03-model-building-02-64x64.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:03.817619Z",
     "start_time": "2020-11-21T17:23:01.554136Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt \n",
    "from PIL import Image \n",
    "import time \n",
    "from IPython import display\n",
    "\n",
    "%matplotlib inline "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:04.321788Z",
     "start_time": "2020-11-21T17:23:04.317798Z"
    }
   },
   "outputs": [],
   "source": [
    "import utils as utils"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:04.633952Z",
     "start_time": "2020-11-21T17:23:04.612012Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From <ipython-input-3-17bb7203622b>:1: is_gpu_available (from tensorflow.python.framework.test_util) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use `tf.config.list_physical_devices('GPU')` instead.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.test.is_gpu_available()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:04.841414Z",
     "start_time": "2020-11-21T17:23:04.836427Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.config.list_physical_devices('GPU') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Prepare Dataset "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:07.186680Z",
     "start_time": "2020-11-21T17:23:07.182690Z"
    }
   },
   "outputs": [],
   "source": [
    "BATCH_SIZE = 32 \n",
    "AUTOTUNE = tf.data.experimental.AUTOTUNE # Adapt preprocessing and prefetching dynamically\n",
    "SHUFFLE_BUFFER_SIZE = 8    # Shuffle the training data by a chunck of this many observations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:07.932717Z",
     "start_time": "2020-11-21T17:23:07.344279Z"
    }
   },
   "outputs": [],
   "source": [
    "filenames = utils.get_filenames_list('data/dataset64-easy-gaus/')\n",
    "train_ds = utils.create_dataset(filenames, SHUFFLE_BUFFER_SIZE, AUTOTUNE, BATCH_SIZE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:08.016495Z",
     "start_time": "2020-11-21T17:23:07.958649Z"
    }
   },
   "outputs": [],
   "source": [
    "from tensorflow.compat.v1 import ConfigProto\n",
    "from tensorflow.compat.v1 import InteractiveSession\n",
    "\n",
    "config = ConfigProto()\n",
    "config.gpu_options.allow_growth = True\n",
    "session = InteractiveSession(config=config)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model building - attempt 01 "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Architecture "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:08.201000Z",
     "start_time": "2020-11-21T17:23:08.197010Z"
    }
   },
   "outputs": [],
   "source": [
    "LATENT_DIM = 10"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Model architecture from tf tutorials on MNIST dataset "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:08.660277Z",
     "start_time": "2020-11-21T17:23:08.543590Z"
    }
   },
   "outputs": [],
   "source": [
    "encoder = tf.keras.Sequential([\n",
    "        tf.keras.layers.InputLayer(input_shape=(64, 64, 1)),\n",
    "        tf.keras.layers.Conv2D(filters=32, kernel_size=3, strides=(2, 2), activation='relu'),\n",
