_1yJDd+#u`0cD2)k;k$eKm(xLhn zEv0t?+_6pM&u@AJ`$zzqaeA6Ryf#&6mfvHxKh_NY{J0JUNr{7=cly0ckkWImZkKf+ z$5YtVC#lOY2heN7O3m5&dIuSX+Q;waIB?eWdVu=Fc@lP0o8#TVvGcsn$Hk16H-;%e z<(3rbi8hB4D_7FyRV3HZZK|t8Xy|!?X>iSX{o9aB^A8IIPgfjRYJ@@m?SAd2VV4Kj2v<3L1TwAqD zG^Vj{h0~FQo%Ter-W6cIboX#mU4$YY;^zF7y2l4Hw8>EU;Ky6&TyCL%C%JTrIMa1JCvY~23m`J2ei3zEAyT@9YMd`9G&O zAZ+rSCD3=_J!Nck!$a-BLoR4df3(FXXY}6&Xg>a&ZRe?!cQn1>`H}QxLxY*PLyJa4 zF*QsQ{8V+G)d$wi#r!3xGh9Q_I6nL87}|u*pV4mM`7)}&r7jwh%g2o>-T3_mDMZTN zj;mk%y@AGEw)ec(_-D;;U`H|t>nf>woo6@1_71gNtj2WEJ$%}4=J*OtpY2yP|B#ySTP{m8G=h`X4 zzi=YFh-EHx(pQt;=LT$%z00+E$O=&akm0!{O;Bpe%@?#FHzCtD4fgPQ1gufe&6qA8 zbFW&-%jBt?)QC2ie&;xG$@UgVu|Iw&Y7OI>rYy1mC(K$i?vYUv&cUR=t@ahbQCs2B zpL5+E1eX(UuH|?VQzYZ{6-_iYU(Q~XZa25pOm}GjOdhz%Ds0?I+5o~JSzrlRe6Q`A z#E?D!yo6^?EX6vli0iH`qXLU^_(t=4$EtQ(P2%u0;GrobX9^Mv2L{SqzFB%GMl#?y zGkP+`k4|o_!Jb#Obd#nj9_xf2@|4P>4+u|Qw;T%RF5di9FF2G{p1c?iXL!oV!N%a9 zNgF^hb%4o%RzR3N z9=&|YG_BTnv=&sie(UvVd9y|b~( zt05&rL>U#km0<=4#Us=M>eADEh-{NCyJx2gUlBz^!CT9#ifWH7kxFMx!IRRi7R93* z>$w?=<+9sCmo|Fq9 zai?%Ww%pn6{_!!)rXFvbbR*nk%vY!XBRP^s3UtKh$j6c!i_Y;JPF6?{Ci}j~knlu- zx^=YMjoVqz#6Ekf6d07DdrSfkjn8|7DdkAE)63UJur9q&&J$_Vx;zZq>h6;>Gx;Hy zMA1k5_3|DB>p0d$se9RYblhQv(UcNa8DozUzMp*0j@-Cyw~z*U zZL>q-NcDFgqKaSCA80JflvAaPn%_y8p{Qo31|HXp&wbS5xb&nFUX3YF^VPcx*RV0|K0P=(KT=ZN!$;ntkt{m7#N&Aok~ajF##ARg-+<$2z9NnQ zgFI>P@PgUmp`ta0eO}6IeJ~7H+sUYe!9dfHP@JRI)I*S4g}E{`5V6z|-+}j*&w%wf zne>Ch7+mGfS>C)QUsCFo#awG(K!Z3g>uP3UrY;p%CJ&FE`?T)eU_T?6Vj+;hvT-$! zg2bS@Fp87OYN(Wtgnpd7S++6!J_0jAv-1-T%Y@7iX^@Hxh2x`ouRZ&aCDB+`neS7 zaZd8TZMhAP8IYcrIo% zJRVi_G`@~M+pKC%846xRPJ=ozX-A7DT`wWg6LW1hzj+!4eve|U>u4a|T)SSeCZSd3 zrfTS{z`VnJ5Iuav3+UpW?oBWw8(y=^mT^!2!BSl74js=vP;@G7w#~Z_1;>=;@?3zi zQC271VRPzVe4cD0PwqBV%yse8=}ovUQj3Zx3M?AelkEBI_eA#hG|-8NS#`)27(x}i zjbhac_wFNP0uO5XxMEg_un3Dg6E4(X@d;HdW+_m!q*-p>Zu|n(ol>`rDyMF5xi#?= z*A{OEQWrSgZ;Z_czQ%cseQyXTnB z(PHB6=auaRj`X1)_FFl9IJJIi3-xXRkL3Ac$$hwlflF|~Yy9!U`RXC_xL1g>|9d5u zAv(D6C56=MxRDW$HEHvQeg zm3KE#hi`*wmp0HyNt?rW-#*w^=YF%s~-Pk3l9ZOKwKQ{!)H+|q7^+@A%V zjE0bFzE{aHZL8)P=caux+YiZTdaZe9i}d2WQ-8o6?J3!Y@txFXnTe)nVDE7E8NY66 zT#blc5C;Ly13`xdhKS2gPrIA+ne7rL>E}Sovx=8D-%52KZ~gkMR0d^X+PyudmN`YQ z%EUVxunJurxuY-Sh~<&Eoo`Zp9DOTb=yqVmFYBqsnw6%H9XlZHOb@}mZ{&iH2_Za=+cUvS?d*$h+)Y9-JEDg|wtDQ`9wpbSpW?yNp3L&UG8$~K zn`g(GZn}c!^Uk+7E{H#7s3Bc}!HEf`U?kRAB3@$$E|c|4$D#GzQDClKdRoLlK$JKI zg^P-vipl<{hJP8p`{sOSDqp(B{d1KG zl1?SxUOh04J!v+p%vo9x@@|G#9e(DLndPW!%_H&NTwcuJf1(#gG)JF*zt#VhOg~8p zRn?vAz%SZIlJ6;#5J?uIx9BaD$Oru{Y~l3{HgZ{!k;IX;PC!r(-mQj0bw9QFfkY3g zJ!HB+jM@9qk+EoW-%mom=w95?KWc@rvm#DQ5FWNz!``sgQdW^Zw?Q!3BVIaiY$7F3 zF+IDPl3*SfWW4K{WGKM}yVv4^cMJL>fcCQNnG8cGhX<> z=`MX!0z-*ZYn6G|Va7rjDoVwCEV^Brb@JT!;p`Ogsr_0spUh{`9jWy$nvPIb(5d>1 zD_EJIHop62X;pzIGmJ}6MV3dInY~h&tI7md+;~YSmmV}g5N6hpaaGh2YtF)9(I??d zf4)D2S<=9;bztUpy`1H({m@y~PNTK3Re3pzUSQVhQIHM+m5loT#bnBs#D zeAwd6RNj~O4Fyfe+Xt#im^Q{O{apn@I#ctj2A<7}+Dqv9t1Rp4v7a@HotEbboUg}b zPc%gc#473ENFw~`O+{~vVnsBG&-<2sJ?<=D(h@>*Eneno#%GwOS*8unw^#SvMosF+ zPwx7z9;w%yu2ssGKrEiVW|+X%t5t6(ulVvCH4W>EFAM|MnA>;NnqoPng`lm;?=X>b zSBxaV`jOv6aOVA*!$nCaOQheJz=U#?Vw$-SxrjIz#t14yhB&!82c_* zw>xcp3CBjftZ;wOC37goDs{B$~p|Zi* zA+m-(C;GNYK{4$N)`-m>Q`s>07WpL*UZe8oj&Mk)HTE}TO3f7-Pet{t@DykdF$At@ z$O}cZW$w#EPgs6UYi<9wOlWSTvq7@~Zj;+Mjj7srh%eveX84XdMnhI0$&MuA?+IG? zx`-c&!r!8xpLOYCxF20;zCVs}L{0AB_2w(DuVzzpc8kIbg+4*zqAjKO94$iK9irDW zuWer_iz{~1%r|kESkH%au@RY9#+;|F;qI>M>Efq`o_IJs`q6g#^p2~;VvtuB^QT z3#b0Q6QpptzGCohIaF=f%xPzEGc*+uyg;8RE=-3SpSuKSNDM-c;kWLi!7T%Q3c`u- zIEeY{_mj5(QJOGu&qBCX!jumcVngzlVzH*W^!