    "        tf.keras.layers.Conv2D(filters=64, kernel_size=3, strides=(2, 2), activation='relu'),\n",
    "        tf.keras.layers.Flatten(),\n",
    "        # No activation\n",
    "        tf.keras.layers.Dense(LATENT_DIM + LATENT_DIM) \n",
    "        ])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:08.816859Z",
     "start_time": "2020-11-21T17:23:08.718124Z"
    }
   },
   "outputs": [],
   "source": [
    "decoder = tf.keras.Sequential([\n",
    "        tf.keras.layers.InputLayer(input_shape=(LATENT_DIM,)),\n",
    "        tf.keras.layers.Dense(units=8*8*32, activation=tf.nn.relu),\n",
    "        tf.keras.layers.Reshape(target_shape=(8, 8, 32)),\n",
    "        tf.keras.layers.Conv2DTranspose(filters=64, kernel_size=3, strides=4, padding='same',\n",
    "            activation='relu'),\n",
    "        tf.keras.layers.Conv2DTranspose(filters=32, kernel_size=3, strides=2, padding='same',\n",
    "            activation='relu'),\n",
    "        # No activation\n",
    "        tf.keras.layers.Conv2DTranspose(filters=1, kernel_size=3, strides=1, padding='same'),\n",
    "        ])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:08.898640Z",
     "start_time": "2020-11-21T17:23:08.894651Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_1 (Dense)              (None, 2048)              22528     \n",
      "_________________________________________________________________\n",
      "reshape (Reshape)            (None, 8, 8, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_transpose (Conv2DTran (None, 32, 32, 64)        18496     \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_1 (Conv2DTr (None, 64, 64, 32)        18464     \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_2 (Conv2DTr (None, 64, 64, 1)         289       \n",
      "=================================================================\n",
      "Total params: 59,777\n",
      "Trainable params: 59,777\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "decoder.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Utilities for VAE "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:09.244760Z",
     "start_time": "2020-11-21T17:23:09.238250Z"
    }
   },
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def sample(epsilon=None):\n",
    "    \"\"\" Forward pass through second half of network.  Used for generating data samples.    \n",
    "        Sample epsilon using tensorflow utilities \n",
    "    \"\"\"\n",
    "    if epsilon is None:\n",
    "        epsilon = tf.random.normal(shape=(100, LATENT_DIM))\n",
    "    return decode(epsilon, apply_sigmoid=True)\n",
    "\n",
    "\n",
    "\n",
    "def encode(X): \n",
    "    \"\"\" Forward pass of encoder network \n",
    "    \n",
    "        Input is data vector, X.  \n",
    "    \"\"\"\n",
    "    mean, logvar = tf.split(encoder(X), num_or_size_splits=2, axis=1)\n",
    "    return mean, logvar \n",
    "\n",
    "\n",
    "def reparameterize(mean, logvar):\n",
    "    \"\"\" Forward pass through bottleneck section of network.  Outputs z vector.  \n",
    "    \"\"\"\n",
    "    epsilon = tf.random.normal(shape=mean.shape)\n",
    "    return epsilon * tf.exp(logvar * .5) + mean\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "def decode(z, apply_sigmoid=False): \n",
    "    \"\"\" Forward pass of decoder network.  \n",
    "    \n",
    "        Input is a sampled z vector \n",
    "    \"\"\"\n",
    "    logits = decoder(z) \n",
    "    \n",
    "    if apply_sigmoid: \n",
    "        probs = tf.sigmoid(logits)\n",
    "        return probs \n",
    "    return logits \n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Optimizer and loss "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:09.593337Z",
     "start_time": "2020-11-21T17:23:09.586355Z"
    }
   },
   "outputs": [],
   "source": [
    "optimizer = tf.keras.optimizers.Adam(1e-4)\n",
    "\n",
    "def log_normal_pdf(sample, mean, logvar, raxis=1):\n",
    "    \"\"\" Utility for ELBO.\n",
    "    \"\"\"\n",
    "    log2pi = tf.math.log(2. * np.pi)\n",
    "    return tf.reduce_sum(\n",
    "      -.5 * ((sample - mean) ** 2. * tf.exp(-logvar) + logvar + log2pi),\n",
    "      axis=raxis)\n",
    "\n",
    "\n",
    "def compute_loss(x, y): \n",
    "    \"\"\" Calculates model loss \n",
    "    \"\"\"\n",
    "    # Forward pass through the network \n",
    "    mean, logvar = encode(x)\n",
    "    z = reparameterize(mean, logvar)\n",
    "    x_logit = decode(z)\n",
    "    \n",
    "    # Calculate reconstruction loss (binary cross entropy)\n",
    "    # Reconstruction loss will evaluate the similarity between inp and out. \n",
    "    cross_ent = tf.nn.sigmoid_cross_entropy_with_logits(logits=x_logit, labels=y)\n",
    "    \n",
    "    # Calculate regularization portion (ELBO) \n",
    "    logpx_z = -tf.reduce_sum(cross_ent, axis=[1, 2, 3])\n",
    "    logpz = log_normal_pdf(z, 0., 0.)\n",
    "    logqz_x = log_normal_pdf(z, mean, logvar)\n",
    "    return -tf.reduce_mean(logpx_z + logpz - logqz_x)\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Training functions "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:09.951377Z",
     "start_time": "2020-11-21T17:23:09.946390Z"
    }
   },
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def train_step(x, y, optimizer):\n",
    "    \"\"\"Executes one training step and returns the loss.  \n",
    "\n",
    "    This function computes the loss and gradients, and uses the latter to\n",
    "    update the model's parameters.\n",
    "    \"\"\"\n",
    "    \n",
    "    # Use GradientTape to record a computation graph \n",
    "    with tf.GradientTape() as tape:\n",
    "        loss = compute_loss(x, y)\n",
    "        \n",
    "    # The tape is used to compute gradients using the computation graph \n",
    "    trainable_vars = encoder.trainable_variables + decoder.trainable_variables\n",
    "    gradients = tape.gradient(loss, trainable_vars)\n",
    "    optimizer.apply_gradients(zip(gradients, trainable_vars))\n",
    "    \n",
    "    return loss \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Utilities for image generation\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:10.484458Z",
     "start_time": "2020-11-21T17:23:10.479471Z"
    }
   },