+7t21em|b5L@4kSA45v8Hq~$Ulz|Y+5scF7wA4gka+x$l+xgKEUFJn& zoyNqrR1=zt>P$0*`m;k#ybkO3CGdMcNc4gtq=GzCrTMvxI$zYg*>6hmeEoLn@&=|& z$5Un652o-D@HcjCGrR7QIEahNf;Jz%?xib_G_+c)z64@4Ng3D68}K8tk4EC~&HRAN z@>Lg$Osj5|64l+4Md9m&cf7f!?BGJ2db~PI2iG<2EI7-TmhfiJI_6ul`J7^03m9DU zz*v+UN`WiAr~eU&e$ms2NW8r%Nw}MX5UM!}tKLBdQ-TDV zA3h&BV*|VIOZT4M!qV9)aPW=0wPHNhszFsV6ko4Z*@1l}^c5~23iXHGXJ)7n3v9V)>}v==xO>$f_!7$rQ-8<$9d}nI?vScZ z#>n~xRl<{JOZBWz`$rYw2y&Q4kozaVoJa&h{aV`n49BI8+FOK1^PWy=3_MQAJsAL`o_ls*L(a{l0298n?p47jfM=!+Z6iQry=R09M=ZuBh*8aB`F=7bzQ zX$A}ZpoQ_{Y)>XDCaMbG!+WGF4mf(cXWa{Ao9n*#Ik`koG^M{2mDBdTK>V<}hi{;; z=29;y8h?VW=+T`bPRx^0tdQxwSC(Ag7JB2v^f6FV^tIkRZhYq;o07|4nb8N8A4~nA ze%-D*@~Xvk-_#f48GOm(LcP#MNPCodOv(5;H`t)L=6m&grVD8VxBV7d>=z0;Ywfv| zpA33Aa}#9a#NgcGqb^Pa zNh$&p6lsy3yMzYVJm2#cUo>`;iJU-su~*RX{j*KPve!|v7>s~;$vOT(#hXkYA4*l@ z1+}#C!bznU^#|Hvz6#?E&+x~^YENb!C@@M%X|?)qGv*p=EKbo3Y72dvR$fB_1S;MW z+6SDHOa(Z?yg9(itb+lo=Z-nBhezXG<1alrWX>;%tl(q3#J79+dEyLN8!yWmirLxJ zBw_R1TBb2X;?D|J^I7_OauLMH>J^~f3)<7A|oC2k%(nd#MNT7V_Z zm3Yeuo_V|I_lg}0R@Q~{!!$vf(24IS15gX#LByZX51_T^Iq0SiiOE8wj@E`Id9^I@ zFk^bU6d&Gm_t*>-fBJMl3praxFtijxXOk04y;gqz+BrVH)XifY|H>kblZ!LsQJ1HU zzVIkB{1&z(Ct@dyUc;2#5WrX6(44=xH?usuQCa2efJ345HEigpP-FYyAjO&qmEG*{ ze66GcGaJY<4?OoYjOlS*s^yZ0l^*NxUoEb^`QX@uV_*GF^@cfJh481++X}vKs9q$w z(+e1%dsr{oLXfYi4Ma>=sc^m`4tmARf7t%Ikm#Uo-u`Xq2C3YuK^K_0XJa=yDqAB; z{76w-9v(!NE)k1|>@-K9Jf%&~t@K8XaJh|~8KQL+DPDq;ei4&8VNw4B2e-a@Y9s zgF1XyMeaq+?2MexA6rLaekftb5Svk(3~P1!`jA8webSQ=Xo{l@BqwT7WA{od2rc4& zJ*Dk5>W&%2^U@{TO{1n>^MpT>6zU{u;*c=ubzo6CRXEg1m6b+_Ymg4EHk)UXo@AOT z;Cf0EzJD*Z5pm60m?=wnVt#L zbxjuc)pGewQug;rm6Lo)c(Fi%i7xFST!iR1av6u zJ?09Ot>Yrf29|g`CKnlM7%GI4ujr>P7%j8#EltfV>?S6QmMYLgA6hY-dQB}aQw_V0 z#E5l>`qOa>A9c^IW41b_hSErIgCOze)U7()8K+rC~=kB*OyvShW&5 z*xhbH-Y)K*&r1)>(H~wCGXmG=RUy*ZfCS^7etya*?JH5>rg$yxV7x&Vp>JaK4&>q2 za1QbDoZfA-w^Xuj90Mud82rO2aG?`a2`s?{*md&yAL*FAV^NE zaysPB3FtSJYCp#=s5W=UC8Sjky5pwK&etX&k1mfEeo92e5!V;?i8-~oo0%Pwu2WE5 zvb1&@1_LVy_k}pENTG9f(jxwdU6*AEOXw;0fkKvx-)K6xj@Fnz3Y#gB7I)v=#ray; z>!q=>)+g)=b!enCDn^~3(-b@|A#}J;ao?$Q*G%$MlDQ>;Gx#p%5YlJf0W0`EY41yv zO4w%$g@f$OP6KK0I+nQi-su^|2_l1M;AHoFkiUJW zos)}kCzWw$|3I#-dg`fyQN5ptWIUB5Z942_ks=`36(RD@M0omq+m7x$W2}=eEBGZ$ zPn1F?9rF%os+ls z-45uCtX)J^X;rKW&CF1r3oeP6DT^OqG#ErF4y~4K7YU_Qx_3Ah%5hJe6zm==-OR!p zpkb=7>D0c{6&cA^fJ20 zbwl>Zo}#kLO6~#m_FU`kw0KzSuscjJHgrpFbH)i?POALTzQ~OLRXiU?u{D5t4UbHj z%KT%T1H-Ox`;@2m#~@S4xFmVB%s0n(Ps|Vdw{qWHENwSRp3Gyp0|V1$M8mDY%hJs( zi1ih*6DIuV<+^gV-#Si=<#weZ%A795uEA~<7}{QQo8Fc~{9JFQy{BTt%bF034b~^b zyu5_{L4x;IcSQ6@KHU3`3YN+qP}nww?6q!JWNtsPV23@%!Nqz$3GO zP{Wg=1JE#45z7>*z)Tw6+=GJ;?X;4#>MY8o0f1B99Ir({Ow^angH!$yg-G7XePLL= zm4f{my>fI!qg>Fu@3aUFGsIB^q+BEgWR+#ZMsuucw2PXN$LEI}M*%}PDW*{S@&RS9 z`a8?+s(RTEi9y^g_UM>ZB_Hl}Lj8BLQmdqqcP&2ZyEy4tk`K8HliyYxKqIpF2hs?#gw7 z^XxyD)ihDAI=&dRd$n1GXzl*lZ?(B(mvRr|*+v0e(I zL~PeW3%s$#Ry;5JLZ!r|*^HXzn=u{t6Jz^t1OC@`^2Aq*B+=A7q#aKLB=Rgd}+qI$MWt2-rK!7#s>wF7;=Sjv$~ z)Y_{G&mf&BLl+t$SZFFX)OwRL|Ax~)Iy^06NkOU4KWRiB^u}c0{_Y${CcWSq0B03Y zbLsQF>rGD(9WyfDBv7<9U?}zOSxJU5w{YnceBqsUDv}B5GQxdPBO21W$Sj>YB<9~jPq7v3}hz&q!w)t?yCnyPf*$+wnw zVk4BxnS$|+aNDX8TkY)rr&Iyj%7)^Jv&x>IoGZY%~;Z|kgzf()(=&VzlpDti9{?&5UB=1Exa zx_vezuBad?k>d!tzC(xxYFLoVwrzFUuGw@q!$1fE`2%!77PCB~H~U9E<7Su`XIGiC z4~1v9kMUt-^pW4aK0!q1+olNZsmA-~Ek>IMa{Choj(Ry5CM15p*6sYdVr%KT^&3myF!(!Y@+jwxC;nH>kRWfH!2I`)Ps+`!>Xhd z4awhw%HRR6vcLpVbNaJ;oWa-Xfk&s<}`0Ek)Ci)}p$t=K2rU<5N={5+eC$`?lkDi7(MyNHk3{X3&OO$J!yZjikSlq2BJbeD`N*V@X6Mc zawwd1RY~ooo$>L&h|V{a6rdg40KrU45hxLoytnOESDbz^QD^sh*$v5(qzI#ytM0|G%jTF_-L78JDe#q~y=A@!