   "outputs": [],
   "source": [
    "num_examples_to_generate = 16\n",
    "batch_size=32 \n",
    "\n",
    "# keeping the random vector constant for generation (prediction) so\n",
    "# it will be easier to see the improvement.\n",
    "random_vector_for_generation = tf.random.normal(\n",
    "    shape=[num_examples_to_generate, LATENT_DIM])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:12.191410Z",
     "start_time": "2020-11-21T17:23:12.186423Z"
    }
   },
   "outputs": [],
   "source": [
    "def generate_and_save_images(epoch, test_sample):\n",
    "    \"\"\" Utility for saving image during training epoch   \n",
    "    \"\"\"\n",
    "    mean, logvar = encode(test_sample)\n",
    "    z = reparameterize(mean, logvar)\n",
    "    # Use the sample function to generate a new sample z vector \n",
    "    predictions = sample(z)\n",
    "    fig = plt.figure(figsize=(4, 4))\n",
    "\n",
    "    for i in range(predictions.shape[0]):\n",
    "        plt.subplot(4, 4, i + 1)\n",
    "        plt.imshow(predictions[i, :, :, 0], cmap='gray')\n",
    "        plt.axis('off')\n",
    "\n",
    "    plt.savefig('image_at_epoch_{:04d}.png'.format(epoch))\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:13.567249Z",
     "start_time": "2020-11-21T17:23:12.401847Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Set aside a sample of the test set for generating output images\n",
    "assert batch_size >= num_examples_to_generate\n",
    "\n",
    "for test_batch in train_ds.take(1):\n",
    "    test_sample = test_batch[0][0:num_examples_to_generate, :, :, :]\n",
    "generate_and_save_images(0, test_sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Trainin' "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:18.192919Z",
     "start_time": "2020-11-21T17:23:18.189927Z"
    }
   },
   "outputs": [],
   "source": [
    "epochs = 100\n",
    "num_examples_to_generate = 16"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-21T17:23:18.393889Z",
     "start_time": "2020-11-21T17:23:18.389899Z"
    }
   },
   "outputs": [],
   "source": [
    "losses = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2020-11-21T17:23:18.579Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 32, time elapse for current epoch: 4.908487558364868\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOwAAADnCAYAAAAdFLrXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAACYSElEQVR4nO39x26kW5qegb5haMIbmiTT7d21q7pLqiq1pB7pJnQLugpNNdAd6A7UM40ECBrqAgQJEFCTagmq2iYzdzqaYHgGyTBnwPN88cbKYDD+1DlobIALIJhJBv9/mc++n1m5xWKhp/E0nsYvY+T/sSfwNJ7G09h+PDHs03gav6DxxLBP42n8gsYTwz6Np/ELGk8M+zSexi9oFDf98vnz54ubmxvd3t5qPp8rl8tpsVioUCjo7u4u/p3P55XP59VoNFQoFNRoNPT8+XM1Gg2dnJzoD3/4g5rNpg4ODlStVpXL5TSfz7W3tydJmk6n6vf76vf76vV6WiwW+vHHH/V//s//0Y8//qjBYKB+v6/xeKybmxvN53Pt7OyoWLyf/mKx0O3trRaLhXZ3d7W3t6d2u60//elPuW024V/9q3+1ODo60rNnz/Tq1SsdHh6q3W6r1WqpWCyqUChoNpspl8vp7u5O/X5f+Xw+3jubzVSpVHRycqJKpaJSqaRisaj5fK75fB57lMvlNJvNNJ1OdXd3p2KxqLu7O41GI/V6Pd3d3enm5kaDwUC7u7sqlUoqFArxnOl0Kkm6ubnR5eWlzs7ONBqN9B/+w3/Yap2S9Pvf/37BvCWpWCxqOp1qOp3q+vpas9lM+XxexWJR+Xxe+/v7KhQKqtVqevbsmer1ul69eqW/+7u/U71eV7PZ1P7+vmazmW5vb1UsFuN8R6OR+v2+Pnz4oHfv3unDhw/6+PGjLi8