`qr$@*&15cZ z%KpA=Nu0ce!4eW`o-DPe8mM%;)P9W$PZtX#SMi#-A~TTL8t5@c2-L-fOhkKE!UGt- zf?)w5WSL2>4seK4rRS+rYlPNiIVS_Om-lh@$aa?ej8hlQycayq4M#ipvp;yj2l52l zY9TffkfPLUfdc5gQlERI@$M}87kG?ZMk)0w8GJH3^}f<4M+iBF8t~TQzRTBYj)>qq z#c%9#sGHlzOn{2edPY+|73eO}HR*MzoAj&Ap#d7<+)g`q3IFB2)zY#`EDw@f13ta2 zLZaJKLU93zjH|Agx*WedOF;G#dGyM^gQ0RCKJV>&1LBDOjAG zHrBVjao(GlyRQ0Wp|RqK3@cq42Pjmq{P{aom zHsNPu0C0qJ4N3 zmB#JeA6u6Nc2x3k(OK0-8feav31V^`$4vGD2wC846kJh4e}LlP#0r8{E*G~iLGW+p zcdglbEFb3ESM)nTHuqQ3d(n&rtoF5GU=EcY@66v(nIH4S{g+biRkWF`sl{nvT8Kz{ z6@~&f`y3`iJKEr5mQ6~XRmeGJ_sIF^2?;jvXe6zpAo-l;ka9eFJGYMIGv*i3wIlw*_n}8L=02o2wB`ul35}RTFE2(J<9tFM z=M>T{=_#&dA~mfL;Y(k*)92IEN(7g3U!1KmSqVuFL9q;#Gph&7h8?84>)`J_C!bwU z>WSoZ9OD(~z1dVjqb_N%RyQm8W|n6y7_}GNDQV5yl4&*LJPA#fBB0Sh=QFKJ%P)7_ zBn7FZg&d5@C==kKNOewn0vCGCD7)7gw6V!mnKykh?rnG{^FfCVUnYMl_TekGq-$4v z2B`xU|D8-Ogh$4NKLU~*n}Ix`(l62M?^bfwKAw6UIprxQo3o^b>S`OiB;qHZzD|=g zc~HO*(;L0Azc@r&gaGkTpvx(KEw@QM-+I-Tuv-*lL$t7`#e)#v-*Lx=)X`9K6?YnZo#zL8i^FBt11qYy$bOKh z#OW7M0)EbNbcQ7r)ZE$H;-ktfchISt*fWfJet@Ijq3RzxZPCUnX>v}oY#=Wpgb^nr z*F))PzI!oF;Pq_!0SH)ysfj!4;C6vlV-<94WAz1UUF?dJ)D~Aht*l?lmtg= z3OHPl?%IG(26~LYEW~NuX-xCnjm+?GZfh5%Xik|GeNU1KK%Vyf{K>tJ&#U*);k&qZ zQc6@#rTSC|f0*OYTZ}GLo3H4@LY$us$U)m#F6ue)kgQYsMH4;I(_az1V`{r8)^>GT zJQ*IjC^X|qAZH|qy?Mr|Zg$_k3RNs8e}pAt?B2YPjx?zIKCHo7Jq=kOI0mlPqbC+z zgf~O-S`sgHhLD1~6tDA6AIl_q&O68_tj|jHWt`gELOIRpZ2dqWNqCIZ7~mZQW6voZ z3Vzkmw+|<&oe<9}nFVruK>_Sd_3@jrUGod$l0b|xDecu*%ID{JHsLDnySkkPb-2p20OIo-mpb~(nAN_-rC}x zzKiCqFcH~DPG>}8x7^@ITk|UeLM~BAUnRm7lplDw{=3&QWWClj9$f(~9aV|qCh0_X zF*RVlQjZGENbzv$5(nzi5H*g>r@bs3>EWG0*U!(PJx(N>XwSF(HE`m+*oyPf98U%` z?@KacX?RL?AJTihn_y$>MhSe0wzTw@lm{x4g$2wwSX^|vvet7RZy-Qf!skS~v2-%_ zU0`!n5x6zI#tAM^Tb5*OqYj_*G&iIF?J|=8kfb=n$nIMTed1x}d9lZSavkv0#c&xPTir1x;`{#09R}I?9NXnlESI|%pi|S3)x}GBO;lM^Std|3 z9EJQnXPhY*4)8pVTRq>PP@+uG<)-M_j6{#Ha&P3Zcd6isj!^V$jnyyF5-hZo;(*Rf{n-*UbpEd+D|3j1edi-o2YJ$Zl`fL;*wz?Kc zSKpR-OFqUmn8K7jb0q`k5aW52#$)B;O5~ET1Pu2QLj&#%>f;*fMubmivEfkLIQLv| zwW!o(d|>&oK+>JH@(=}tb2qQq71WWf5``o( zZEp%8Dxd{qt~>C15esBB(lkP1XF%iPUqU3x!G}2ct2?2V5Wu!SyfbSC{~)sLcwk4! zw_8tBQobGWR`=r*X>}W!_{_XsbT-pQ$0AF`YD4=cq_T4~;?OZ#?*(ZXSyjR&U9kqvsWrsjJW?MJC66qA4d@ zbM7;~Rztbv!FaNkUIvT$98GWNyxg3TX=v zx#~z(QPm}hBgpyT^M?3#O(j*oIfs0O)=il`UwpOMYD?g%bBl^`$sebag;QbL7An}x zI^yEfocr*x3anI&+iPAHV8^H)exLj&i~9hfodc}tIn)pZ+P3J>`_YchWi53HETkKw zOhX}_;mA7;GRuW23ky1Qx{QkiA-rUcpEKoAE-qw0EB4{D0P|D`Sy}~a3<_I!M@^ju zR}?&DTc`_i5~az{pi`1WP?E*)y)X%53WGu6Z7ZBueVd6B*Y=YX|3_id(AO?s;lFQp zyd~~_gGZn)Kp*c2;RzSWB*zY(Yb8=b;)jYKVDxtzl6rwOU7wPUXZbHFr9T?UuOD^C zBWz1;aw7QgJ}xl>w(6_+@j%xt)df>%5N)+yR^%*ds-u!4?ZT_ z7B{bjp()Ojlv=5+VwoU|BYVr@m1u09e=>#8gzeX2w!=Lqo+^~38tN(rr3lx}sZECe z3VXBYDle_KHir7<{K2~rpqXhsw1DoFnNaEe2$tip{&fc8VgIn3Nk-Th_)!1R43oi% zsSAwlyf_nxRO#I;-3jgf`bJ$&pZY7QWS7hB>curBxvIi)xcWis_i7&`_mIu}t_p)a zY0%nl%K_4Oa=ur(SPj_dCWqj{*rZ|JPE+`M3d{?X#?Fbm{`@NJrd-MHsC`08)-jBg zM3aKNaGF!`sqX>BzY1z(z{{PSj8NfFf^#Xw;b?nGp%cadfZ?@JBH#Zd=PX7Pk) zk2it&Kg(?Z9+ed-O6i`sj*-+T*5hT^fQikoCTLe%xWkL1IgtvB_u=Z}9mpqfU_|4rI z4HrXoQ2ww2zbh98`AnU_(CU;Q1^vI-By0mj{e3F-N0RGgX8L8pPVZ18*1%WQ1}G55*|7Db?Qsi*c-V7OQ zXQJ4iD{CPCR8~faK|u$hyoyW+E5+&~PoM<^f-&79-LZh-&rizr#dyssJ$qZdMrrdW zVo2!qpxv0X^ra7+k*>(fe+N~L{gmXwmu~J?fw=-e3MgYMKg#-k1Z!RqgzuYdk02vZ z$!V3liOx)XXl72jvH#3bPaa*yaMj=j4S=y&5ar62hnvhI5(%&^M}?EGj%_h>qJuQd z`;R=NqbO^5O%qbTAO=G4FvKhqG2-57pqh|SxTtmS9MV!`qy3p!^mRU9(UU%6d~3^9 zW!7`0mV&ld$eS2t?TTF-Ma=#t&kDRbZ>H@WKMuYrQ?n(>B5Im|!T`n{;OvCd%&I0Y zu~l1**(wMWFluq;kqeBXj88;Qrw?Kj6C|!&)BdtiTO*H zJfh?1sI{t5q3AiFv8iOor`g1T*Kx+~Rie(gl#tM3kI}<$DiF~Z^ZmKh7JY&SC)ao1 z>Uo2hdFhRX3H^Q_OTV7i5?$ebozGV0+aw3k#7^AI0nwpg>7Gd1soZ%2sIs2FcKVCN zw$8Vfn~SiGtFhGfbeh-UsQz=sqlC>xcg|1I_e8fJd{htFvf3O<15k->pwU?Jvtb_o z!!FVMsyfrF!LwPRnBPNx(0_}YZJac;QL7Kw8KZkU}= zTN{Pg&G`7$GI~n$#d*^mv}<9(>-}A2(`qGXI3XmV6}V}A!e@B zQ&qWRm1~RbV5%dwOQp_CTJ(z{_q*~2iIbT{029vEk=6+m4xUufCjTlF^q^O%xF_R0 z)oMh+R|+h2+z<2XNf@@MMhCH-j*ddBx@ONq5i;==fbN^N^*qp_nFu9j-6R}xkU!cG ztKC{p%5-(+#Fq9Xps*s3pdcO6&#WZ6wr)FaeTmy+nYnM5(7g3PZ_P?yaxY2kG`#^3 z`cvNclk@;PoEb({T|2hf*^AfqswIBzDA*wJMqJQx3*F;E?al`$_8lBBowxbWm@m*_ zCa_cR-ckR65NqP?9Nq0-@w~hq&BYk^=oLjym6bXyc_bzoh_+@HKoDQo9+=J9T#b)G z>L0jiXIr$-gl4b7h&Nw@x6}}_g6@rk(JB`-AE}Jm&hI);vq5i* z5LV*k-mTMdBm-Ai8@ROFwedfQua($mDznI7K8Ewv06}{h#Wet8>tPPFz=GZ*!s*%H zTSMRB1%;ECgZorh!o+}Bofr2Ysj43S;S2$I*}{vIHa7WS70-1LLaI8So?xLCV>vS~ zj_oJgx!gmW|3z-K}papB{fCt4@Q96Prb2>=}x>2aPvIuXTL>WD!l#@onkvcrD4ScZy8q_S6IwBD>&IH+QS}s zt`s?5MT3dV?|CgZ+x$ydi)J&~W@#as7Gi4d;+vj(}4qqL|2^Vt%B~n2<4uzOrz42VC0^>sp{?5heLM zf3u1H8+!?cV}Tk;7=5Hr+7ID+EBu=}j8KHRKJJ!#KFXRjfF=ShVMX6{^b`ywmYaq> zda@x}9C)4CN{L1$_B%aHtbFT8EE4@q@Jbt!wgKGXL6le|19Hxq{gfAYDCs8(2PCG+?Dli!`MiTS%7wle&7y!JgBUT2-3eANkc z8}Ix40{ckdStcE0;W4y)-`pY9wVQ^R_eC$Z0K`&NRFAY~&>4qw2X3slLPK*2T zv#b`u!z4i7b@1!qVOQ=9S8ZP?bO&KpHQgR$YV8{R30alkEXs{?Q(hdwV8kNlXyR+^Y&Xc9lclOioyIHnw*U8pevlY8_{X5sA^YsX6H35>l1a4cch zE*jhBj?IZ}XUDc}JK3?lW81cE+qUgw=jJ>AJ*Q69{qMb}YN}`T`>yVH^-N98TK%kk zp0tIdG4(CVDms)OUF=)zWdJ>p_1zI7=kVNu%kA<4o7LzJHlSgnL$xb^DL;naSNabn z*FUn$T1W^pyG#rmW|DYyX&8bFki-27{)oofr7vR(6&&+W!CYO|qylwTsc8%BPxD`x z)1ii|`~P^02M+?h07O4oluZLM$mufCV70WD5f9cR>4%z^6{uSEL@Y-+`1+PK@@X?+M8o7!1LLPI6rUj4_%R< z?c*<6;GxjAK-#>t3}e8y*1F_%`lj|7{-zu2d^I~z2!&a3r;hgfOhr677EN9!4gk+0 zAGZ9Ts-R>u4G>%qcfbz^R%L6Z+L`MW$PIetud7rNo!&~-zDW^Fi1eDcrjzSadB&_G z@F=a6G(*30Zx1kZ%8|-|X~O!_8FpEm@L@2$DayTFV*(gmo1`;d#cSxjP*=3GEyele z4frfrQ&dmQ$;?m=c>Qyf#GgkK65ki1`fxU?vcZnl*pkWIeAs4};PbZXLaB!Bqc}cd zguZ<^clfQ($r6p!8BNXUI9Ov1>_z*{tTMbmD2pLNANxmf9HSOD;N1=3$#!m0aAFkh z1p-?vq6TEDX<0^zq4jve*`hpN8O2#JQX-iIbtQnx@Y6_V};!zL34n3gO zaf2U!#6e3hcmkepq>q^2Tn6A-Z%iiL}~9g`0Gv zK;9eL*6(PtMy9r*9P50ec!a!}mR#4vY9?OGyg9eP&yw_qei$FsuwCUu0{ow4?%e=VwydG(s0EMOCSfJ%Vv-2>I_1wnK zIy=}PD_pL(#IijQG?WB3e@jW++{O~FfJBXoJ0d(^~)~3JBE%O$lw8Vc# zLAMA9w5O%E{T$+Hi()oU^7a}?McS2U_%Ik!zRyrskt@9P=m=ObQnQw-T*<%9@9 zM{_~>nFG45p5Kf})A$&_8h*CW#Av#F+8m|I1WBqm5fYRno7q%u4L5Xh)M>7xX`y}M zKjhTCJ1kS+VHYbej7n5Kxg+5Ot?0IL-UH_)v+2b!JKsIAZt39ag0>iC3o>lVOmvDpcE+fb)14 zI3-RX9TIP|79$wSVnp9e#09AvVb>D{saPTOgnxdfm@z8Q`0+Dl0#myimP3@GV4ILG=H37rD-1fN17^%1$QwpH1c zOf6E5t#r|i)It+^gQIn=RyuJxBo;(`QW(*+mDmDztGEd6Qi;G+o0^cFLz|qaxwg(l z-V5bNF-f!T2RP;u{tXP^1OwiIAX_Q%zQ~;Qh*_rMjR6Wao_22wvC0B<3!An{DSoMi zWfyPbd_z`0kyn&T7)%@1_=2NwfCoiW*uf#+1O*yM@>Xlfk zx3O^$^`1;wWix_^s(U*KI+0Q_v+< z$Ej#f{aPdz7v5L7L=gXQE)ODOE3o!Y?Ho;nov5+FYzs?mHxwARAT7pirntlz>0y$h zT_{*55g$wMPMBEIk$9#}Gc(1i8VPnovvd7}#iuGqyO$ zV)LTLsq!6gYf)+P%SFwv@qH@~uX^2BXiC+@5W%MR3{cztDc996B) zoS85s@upoVwRTr{V?+&T{ogc}rB7FtXeep~%z7|8?}0wofeH6a&#l=|`vqYXN`o8<*W|JaS2)G4mgO z5+mo4=%}h?NK?ho&cHhf?O<^^tZ}a_Mks~xa89X|94l3tpih#OuJXGUq5C=8-hFl^ z?Bxi+6*fDsFA;WY);Q>U=%@Zsx)P~tPtVBl_2%z=9Ow*O`sVUV)XzEnudf_tA3|+i}HzUAgD5 zzB?|;m1#EYhKa|eO)2D&U9`1-AQO8PiP^#Dk;AOqMdJD}N?9P3%PcK@pIh{&9^J%E!;ru`kE*KPi;Ok~qg zY~nFm*{crKnPI*$hs`C?9>nINkCade!g1OnE8(5OY~|onTgM>E7)6_aDYA<|isSKi zVi#b`$b*Scx}8FTDc57u^J$*)uY53%37O)ca4Swbsn~FN0N;EVFBwJKG)6%nA7loPmMGZ z*Ap6PB!*jPW(>bE`SXF2rDp9ftr4+34eXT&S;v3ggSuF*(yc+Y^%wf@~PlLm$7Wa-=0e?XN zLA;#f`dsgQgET0qJX00DEs__mgXCL&pfVv)8Q||3)HsAh8}3$x+tv7YWJtxWjl+^3 z2P6LAohHL~r$}fnc^a^DksF`|;hr!J6nc7$<34_WuM{RaY@(HYZ6Q!