vdXd3F3R1e3urfD6vcrmsSqWinZ0d5fN5jcdjSdLe3p6KxaIajYb+y3/5L1ut9V//63+9aLVams/nurm50e7uru7u7nR1daXBYKDpdKpyuazT01M1m03l83nt7OzE2fG+58+fq1QqaW9vL85vMpno9vZWuVxOOzs7yuVyur291Wg00tXVlW5vb3V9fa3BYBCf73Q6Go1Gms/nWiwWyufz2tvb0+HhoebzufL5vAqFQtDUf/yP/3HtOjcyLBOaz+dBsDBuoVCQpPiey+WUy+VULpfVbDaDQQ8ODlSr1VSpVFSpVFStVrWzs6NCoRDPnk6n2tnZ0c7Ojvb395XL5TQYDNRut9XtdjWdTjUajYKQ5vN5bOpisdBisYh/7+zsaG9vT7u7u9vScBzO0dGR2u22Tk5OdHh4qFqtFnNdLBZxYI1GI953c3Oju7s77ezsqN1uq1wuq1wuhzDxvcnn87He2WwWzDIej1WtVnV3d6fJZKLhcKhSqaRSqRR/M5/PdX19rcVioclksiIQsoxisRj7PJvNtLOzE+9dLBbBsPl8Pj5frVZ1eHioo6MjtVotHR4eqlwuxxzL5XKcgwsYGO/m5kaTySS+7u7udH19HXuzu7sb74E+FouFSqWSJMXv9/f3t17nycmJXr9+rXw+r8lkot3dXU2nU33+/DnWure3p5OTEzWbTUkKOoWmS6WSDg8Pg544C4RNPp8POkPwdjqdEMjD4VC3t7chhKEVeGlnZ0eNRkM7Ozva3d3VYrHQdDrV+fn5w+f32OEy0bu7O0mKQ57NZiuMwsaWy+XY+Fqtpv39fY3H4zjM2WymcrkcBIPGub6+1u3trabTaQiEUqmkRqOhyWSi8XgcjIrW4t2z2UySQsMiEbcdaBi0NMyVy90LuUKhoFwuF0wI0c9mM43H49gbnuFzRBPzrNlsFkKP5+3t7Wk6nYZWQ2NDoPzN3d2dptNpMP1iscjMsPv7+6rX60FUMOz+/r5ubm40nU6DWefzufb399VsNnV0dKTj42M1m01VKhVNJpN4NwLErYjFYqHr62tNJpOgmWq1qkajocVioV6vF/vIWZZKJVWrVRWLxWAqhLALwG1Gu93W8+fPlcvlNBwOtbOzE3uGdQhD8k4YFsaEDhFAd3d3YWWiPPh9oVAI2ptOp9rf31exWAyabjQawbyTyUSFQiH+Zn9/PyyAu7s7dbvdB9e1cReq1arK5bKur69Dnc/nczUajRXTCUKt1+s6ODjQ8+fPdXp6qoODAxWLxTBB3LycTCaxiTAsUsjNsNPT0xVic9MQkxKzPZfLqVarqV6vh3TeZqDVhsNhmNx+QAgKzJa9vb2Yey6Xi/VDZG4qpf9GQEn3Ag7hsLOzI+meSXZ3d4NIEUaSwuzkXQi5LAM3ZbFYhMkJQ8GArH88Hmt3d1cnJyf65ptvwiTmb9kfBBYCHIa9ubnRcDjUeDzW/v6+Dg4OtLu7q+PjY11eXmo4HGp/fz+IW5IqlYpms5mGw2EID6yuyWSy9TqPj4/18uVL5fN5dTqdOE+EXrFYDGsQhoFh2QOEI2vE/UH5SAr6QHMWi8Uwn8vlctBTPp/XdDoN89jpqlQqBY1IS6t13XhUw2K67u/vx8HUajVNp9N48M3NjfL5vKrVami3SqWivb29kEAMNCK+jv/c31kul0MT1Gq1kNRsIlpvMpmsLBQfMov5hBCZzWbxXDcLkbQwLF/OjGgD/zd/y3N8rf6e9Nk8I903nsd6fY5Z1sq5IQClJaNICr8SYoKQsZh8/yE83ye3ejB5fa2udfFPMXt3d3dDCGE68s4sWXnQEb4izyqVSrE2GHVvb0/7+/shiHm373t6Dm6FMDCRXZH5/mA97e/vhzDa3d1d2Ut3N9eua9OiDw4O4uGo9t3dXT179mxFE8Gwx8fHYca6hiuXy7GJkoLgWAi2O8+bTCYhANDAe3t7qtVq4cdhDiPd+v2+JIWgyKJhZ7NZ+FWY39fX19rb24s5wYQQI0zFWjBxOUg3rX3N+K+Y05hffB4N7s9FOvN/NBxaKMt49uxZ+GWsHcYEBEKbYOVUKhW1Wq1gHAcMYQyEF64O2kxSCH1MRkxwNB0CbH9/P/YYocHfAxxtO/Ap8Q0lBe3CLDAqAgXAxzUxgt99etbCOWPhwSeOzSDcSqWSFotF+P5oaxQf64cuHhobGRZ01wEhFoc/cHNzE9IICQlwADGxQIAn18zSkijdvMzlcmo2m2o0Gnr16pVubm40Ho81Go3igCuVSpjFmIjD4VDSquR7bAyHw5C4l5eXgc5imnKY7gehLWA+vlgPfhnM5v4sZjzvhajQmIA1mNA8k89ICiAOQbXt+Gf/7J/pm2++CYCM+Q2HwxAc7D/MCMA3mUw0GAx0d3cXP9/f349zYI2SgvEw39FknD97Nx6PV4Au1s4XvvB4PN4IxqSj2+2q0+moVqvF/EFl3QWR7pkb5Bp66Pf7mk6nOjw8DPMZkA3LMZfLBV2Ox2MNBoPwzSuVSgBb9Xpd7XY7rAjM6uFwqB9++CGUU7lcXpnXurGRYe/u7laQSDfFXCqhBZGkfnD4OWw+h+nDNRHwOBIYosWXQHpBMBw8INDOzs6KENlmXF9fq1wuB3FBuHd3dzEvGIh1pJoTjbC3t7fWdHOTmf9DqC5cUs3qpjEHjeTG984y8KfcHcEUS4EdF6TsyWw2083NTfxNeqYuaPi30wBM4ci5m+mcqSPO7iduO9yv9lAKc3ST1p8NToDWRKn4mabmK/SO74325XeswYWQnwPMz7M3jY0MW6lUguv39vbCxkdjuk+AKYBNnmoeJ9iUCXwz3cTmEDGB3WzDzAC2n81m2t3d1e7urm5ubjJpHrQ6UhJABRcAYnbEHCZy84l5E6ZKD8DNajf5UsJNtZWHhNDOfGUBYqRliMSJJJ/Ph1WBleMIPogyFg6osDNp+uXuA3NG8OKy7O/vR1yes3ZLy10B369tBhahazRJQcPua08mE/V6vRV3gMgGwgkhyRzZK/d1QXgLhYLG43FYMLgd7A3M6jSVCpCHxqM+7MnJSRwyjFWtVgPJHY/Hmkwmurm5WYmNdrvdICo3oR1Y4d8sFp8MRDqXy4XGKhaLOj4+Xtl8zFVHsNGSj0kqHxDjYDCIJI3BYKBWqxXmKoIKoMtNvpubG11fX0dsEfMvBZ8QTmip0WgUoSR8HEkrUtsPj7XBONfX16EBth37+/vBLAhWzhQg0JH76+vrEGi9Xi8SPABy3D3AwnIg7O7uLoRgqVRSvV6P0M7Ozo5ardZK0oC0FNo8m/3KwrDlcjkALcI4CCZ367AAzs7OdH19vRKiA/wEuQfTkJY+rdNAv9/Xp0+fNJ/PVSqV1Gw2I0HDXQxo1U1zxzigo3XjUQ1bqVRWgroef3I1j2R0ExFJzSTZtNQs5PO+GWgsiMgd9xRa93gsZmyWuB2SkoNl3i713PRzEAri5gD4jJs4LjxIQnHiRPM4+OT76YyONkcAZo1PInxTszg9QzdtXRi6hcMepO4Ca8L64IwBdNhfScHgCA8XcqzZP7/tgFkdAPJ18z6Gm7zu4vn+OprrZ+fRC48SuDvjQix1R9hfp6uHxsbTbrVaarfbYVagLTErOCzXGviSaALMKbTxaDSKzBGegXTq9XoBKrmkQ3NDKGgfGNUJCymfRfO4echwa4DNhqA4pNvb24jduq/jiRD+fPaFuPZoNFr5TBragSnQBIQ8crlcaNushOzglptpnpyAxdDr9SLT7Pr6WldXV4Fsj8dj5XK50MCYmQjn29vbSJ/sdrva3d3VbDZTv9/XbDYLtB9NiCslLRF30HhoKAuQ6ESPdUTkAFrt9/sajUb69OlTpL6SjUX2mwOI7JHTCs9yawfEmDmUSiW9ffs2ACtoX7rPdSDiMhqNdHNzo16v9+C6NjJso9HQ4eGhSqXSioRdLBZhe2NSDAYDXV5exuIcGDk7OwsQSbqP47ZaLR0cHAQx9Pt9nZ+fR5aHx/pgzE6nE/HZ/f39SAGEiDGfrq+vM4UAsAJ8QzGlpKUGZ+0w3mAwULfbDeabz+eqVqvBtB7OkBRodr/f18XFhc7Pz7W/v69Wq7XiI+F7eTCd1L18Pq/hcKhCoRC+VpbR7XY1mUzUarVUq9VWBC5ENB6P1e/39fHjR52dnYVQQlvCQPV6PdB8hDjCCpT23bt36vV6Ojw8DK3Fu9rttmazmer1emQckSSBlbVYLCLck8UkxgxnTrgBaHlynD98+KCff/5ZFxcXurq6CkFSqVR0d3enw8PDmA+Kw/EV3JbxeBxrBgg8Pz/X+fm5Pn/+rHfv3qlcLuv58+d68eJFpL4eHR0FKn1+fq6Liwu9e/fuwXVtZFjPk3WTFWJy34M4JloTk4hNQ/J6mCIFVQgvSMvYnYcAkLKpb8h3N1eyxCc9g8UlI1qWQ2IPUtMU8520S/7WTUtfJ0IFoM7NcbcWGPwcc8vj1VlRYvw0GM+1LFYMWm00GoXWSWO0hDqYn/RlIgfx08FgEOCSnx00QwiE4WZi+tltBznLbp1IWlkDz+z3+6Eh0eooqOvra9VqtRX6ctCJcJGDkR4Rke7pq9/vxxwccHKU2BXDQ2Mjw2J2IU084Iw/6X6B+yeYuusyVPxgU5MM5odJ3XxI0Vg2Kh0uGLYZ7k/CKF5E4L4pzOYaZ50Qcn809QvTeUIYk8kktAkmIZ9xtNYJLkuoQ1KY5NfX1xGsxxqQliYeIA9fvCdFSt1/dWLmTBFICIf0XLwoAL8zn89HXjNMwD5tO1yY8n+eg6C7urrS1dWV+v3+Ch6AMAKYc7eItbNeTG00OH8Dw7uLI32Jb8C8DtZtUjYbGdYJBobM5/Oq1+uhKXix+51IGc/b5HM+WRbmCGMul9N4PF7RsDDpcDhcAZQ8tOCSzQlsm+FS3wEAB9fQgpihEPW6OLWHNDxbxuOWHIw/k0SKcrm8csh8DjQeZko18TaD94xGo0hUZ69BZ8ETSGrgPYCP+MGYiJ5Kh6D1tFZcJFwPxzn6/X4QPYIjn8/HGU4mkzj/rAzr8U7ORbpXRFdXV/r555/19u1bdTqdlcQGsrtAd/3c0jNlPaDvPIPfwSep4EWgoxhcKW6i3UdziTEPPH7kZhkEA4Fh2mFe8hw2zLN43LxF8qBRnRD9PRC4h008Xuu+4LYDs571QaRumjGPND7pQXHm4KmLKerJ+zyEcnt7G+VfWDGORqfhE84CMyrrWjELU7PTzX0Xxn4OzAPmdTQ9dVXQWlhC/K2fKQIZoIa9cyuMubpgeGwg/B1cS90nPz9AsVSbIoBwCdZZjNCuC1nOmM+mQngdMuzzeWhsPG1AEZIn3CR0CB6VzuKQiMRDYQIyXjznlgGhIvUBU/BB3NxwHwLIno0jjprFt3PBRH6pg1psJloe/5OEAEx4ip89AcKZi70gRQ8QxJ+Fvygtkwj48oQJEhiyjvF4rF6vp0qlonq9vnJmWEkQp4dAPBzlQBoMy3mkTFsul9VqtcKfdaHvwseFBH+7WNzX/n78+DGQ3G2Hh+YwV/P5fIBR7CPCh/+79bC7uxt58YBqqUXBXKlGOjo6iugI2pJUxUKhEH7zYDAIHsB6ACXeJJg2Muwf//hHnZ+f6/DwUAcHByqXy9rd3dV8Pg+CxjQYDocrG44ZAPB0e3sbpUvUADIxTCLqLGEQAtL8n8953uV8PtdgMIh3djodnZ2d6ezsbOvDhcBAbI+Pj3V0dLRipgHdd7td9Xq9FQJn3WgTZ3iC7TBwrVaTdK81zs/PQ3pToA/BEP/m/fi5/X4/klLcR9t2/PDDDzF3kiUwUQGy8D8BAhEyrk3ddMOXw9/nnEh0KZfLUdjN8zzxZj6fq9lsfqERZ7OZBoOBPnz4oB9//DGTED4/P9dgMNDp6anq9XqUur158yZoZDabRRXap0+fNB6PA43HJK5Wq9FZg8ISwCLpnnZhRvAHwmEIWQeTyIl2zAKBnVqt68ZGhu12u2o2m6rVaoFoem6kO9SetMD/PfHbkx6k1TKvQqEQEhsfCikFoeC/pWVzHrtz0yKL9vHnOqjippmbqZj9fJ6fwzyuGdkvCN3Nd38efpsXTbjphTZCI6F9smrZXq8X+dIp4p6uJY3zuomez+cj9Y6RhjscPcUqARvwZBMHptC8qekI2rztGA6HX7gRklYUAftHLJnhpjtWoe+/Z+il8Wz+79YDxQu5XC5ivCg+Hx6rf2hsZFjMM8wFzAoPdMMgacaSo6QeB4PwPeyBlMLZ73a7KyijC4X02fwcdNNRxywDUx3QwDshuEnPF0LLfXg0otc4SqtZLg5SVCqV8N0cgPLwkZucaRWPC4ltxzph4GEInumhLv85REkesOcgu9lP7jfZcrg6/n7W66CSryvVPFnWmob9+D80zXslRdUR+wyYhsXk+ddpthN0imVF+aBrYawzlArRDy/LQ1B5VGLdeLRahxxbGmTlcrlAwxaLxUrCvB+uE1iz2YyWMRAHDr6kqPY/ODiIoH0ulwtYnI1EYztAISlMDbKrkJpZhvstFDp4OZRnJ6GhWAsmIaawM1ZKOBwwrVo4TDQa2oRuD8yLfSfEBWNnHQhYN+VdCJDTzT46kzjA1mw2Va/Xo7TOf89+lMtlNRoNjUajOO9+vx/vJvTh8V7qXh3c81DI16wzBUm9YAM6ZW9dG8/ny2Z/DoK5VcJ54ibW6/UVi4k1sAckSUj3qb/49dLSVdzUfGEjw3oCNcTjEgCpl0piJ1KkF0QMwXl4gOehjeli5138XKO7pnWp5dZAiuRtGn5QaaXMOjh+3TohUv7OzWJGGn/zv0t/7mEhNLzvsReHZxngEOvMfT9TtyA4AwgfkCV9t++HCzN6fVFQnrpUjsI6MzDSaMO26/Qcc6cZN8URVCRNwMDshQujFCH26AD/X3fGMLwX5Dv9e/zaAb1149FMp+FwqFwuF43ViKs6HI9GLZfLUZqG9Paq/r29vRUhAPGzCYAyDjRh2qYai4W533V9fa3Pnz/r48ePmQAKBIozZ5o4QFwSiejtVj39Da2c5pwyIGKEYKlUig4XaBPQdSSutGwFSwwVIC/rwGry9RBWcuJGSO7t7YXmx1XAxEWIIvAkrTAdwqhWq4VvRj65+8Gcq8edmcdkMtHl5aWurq4yuTkIG7RpsViMfkqeLOHplAgG6IFyTXfjUiGdMqm7AmR3QQsuhGmJypqgOSIOD42NDPvmzRv1er0ogQJBY5IeS6OcCIIguEweMhB/o9FQrVYLEIqDwpwAeoehOVxMNeZBWMLT66bTqa6urvT27VuNRqOtD9frfTFRMc19c+ksMBqNQiB4lgtz84wdZ1pnVlreoF1gRgQk6XCglZIioeH6+npF42UZHz9+jD178eJFhJdgHISA16pyzpTDkS+NRYRARqPxO+a9s7Oj29tbVavVyJnlfBBKTtD4cwB24/FYFxcXmRj206dPury81OnpaSiP6+tr/fDDD/r06ZOGw2G0b+XcEaLlclm1Wk3VajUYDGHOfrsV4cAWoSA093A4VLfb1YcPH6LxIGWKt7e3+u677yQtC+cfM/83dvBy38LzSV0DebqcI3AOpFDF4IiumwIupRxwcfOTOfhzPUGBd0tLLbLtINXQzWmPK3JogCQeg0YqM2dpCTK5ycTz1qGqDp6hbXmWWx+OvkvLNMIsA83MPhJHx2oAzeUcPe8YwsX3XHem/uWIuJ81P+dM0aa8y0FGCjuyIv+EBx1AolDE/XMEkjctcKALGkOAuLvkkQDXsmnmHutyWvL9d0HOWh8aj+YS88VBs7mSApXFhPGFcphslJuK7sSzYWgnzC6kPn2JHeTxXFrCBh6z3YSyrRuO/noSPnNz4nXf0U0kPyQnWj9Ql5yg7V4hlJqVLt39nQ7GZNWwCDaS4yEkhBwCw/PCPZTEGh3Bd0J2IuZMOU/6EhOzdx9uOp2uJKS4kHTBnWU4PTkNggmgWBAm3rOMv/VuinylQteZ1d0dYr+Et2De+Xwefiq/g87cVVy7pm0OlyAvKJ6n7TmTwrwQsROzmw3+c54Dc9PEzUEl/Ds2ymO8mHGOmqaa7LHhz4OQeY5nbfnaPYYMcaWM6dLTB3tBySFAUJr5gxTGVHKC8bVmGc6wLoRc6LkQdsAHE1VSpJC6dmE/2FP2iSgAuAfn6YkYnAGdKZxOHHDbdriVxvxwpcA+MFcd3ffOIpi43ic6dXHWAWW4Cp5EwzpY/83Njdrttur1ejQRd2viobGRYTkgB3Z8A1x7uGPumtVBCZeoTtipGeWSGTCIZzrS50F9/g4NmYVh3cmHIfwwMI39UDxu5lYCv0vXyOBnbiojxb3wHYvF/Tq3RHz/swwETMqUrNVdFWmJGnsmUwqkrEPEU/OYuUL43nPJzxOt6oUAX4OIo0W9qZ67MqyRfaa4vVAorLzbzzQdLpjX/c7RX3fd3DyGT1IX6KHxqIalFAmgBbDFJbXn/pJz7CaUx7hSCcV31xiedeQdENGmw+FwbWvNu7s7XVxcqNvtrjDzYwMGxTxynwkTzQPtKULodZfputZpWBdsgC5IY5iHxmDVajUEp5eFkbGUVcMuFouIfWKCUtqGRnLNSoiDfYahHjpPF3A+MAFpF0o+NZldnvCCrwqq3Ol0VuL824yDg4N4/u3t/T04V1dXuri4WIkLOyLuysYTRVhPGtbxdSOQAFlZK+dWKpUCm6BzC105SLa4vLzUxcXF19fDkiRwd3cXYZ2Tk5NAv8iZJKFgOp0GAXpqGxtCVoznqUpaYUjMJ/KU2ah+v7+SNIH0IqGDQyDpIEt5nQMB+Ol3d3cR+kAYuaXh5hqJFaSdcdju06TCabFYdu0gaYQ9o0UIpvB8vmzc3e129f79e/V6vUhezzLoqDAcDlcuaELoknDCfniMerFYxN8fHR2F5eNm4Tr/i5Afghsh7JgEHTwI68HIDgZlEcKtVkunp6fR6aLf7+v9+/f685//rIuLC+3u7ur58+crJiuljbncst8wIS/2BI3pliVMvbu7q3q9HhbD3t5eXJz25s2b4CVo8+bmRv/7f//vKI74/Pmzzs/PNxY5PFoPm/qqqZ/lGsircyB8kEegbo9juXnNSP1fzBa6Wbhv4uaYV85kTU0ko4rnssbJZBJCC0HhTOlmKpqHOacmsX/HHGIfeZb7y4VCIdaEn+suipS9UF9SxCQd0OG5nJXX3ToKzFpzufua5UajsfJsN/M5G9bqjA0CDJLL/oN/pMkjfD4LIk4HTheQMC7gDudaKNxfH0JbGLecHIRMTXx3D5i/WyJeT825ua+KdYq7USqV4ucPjY0M68FiCMwzldblhrq/4M9gss5s6/w8/zmE4nFWiNRT1+j4n5a+bTtcYyJBi8VioKVuEmNFODroB5Oipawh9fPZE9bAnJ04mJszte9f6ldvM3gvQB3n5OvE/ExDOY5uctYpILTOn/X38nl+73u/LkSYzmHb4cqEKjGys0BmXdhRpcTPSDFMacNxFs4zdZH8rHO5ZdpqpVKJVF7O0