@^>>jq8TKKJ zCF*)WeV_UbcRrr*K;>mgYRQxuE-~C+lf}ZnV)|w7%tvjTz|M>-0>lSa&OzpXnD|`R zcL~Ues+~`#vZKyX;PCR<&w`TeD;nm^VIGm^NMB??fKyA7{e(lxCHB)IEfEo)(`uqa zR&Rjaa3r~<0myKaGhuo z&0;995a&#f*oSBjzs7FQ1aH?#4r#taw5bpHWyyWL2V+djG!ps#s%O`d%VkH-$Yv7g zqpkYKpYt1B7HnXvB@bZsX5&a<59j)5AtZ?7u4p4WB+y{+Wrl#}c6UbMit&24`|wV% z;58xBZ;Q5yYI=v^TzHzQvOes?ef$}PG#Ecgqy{@47R91%D;k8V6@UyP5AjV2jbDwJ zd31FFF-I%}E(%VR^viKfU+_ew5B3{<%@ikJAhQ3rR;E`6gJc|FCjK6pwfB^N+Ra-Q zTbPIr@=xI!^sogLQ@Xs`6_`}*ybKE3uN1%H61?^*T^{UUtG6eVw*x92AS(H2|Bp+f z#gtt?u_=(9BwN@F8LHnD>4)i~59v<~8#QE1^q~Vtmn|bcV&e8iD&m0qJ!OE6qK6N! zl9Lkw32mqe($!*kw&Zq}UlL?9q}f!w_8HvDg=aVi9sd4;=VsLUha_ta1wyc7OMe)` zb;CS}3?0{I-68g&ZL~9vjwWXnM=;2)Nbk3O)bvyZ@Aku%JFjo)g9NumhQ_spC-{W4 zg}i?l^i=B0xW|U`B6-*E*a(*;I0X27EwX%2o8|y}hQ~lWofYr+1{f4P&?_L+k7!Jv zi)~Om#(DgOJcCUBi=+g$enbZEfea+qGv9RXSNK8>WN+w7-$x=h&E7@3C~kJ-Zc@cFpiTz3Z67R=cRhW zIh_jTBce3|Hlw+_J2gka#*H7QbcW!`ay=gAQ4J_Jkf)7Ub~J9hU?X=E$Ft!m8gEjm zMR~0>U{W|^nHz&LqD}(}-M>^D8lPF~?2|XqM|5B#P$e5c090|n>{uZGeWzeUgp;?~ z+N41`cZ9C!Pc!)X#<~KwAh2nd*^=J%dbXw&Tm4YOv&Ou{hubY=n5BnAUq0%LbpzZ= zM9c1%|GT}4H@Nq%d!0lvVv}Beq(SN0lq{b7S}zf(YEcM)<>U=bWUr``LjHX!&(0#L z`k1DqGO3kYejO(Fmha9tg**ZdXaJi~x5N#oUdS*jTqq+-pQ7;7%K-*8H>TM2kJn`t4(i&1SWM^8=%z35i9#eXUrr)BJmD1U;B=*V&5w?4n3i z@9dFWlE)G2G+8w9jC@()X%excr>R29wuE#9F4$(TU{ga*npV=D7`8b#+@inu?UbGI zpb-7Yz^E$+r1i}NXjv5R_mvOr9#8cCxcR3raXT0Dg)UE`dz*`>b>J&E)+#kpfh+6V?0_H!y-yQ|nZxmewM6>C&;dSDRbXdyM!hClS|7^zW z&YtKXb?^WZXm8FuO{zKar`Z{zI(OFie_YKGoJGi%g|Dvy?0@Nf7oFy|%HVDVHc#DD z-<>I|)X*Vp@1-3Ascc zAx79~ibl&cs7e#yRY~du$Sn_5N14R04w?Ryfo?Eso=$#sK-xfsQhZOS2-a@ zL6sVqwtxw1zY@e?k5?IgRB<#K%xp^~IvtpcSF?g{s)`pQ)x9J?v}1RN6TT&3SAHr*4k))XK`z`n7#r zL_%n_NWo<~_HP72SsSgeAv}VgD~C);E%#q8FrY4-H7QMFJb6uZTYDV7cMbO5klG4E zdMOd5DxjgHc)n)3OjV&J-c3efF~!|QY0Gx|1!Yk4fOXlpR+(rlL>(z#IL@?-+~5?r z?|eavI%Fop3jxLD+(?SQm0uuOzp9h2nw^A{jR8v-X+?m?Qpa$g4g9OZ>9No4VK#*+ zS|Q11rhQWbHYT9Hm2ZO8YfI1vQ7M)#*Sk~*Yj=j6T$SB|TU_Vsmud)eMSi=wY5ptM zF*5bkX){Q8Y=X_yFYs|i3-8RC?#S^PBM(n`fBuO$3sb3HM|e|pZfwbexBH+h{%Fe8 zKg+*+5%H9q-=w1S&!VR`0rTWK5cC>&4nKOTY={gWN?(UsB7-;1i1ij9Uf|LQWBS)E z3j~T;p0&Qf+tq~fgZ~;;GsH;y?8eF+qM{}4#F0hA`WGCEuy4jY#R8s$&sN0wM&EM7 zxD$D02mlT;$yg!b;kQ%LURWk1<)H>|kiyDV>><3-vY!~Aea0|jtg#enYvBfDWm#KC z5k1BG>2zV3(?<>M#KVC+kO3w2bPifSzs9B6b9rGmcu*_;n$>>#zvJq^Ni4r?PmJ{P zJ#%bv8ctwb8F+d?hG(`Auo$(zO$ok5r5BandZpJgR!uApOzL66IfyMuRsS_%$4q|W z1vv&cxbIDr^f-ljnRV1lHo%Ixvjg+%%FlOKU6_q@C5#k6-$@SYFPDVDU0a>3ju|v! zCGcM+EbQ!8bwQZF^B=$fF zT?`^jMoP7ADEahxhebnhN8iqDTtx)~ySI;#C*R=Ipo0mr{@U<9B!beh{Me-pA)V3z zYqdG?w4?PUb%THy0pLC=I&o7xF-Tgo*x619w1^1o>wi21AEv5_8E-(i4-i-sTz+5e zXP!Ci{hAz=I*t2<0~Le05zS{wT_tB%(@VY^#oXwhWr~ctVJUXeXg2ncKeT`P-v8$G zl6SiNGTNZaXC+im-@Sj;60oZjRl(|uHvMjAuY66s4^@9<%)gaA_Kz}>r6ZC6+|az_ zm{5YstO*2`1$_{`MXm-~W;h)6F&I)2HiZzjUz*8ZO-QD#ay%vQG{YL)iqnqgPkr~D z)vIKv{8y$Pqx%`F=3oITGcYC;sO|Cv7w^(ZA-DcjG(9}Q%36|RF7#c9+V^9E@ceqJ zFsd`E(fueVKYS=YK!&xRYFfjxzlvduwA)qkZvwVf+y^(=*?6l*XLdnVNlNVPNcP`H zsCyPjBS@m@=LG8wnxt}~|4$G>rbt1{;hHX0K~|*N!~Cf> zVf8sjX7X%4SLNe9*&TVwY88xE<17|=#OtqG7b)6FHsoj_U5oL|wT@q&lr%|VI1 zoc_^Kg($i=$2!9?V^0{yfC1^NynsoT$z0#E8pd1S0Xr6q9&rgKQ1k`3a3n9rN8< zy+it((<=iLWGv7urapCa|JnMSr?R|xDmfp|z%g^%fX!d<-85?Fsg;ly?WxzY?xNYU zX}%}r@-HS>w$Uy&B)|iQ5-Q$yUTv zQCfz-WOA}7>I|f6eY2<2;kOr@L#I-h#mmes>!I9sF~Il)9UdegKw)+SBLXvyHhr~b zW49rZe|6&d$I+-Fe7WP}Yxt@HrbS@{?+<`iJnzo80O*9E#F>2sAGq$_(!iA~rT=&~ zcyyoSf>|QmjLfQhaiPj<8ke9D&r5xbqQ9!qP1t!$B8~Zz%1aHFs0!{m7C=jCMMv$6 zY~M=k5ZS@`u=8)4zY6tlCbS9AP1@+?-bOU9rFky`IvU4s1Gr4 zsBCqdoES2smea7gsSx4optugZ2XQq7siW}Wu~}}8k+YKwqM8Ysi8f0Uu>oiW^4{b? z26%ucnx=Nbnn{4b-f#+tiZC8z4P$vtP>HGnGU0YWF)Y$@v8h|V?ISZd{1o|}xl@d^ z`95^V%UtE#iHOCwB%d!v5BR6gbrg4sN7(&0AEVk&OtCoHrQUl!2w&O3?yw(tuf+eJ zp3ah7-*UUFrKq?-K=k4X4ssMj&%#aP%sTu0gsfwoJB_*7m8-ZIHe z-%^JgTo6BqQGHfCq1-2=%Xm&1A9M~MbYwJ@7%n|5hDMlPX3^aH`YPKhfGK$+BkK=%ZfVn`vQUeO_^sH5U z5o&HpB}z{BlA*_n=Zb#%Jmr1zV%CEZmkmBbT;QCYvhBC}D7tYy;La+0 zc|hJzs;t@L4M>0y9vN%WPB%AQ@Q6pCKw4q3SUzmZ`;9yeCgavXi_>sH%H;(al}U0l za>9I3cWU?n7}x9w*dnxbzpCH@>HSTK-kPx6i7Z*7X4KL(-5^AxWKr9jIEEoBD@3_D z5T{#VEl}(laedKTdB7_1`I_GXefLrF3NvdlNziIYUb&XB!v>Swa9KBMgI>g`<-*AtyWY z|I|(=WM*Y%`X2)6g#Vefb~bS&WDv79a5fP!F|so@f#KtWadLJvF|dJgU+>b?u-o9m z@O`P-J%EG2j^KvY*3Om8sJ2)*ZkFUe4K$<)KlYTgX=Cr|@tsC43$CX%13=IbN1)8y zXC*bJ9zsh(C2Aw(ks+WdEbsOrRp2As#x~uGGxozRaG(THK+~f|*c*Vw!0m}8-Ej72 z9gvWdkpwu&rUIq+lM4uNh!z5!GoFMg7iThyXcp4qkx(l@$ABqpl8OYNNddW(!7J2f zCW}~{i79HuB|D@7*Wo7^W+G1sm=@{xR*^$EFDOdoqe?fL3gB)NuVZC8ig4#vVUvXT zi)l)z=N&=mv4jN#x(P6VFQWO2_R91{x!gk9*FtYf#w|F?3ouJp>I=vZkU+aZ1T{tM z!MKY*7=z7(vl5noH&ubVXCB*wItGAYVD+OUONinLkD@}_*#nOk{aaWS>olBA>{|r$QTDWfi%XM zw~KH{&Z-h-N{X2iB~z-0CL4!C$OMT7tZQKH!4#kp!&L(!}W zN6q@e{Y4r7oBWp~53HPSmp7aSOn(npU;{Rs2JJNPUmsR3EQxj={;cqf$!S8s0Sqhz zCY~Axuw0)+6h%}l7%B${CZ@inKCC|>ih5rgm}e;`roL88y3^l31;=<~yQQ4s(6b0p z1z>orC+q!QrJU^3(Hxct+oMq8Q$IRI1;+xv453*up=b9?IT-^>g{HtE;W;ho<#l7y zVF&R~`gY5s_^<@AqJ2MXM802LGLStxQ!!^6mfX{JFd~miJNe2aF>r54VvWTY4xh>MK4bSF)zvSy=n=d3)t ze)NyE=5f;(_|N{zF037q7A;seEpY10YI$bRJq!hHA3IB`IVb6XhA*+`Y|RE> zuuHsBbQScR?tDMWASocx(A8KTX)iD00by+(+nqFKnc05;Nri^2sVFFTmnqUqKZ3hFAhJJkwn zKn-g|M6x0fZuALp=dRheVV!;;&XHxp5^_GO;6Eel`knIVm=^Or8DZh7qI^K+Sd?;c z#zKSi?9Hh=Q(e4u7{4YyQ#}4gy3;XYV^R`P)w@NbYRO zuq%k8Cy%|y)~=p2^}dI1Rye5SfQ-@K?~Nx=UawaV>cu&2N}6aQ9uwISXkYy#;LKCh zZpWJOVpjF z`-IyF$LLL*oiP$$km8r}TXs&ikO^Ya&E?9aBS7bB{C+m_7l+O@ufDXQ=BO~4xAtA) zMn~!6vuj)HgRe%>B{q~6#Z{&38ZcREssS4u{9nEFnVNeK-Om8v$*OvXH0(LFzM}`u z*GFq!y8DeY2CkWECGUX}o=3Ouex|HLBX^&q8AqV7>XTc=5gUN$aX8mSm zQ$BaN%ixBl`1EA*WT?FL+$~glAmGVA0$J0!(OW*``}6Ad^-yWn(X0ndm8bVBm5P0& zQe}{sOpBWQ<214SSJ`>&bRBcL;IKx~L-R%}?j!**1c)92b7 zxUa%VPftBQeL0&}x=;(HK9yjdzCX#P*o)dZIsbEPkp5e>Ion}+CbCTui0a}E>ALZV z4oL3~1d-Y)=q?#_yx|$NdCZSK&^$0LVcH4N>0&<3(k zN(k<-|G5Vrm96TnBihhVoOep)!M7-+F^*bX@xcITBLPo4Sjc~ z_gY`owv??zX$MMwaWNh7HtTI#R~UzV>2gwR2K*cGojYmvMx^@&`@YUu(5ivncCze^ zXnWO082p~L8i!4@hU)4U*zogp4^5WwkH@C1dJ{npcaQsb6W5TRgPQRJ`r)e}Anht5 zO$4hPBgA&0SC&_MT&L&4I!K}(yMH4~-kSLAS3FRtAL1I41BvYBG7qJ8Jtnf-XK@`- z{<^=Rve>Jx$fn0@Bl5iEDlUn%UH|DXWaf@(Gd2;H7YqfC4ou`6Wm!fp%e&j4Qs6in zsgYm6yURvj@Ox}1&DL&A?#)XABQM08&kwKg9rc2T&S-tM4E?-SGDr$Jg!zud>_@2I zCx$1Q^LsL&+h0N4GBpiLVZZ2>-N`uuKFDkd#%n~zpX?r>oevX|tityv^@{G*(2reG zUl5+Y6n;0(go-?zTYjm8cwt{B{d_@3JjkeF?+8FWZ_sMro1}93hN+<=VDP_ZdrL{R zn|drCz{tsw>WHR$82L;VaZFYHGdF87B9-ND`M4fbc0 zGa#cuAfi`C*@WTmEQeSkc_3A7p6>(qq~*Yd3t*zfg7wq)0ihueP^DdkJ%u=?RfmH> zm-}F>GM=Ok;yLKM$sK{mYlSmCEKdhqLRY%mGY#tul?y2{>cW>}|3j+^DjgmDlK~}( z0|Cr#kYJhXR%ZxWc{y#(KrE@EtQ3I6bsD`R{$5BW%qJ~$ZXma7=tHCKD}ScLQ6?m$ zdUx7SLhkBG@1(_p{bIVTVJ&m-eAN^q(KMUr$Dl!@ok{<`++lok{bl2G7w(y zt%QJsKpZPrk@u|Ioi;oO!Nq^htOT^6r-P7#@IPI%%x zWEtb~>wH?5snT|q4X}Rn(oLtJ+<|RNO4C)rXY#o1qR*xM+X!-@PFQrA8Q_DH4&gBn zqIv=4u@Ks65w~I*ryAs4qunm38!4z6kMEp=?`#Bl^llNnRBqN}d8=YM8^IGWT0dHx zl*wlV_DP*lPCqkDU2c>V5pG8H@u(7wUx=-Zzb5#!CdlR*XBeky9ID%9Yy`O%Vjez9 zT|YyeUl$=C*eD`W!SE3VVN(Zg)1&$5(Pch+Y&y#Cp2@@o&>K7ZB3aG9N8H*Q!>zbo zrRJG-=#<%Fq2pR`4GEdCV$LO8phZKHRtM;(KDk4E;od_G4)Ujr8tig!WC_vneP6P zjDSgQ@U{jn>e0>`!iQqQ2U)@E`3Sf<8nyr7h;j9?kXDT5%GGy^eGXdkWBGia(+t*)`=O?@25Ca$Outi#awzzhag2!hGlulM?*N5 zzj0!)nfuZ>oQx-No`_w}ZyNoh_g$GX3V#zhB z4gt~iloM59mN16tQqzLhA|~>2Gz#c#n_-&L_i}`?2cp-)JcHqFKup_7kRma!$Y z;5&)TM@y=A9QCQSh{38Ast7f-FbMEK-TE~Ev#bLVw>zk6-ubi8R7yeFx~z7Gk=sby z#CPX9U1K#q=1h`8F(RI>6208lL`1mY7OI{w1;l_zN<{kE3$$T*3St`{OcSlVJ+HpNGLG#v;iUH~;VHYEoUSEj~bnFuJq!*Vzgx5 z*)eV#&gwa)67}d?0CYoS z5{&>}d&}Q$|DIEq8)mT4W6W}vO|-UJrZ^g1z1D~P)f~*35$JZ-9UasU(j7{_rd3@O zEGkMDv6D1yZA{E)s;y<3FX%Rgy8PQcK}{U>=v)0Vb?bD<9~pBmHS1;3y$e6()gk`10WuahLLem%-B0vWit!R@1ip<@$Mf#dNWz z9i=P2%aXhqKb=)XOCqeQ8N*kS>UqMDiqwj@S10a(8743LoV9nA=F(;E3B)Q#Q;9|hYE>-;EZy6G;P>@Z7WrBK@T#!En{Gnr-nO_&;i z$;j;PG~IB2{J=tEHw8RCo*}nf+4;s{OlFu7oGo~)R^a9z|7Y%vKCZQ~Jg!xIp3p%~ zM7>I@V~q?yAyPRr%{?tl$PYi6vi2+$G|ff*GPfQf5?_QZ#V@x`$7Cno`;Gdg*FADN zcy=Mdhp|o2Ffauh1-E^Z$e|a4)Jp$NqQyJO2C7h+7{wE3%~Kc}CGxL8tQP^8=jH|B zX6g!+>tK8}N>p!*G$RN%eIL|iD;P=b!OwtfON(GOo9;k)P3Sz1Dlq3(NGTL&CT&;( zQ5|8#W=g}0Y;1B(f1G-^ROp}}ew=!#d&5+@^4-?LJ8%D4 zUemBjaP~G`bvjX9E4AQ$CWZtaJ@9g9QSJV)kqlPnFHOu|ftgylihy5gT-2uyNg&{r z8rm|a*kcg&WD@lDcNdGA2bS0)HX*EeX37*)aq5>od}ioXPH?t}O{`4hymqi?nCKewt>vTj!!n}DY+@O5 zF*euKnr_Z~QrLP;q`y43Jq2TpCZuA^jRux}z&_zK4N*1-d^fonlH+f)_Gh22weYfcp9;92Dz1ZUr}JWw+OsWAg&tnnmEV z3dWANW&!ET9&YmN)t-H`;TtlE!jQHrSfSG!)yy%~#E@7#NB06RGjjZ2_eby*5;PYO zFgr?D&^!O4f13X6&koO;3*r6jN<^@qt8BDs?_{BpT$oU&sB2Uge=r46)nqTX`;=Fw=h^2 zU(_ir3SR)%-eoLLv44seBdDvKc^q_I4ozQ!1~kV(O1Hmf3kkGG6ncUu>s6hjtIr>i z<$Rt!E6NJ#=#}?ioDr{hqe-xHKD!NFs+@t)%Ksjssl-|?Z=ud^ePbtmIZm|O%kF^W zyOkiQ{eHNk@Oe4%@XyS4tD>;az#}=&cFO<@E)=7ggPgAfd-&LfU<;qVnMN?Zx1nz6 zMW_i~24{r3{qSy*6p(k8?UQrv6_7I|=@UvBgONX`tafNxDQg-IZW)YWhI@pQmRpA2Rt$67NKDetqe*7qS-XZB@VfHA zm+-0VSjNG$=&+}srHv5wS5?STVu>)d7(Td+5wdpv3Sjg}?f8kLG1UAZ?HW0eYP(?L@RDB1GLG6D^!9%LKlW8V$xFXLy0X=3u@*GLB3yc++WF`{ zY5iMIa~&Yub+juL7vM(O3byG(5iQKd_=7dGti6Jv4d)QInf5dF z?Bz{-vZ^E+Yp(u|LffBBr7ahJoJoUCM~uu;aiYIh@oR@Pt(CR7U%O{vC2Hq{^A@yH zOL!Hxf#c*^dWtsa94kWyW{uVMf!}X$p%L+v{R;XbZUg#ciY;g-moS$gGwG~qkF_qf zPgd#;v>#{w@45Y3@6xKirz^W7HV}H|BNInTdt=idH#z`43nM*%or*!)#KXiSZ94WMJ)N0>dD$EUHc~ z=3;GaXkcsm0~Jv!n_D;${=EJx23kT@6Gx|?1IwpFy{~6nl|IhXO zws18e{7)cMgjV(j&gM=g286Ev_i=3WO!Q1t|3RHhY>n*YSRp@-Ce_DZ!aFZI7b68HRYH+rh(x zUfei16$OD-(FyMeiY^<}0a@q-hW-s(7YDQJL`EJ*9*F`{dhn_Ov8%}m8l1r_pu!#! zRBrp%|HvfT*fpQ0fSwH~E(R6jmATYbnFd?g33M#CT}gATbQ-rPX%3bcQ#*N2uP`oF z&R<5oC^u~SlC_nyA?ChlG7gyc-ef>ss%6k$599U8HNCc1?^eJe3&Wb%yt^+Z&f=IBmPJY3edu42u$iRo3Q}wm=^fLCX{7R#;zJ8InC>pD^AZJJwcY2 zSeX3ay+Cu{U7E0C$nJpu7+4zwuPLZN#Xna8V!p_!gchhQu?_b70hGa^QgBLl{#L0w zVXr2V7pVLoNEAkK*b(oSwG#nK?Uav1)K$Na1kUE7WYNIZ40GIphETE@H0;zZoqbU& zhU$JNL+R<73AgA7l0*TmtbOQkqN)$1C3RZBdN+(e4VIMd25y#8$m{MPuLpecveM$& ztj&D+nt-6_aloHH=f9Iom=6B9H*=bV$rQwM2%%ZkD>v8)&!!246()QNsS(GDGKZg7p9s@X zRmdtr^rx-X=ZKgxOcOaT1+Y=G@=!)#XkxmkO(TLF%?gf}`vdKD@+d2m1q-7kIisgz zow0%9D`Z6?&KF0L9PWYXsi`5A8jvpi!J@VxZIpK@##Joks<@%IU?y0tdGOUitW^^i zGuJOX*p3t6N{0h3%9Sa|s7btNWHqyvJm4d&x&nWyODgJr$tpV%W&?*em{2az zd74^ep8WI~M;z)EQeBN5fg#<>+9z!wNJkYGpqYm*P&KJltZ#u?hv|Q@=bBd`rLdVM zd5FetO2(L*tZa@4d~A&a6Bks9UEQ4(+3{~qW-(#MX*$1XRSv=+OGcLs^OG^_3rUihTzEI0XD5T7__2DS)axw}TW zTex3hmU}Twx%~hhOh=u2n>0?}RbCo=6}R(*>xX@EM~!Pou*&VIF!?%iXBpa=hKlM63trV+dHxnWRlnq5JLiBZZ%T>-)|a(CyRSIntJH zco_}zP=UK2(!R$*y>#Af`vT>10JWbw1c)^1ZYKQUQ(MF&5$JHMIKm|6FmxE)4;xBT z67PHOs)-~S`Tn*2?`;G;)FC1-MgzZr%hcs~(Sec|x0^|GuH>VI0)W@Vb$(o$;k1G$ zUhuWO;u9m!Ep8su#(8|4`kz733kESH#90!^Q-oIT>1%J7*e%}YbCub%j@A>2m=^AB zs{t$mi|U`?eV(P6_dqIE9zI_yomk8N6Ym_~i>0g*C_jz^Gc zbu$ic@F|su91$D)yQa#FrkYF)ALskwX<~t#tZ8}0vWx2ED}qbbO>5$T?6jdP-skyO zTdG{*C;}&g7tPf7@x6KAa422I@wVQ-cV}p~?{>d-zn8nKsjVKmPZG}$iGRn}CHy(t zyxyxi^&PHr*f~SPcE8{D4l*XTYGN@7Y*G1j_&%Mir6|3^3HNtCy38W}+eCun|8cKQ zUD3o8hC$NS*ue0J(az5KKcj3vXHopm zL8f;9+5Jy;@}FH1LM!Cc2P=~>n;%m3*gGFh6HHQtF!$jRg-eT}9fsi-1eL;__*KIM<8fI=pfFF{9@ zKuo6iQY0Z+s6d+CXbam|77t>h`4XB%yN@M?R6yle78nisLlh_zkLKW%)@#Gec<#z1 zgTw9sTOi{^r3or{*+@U{q5JK@agw#Rcj>nMaP^YqfH(41biq+=6tnR6>;DRL4vX<| zi+f!sU!6oV;G*qAJ4NV$9NK=glZ1ItK z;SNwln?O58*aVu<&ZCVH?gTApt7xNyt)LC<2HFVWF3^s46YVHr2k1n*g*HsM8+4)F zM;jvaL3gGFHE2f&_kbR>Mzle~ym$?y4WXSw)9v+Tj-p*e>m}4cjG`@~>CrTZ$Iz~#?ITnIj-zR; z^tF_{6KMC)dI**H%n@#CXc`ik<}{j%)=hW@Ord#b8qTxe99j!n7vXs@jV50y6&Jt^ z+8MM?!dY+;P2-|8mcSg^3|c$kJh+55izW*!fJL-rv^K&ea2ZWLkbRcH3fd-G3*i-z z4U<={MU(Yb!5Uf}S~KA_a2;(AnrwRm+(c6?6bWyE+h_x5viu!TMjJ+JBwPn~(NrHQ zmkqFqHjh?Mcn{o1TSrr=Jpfyo7CC6OgxeY7tk?_v4kwg6YBi?VV;3!tRv@I<;~ZK$ z8g(8E1+)&d9HDMdi*^7_AL9w)dbBY#myoBA8__1w^e_45aS=^l#vx?X;$}2VSkW^x z;&BVw2HF-O*^1lHNK}RI8#04R6F0dpwB7xKwTuQta^%8V0W5tC(VskD{%hT_>d2<54sw!3v&?Dfajn+Fi6Y zLW(^;j;1li&oRXwpFn$nc9oD~k58d(p-~Gl#U5wHo(n1L$_gRH9#5fZT&Sa%Vvo0#J zCR9Riqsc@1`Z7ZqjY+U_j!;G-JwB61MMEbOVsD=bpfW`$1M?XEDKrhc%ujm!X|&UX zS>f2zS+RHgITL_?ffJKx2AHIMX3WYdLJ#CX9uz;_%H2i;%~=mC4dKCmD3f&-ur90dK~5EuZ5!5}yShQKg53P!*v7z4+^ zI5-YYgHz@-F0u}KdIFpPCm9^2rwu51dW|R5?BC> zUF#a0isZI=Blqz$Ul{?t=$l3v9F4(W=jY)_NY$ z3NH_|uG6ATKA1FT=TG$sO{J}1mpS)_r8NMp=5_$B;hKO}ZaYCM*afsA(-Leq(89|H zEVqh#z(H^T)S1)!r}`f(yoy@&X_?2bP^E(-Rzbx+bB2GY>sU|~SwR&K1I;x1Ib!ut z95(0F$8;SFr6Mb%;uv5(R2&Daf{HAAiYEb!o#G^5iBmiSSlATL0+usH)-S~ifK^Lz z7O*ZU&I48>MV22$)*HoTz$&A71(d;UP%@|VGdum-My>tF#afpu17%GMoou9ozP zyI=!s0!?4Lfe(6sM)v{O0$3p6vjG+em=r2~W<5R9ut2~vKFFE#-doZNT%x5ewg|{V zkfrDSpVwI|5Rm<3L;OfCbmACR}8)ttwFq>J0l`P7h(ut1O*>9G$eB%wj4IUjvf zp>EJ+&L@`?;*y}-oG(l&)DpYLoG-5_+zZHba0nbQ=jDPSI_XxDT*JfGq-Sk@Bli=>+}e_I%Yb8ye&jX&uO$^}nT%IU=nA zT1)ks&9r70ue_kM4Q89KWZ^^4C~P$QohNj}RFKx2&7N0?X;Ms+GG(Mp87ZboF+pD$MZd59>NCknS;i z-{%xEtE4z5#n~yAN||_4Op{`jwBPLgPwHZPm9lQH95Va;uSp~4>7dyMKcWy{r9);< zKCBQorAN&^`nw9TS&GflF|$u?Wl#0|yHCZOlj5)xbEVo~O`5&*FFLDz&$!tOUsR~A z7PC*P%@k%$wS~f+DXvSk11gz)SIoZuU7clGO4)K(2F!kRD7)2vF6#t_ORr_xOA0YzdJ|+`d*^Q*a|4v#&Z%BKD(dO>l%U(``<$cPSfi&3&p(Vcy))-8y2|QO=p$ z{}Y95R?0PiZAQ7l+>sWYU?)=EVeV9iLN*-bouC!$0&FD89iS8J23^1hhe1#lWsZ$j+(puoWfIJ%-kFQqfqN& z{kL&&987={;3Sv=ljd$ct52K;XUzSz&nr9!N?;b8HTRMC=)`$24K9Ega2Z@Q_tAHs zWe!{d3t$l}ffaBCXve$?t^sX~ZvbtFZvkzB@33KG_gP*x_mBQex7SWsTi*?!4evdm zZLap+atS-!@-_o8Xx`)ho7wEMU&!#nigviy}8eR zPM0)*Msve0h1ep>7D>-fDr^GH=6>nx*+2U0f0L$tDBg%#%>DAmrQx9n=R`Os!W)rq z?#tiRS^N_1G558Hv%9?U!wkQ8L?>`ggh8Uc=Kk{^O2bGI{k6&wbKkltZ2%lJ_w99s z{os&!^?#@EC^&3h!#^n;1UM|hQqiz^%x~GzF)(6Y$GhJR|3%m*8aHqMy90)=q6zax zA9c)|`mYQNf1?w)DLQH1*j0tNGs05Q8S~D(Bn|&X+LsNPH~*qE+!UQR@6rmv%WPlP8UJQ-oBXw|&men@ArXoNc>+!@K8*fClMci9Ou`$P}S``~eXLT#39 zPUuH-*m7$hatJ;1o_SeEIZy{&^PVb7%Yy>g0qV{B_@C%R186kw`HI38P&DtwM-^&S z)n(o{UdzVn?Z>2bfL8Oq@;!xZU>BfwQ)xHv=c78|gHH3_I;(Iu=(1e?9ffvQaID13(RpgWwQQ-{LS(x8ev;k75|;VJ7#g*p@CK>dgb zpe}@3B`1Lz5R-tmL+K2l*-)ZuP+9=zEZ6-L-EN>gs zrJLZY<;H(1Z56Q9FI@v{@JqMBCb(s}vwL*i9Z&}A^Z=yYv)p2hj_w0ZOCy&1hi0^p zbcbfSyK@R%fGxrrPzx|fNcO`Yvlo3Ta}6q&uW_6_Veu9+2b^PljYVBnu%X49R(juR;a=-VCKBn#wRteP~!Y`q^LS#QAtDzb_ z>gP#0 zTfXs^I_d&_pffxBZE5TW{cb>Y`}FI48kaupcc0%GpI)xdD%U?~`G!~@W3A%T#jcE5 zzT?-Wv9tDhqvMZ)!~{f5P(nE-E|$ z=)d^1iv81ow*dZ>Qb&sl!-ltNw(`E>k!x~u*In6v!k_jHy{s(;Dy zGhb9l6V+d|{On^2m%wGdn^si~gzBj(p;tIk0(eLrcd%INhJ8LZ9$y#Tlb$Xit}sz^$eq*V3JhI~{>MOCk?^xluLuQKdahP|ry zetLPOH#!V^m0_{Y!TVc4q-dsQ!Z^j1r+{AlA=X`)t{xRY#jxAgTBsym^G6S}$y zm7Gw?36-2s$qAL5@FFN-tP<>#;Ee?PBzPmiJ_*iAXeB0e786Q5p}m+;zX?Cq35A`o z?@s95B-C$0!6p=JLT@;sY7?q9p=uMVHlb=0sy3n66J9VT6l_AlCRB1l4?dw_6ACt= zLX%;ud$68;)%QN@{6DN(U0VueZe(+Ga%Ev{3T19&Z(?c+b97;Hba--QW(qbjHaIpo K3MC~)PeuxvyMP=3 literal 0 HcmV?d00001