pOKiLlvspq2zvXyuJIfiBMSG4R0RDK6pvAsFk/J8wNxpvENcq3tCLZ/fY35lJqpnvSQFtzzM2kZG2VfnIlScMnfBQFAwGnc0wEaf7a0NN+zatZ0rR6GYy48n8949g2Dc8ISSIWwv8fBLBew/nMGv6cjBmVv6z67zWBPF4tFMAHFCtKypJP/DwaD8OURyIRloDvONBW66wQo+8fecesflUtYo4SMSDzC935oPHqhs2sRPzD/fXpQ/jvPF00l1jrAxLUUxAsy6iYSDORxYQSEpI1SKh0eWnCCZj18MQ8ACmdskM/HQKDUJUjn4YKRWK3n20qraGlWH/ahtWIhpKEL/zc0gKUFUfr5rwOcUrfA9wBN5H6yuzuu9bKsldCQI8P5fD4AHt6PmY8Qllbb3qZz8uFM64zqlpAzLiY9NdJeW+v+/6ac6UcZ1iUtgAjV/xcXF7q8vAxmZENxwAuFgur1erTndLMLeJ9FutQnAYIUQUfN8H1IJPeeP+Q+4yNsO9zPwCTkcij3qwAouB8XYSTdS9BarRYmox8gg7WhTVmvx5GLxWL8bDQaaTAYxDNBOD00kJVhmRPP7/f7+vz5s/L5fLSZJQUUAmc9mJPNZjMuSUtdFBcInpWGJsZ9Abl1i4tOGqRuYrEBXmY5U9bmV2e8e/dOzWYzmIkYK0XqNKYjoYU+T24BwljOG8RVPQvOaREwj0gLlmAul9NgMAjhwl648EjHRoaF2O7u7vu49vt9dTod/elPf1Kv19PV1VWYLxAuAAl9h7kSnufQaMwTxjlYFnlxcRH/B9Lnb0FQKWjm6gfCPe5XZxlIe4hmMBiE3y3dS+xWqxWEulgswpwhZ9qzf1JAyokVPxtzk/8jsWlO1ul0Iu90f38/rhBBoK3zFR8b7rN2Oh19/vxZklZ66t7e3kYuLUCItOzIcXR0pFqtFu4CSGp6pjAbUQTO7OrqKgQWKDyMz174ZcieAbbt+Id/+AcdHh5qZ2dH5+fn+r//9//q8+fParVaAbrt7+/r8PBQx8fHkU/MGYDcw5Ssx8OIrJ28cwAszpQb8QaDQXQrubi40GAw0N7enq6vr/X9999rOp3qm2++icyqTVelPsqwbsZAJKCIHgJBunp2iKfbuT+2DnFz4MbNYg/QYy7CAGl4xXM0s/h4fBaNTzAb8IU5eEolkjgVPh4oT/19/HJfG0hxqVSKWw/WAXbsgftFX+PbuZ+Fht7Z2VkJn/A7GASfLm1c4Gmm6ZnO5/MV/999coiVAQOl5nCKhWQRxGhw5ocgpugk7c+EVYgJTH8upwdfo7sQ7Alr9oQhn7P7v96Xy3nIw4jrxqOgEwQHYfrhsEBP6+LCo1arFR362QB65PjnOSzSs9JLh6jgIbBPOdo6QEdaJfIsg8Ol/I+sI5IbiPf63bGgjZ5E4Lej+cFKCr8F8859YTehOXgQTtLp0oRx1pt1OC6QEhh1pGjYZrMZ6+X3zAlU0zsyuBBCaLNPxKYB97gMjCSO1If3M826VsxqBCtflIhi9h4cHES/JuYvaaVLCr2c0hz4NCuP9bofy98dHR2FQCSzCZrH8lgsFuEePbiubQ7WCZCfkVaGD8CGcHkt2Sy53H3WEAeMCZnWjbqmQuo5goaPXK/X404UJH0aMmHuWQZzKJfL0eAN4m00GnG1BSYP9aD8LYToWmidFeEM68TrhRGFQiGImf39Gl9101o9no4gYR9rtZra7baq1Wp85/Z11ywk07gr4Ag/9EFmFERKKSVaFq3i4TgASgfnspwpjFcoFKIUs9/v6/j4OOiXK0KoX4Uu01CN0zLKILWo0lRa9pLQDXkElBhiRtMTmQ6Wj4UkNzIsxMYG8G+fPBKIg2DhlUolCAIp5HnF60xt3slhO+jD7ymgxkxyk8bR1SwDZneTWlIwLqmDbg56tgtCIhVqHqvlPe4SeLYM9Zi8B6uDon/Pp/XnfI1gYt+lpV9YrVYjoQBLAkJOCRaTkrm4S+DP5n38zgsI3AJCAPh1Gel8s54rc8KfRKuS7MG+wmSe5O9rcB7w80xDcx5FkJZ5+Fiibn2ydpjTgbfHXLlHTWKPsXa7XZ2fn+vVq1dBTLVaLUrhSCfk3yzSpa1rHyd2Dg0fA2YFUcZ3I+NHUiDEnU7ni6sdskpjJyrp/sCPj49VKBSiNAyrgmJ1BAQH42v1g07XiOXA7913hInxKSFUukSMRqOVsriswol9oZyu1+up2Wyq1WqFAKbNJ5rWNSjpeU7oaccJ1uxhKYa7GZ43TS8nr8cF/AOUynKm1WpVksJXLJfLajabev36dVg1mLxcsObWnWvJtJSQ9XkYBjfPIx9uQT1//jz6W6FhiYLQhpV1fnVYxzOSBoOBPn78qPl8rt/85jcql8uBsNVqtQhpuM/ivo8Ti+cToyVghlKppFqtFgitX1k5GAyCeXd2duJAz8/PAyFOmWSbwXwwnRqNhtrttp4/fx7rcR/NtQZgWwoypYCdtLzpnLxgT1bAb5/NZjo4OAhCo58v/ZD7/f4KSpyVYT322uv11Ol0dHh4qG+++UaVSkVHR0dqtVohdL1yhXe6ZYCl4wKYtRMGw3rhs5VKJTQJIQ7Wi0b8+PFj+HZZ8QhJUaIJLvL73/9e7XZbzWYz6MzpFSHMeXB2vjbOkLg4tIOiAuXmOSinQqGg4+PjSJzgpkNudgQVfv/+fdwC+dB4NDWRQ/CY4WQyieyRNLyQIqHpYqUvb5rzw09R5NR0dtTZnyctm3V54sM2w9PsCG67eZ2u5aH0Q9bomsB/7z9HKzM8Fowf5D6RWwEOwmQNYXkiiCeiw1AQGF+sH1eAtbjp7+v0tXrpHJ9zVwIBy+cIl2GN4VLx7ixnSp5y6nKtS/xZBxD6fvk619EAc/MEELe83MJgnR7HB3TEgtnU3ihTLjGhFOx/L+x1p9v9m9Rv49BSQk6ZFZMKX8DNEZA/LypOM6myaFg3wd08TYGwdetxs97X48KMg15HAPweovWDBjUEJEn3iedlGY6ie0KDN5Fz5Ni/fP6saZ2WTwWzfxaGhGnSEjbO0s/la84UzUkYaR2S7SCfA4QMhHjqx3I2LmSZrz/PLSs/L+cl6NdDkZssiq18WJfoxAxrtVrY3umdN0hIl2o+Ummcamkneux5fAFp6Qd5kgKE9zU5th7/I5OKShXXsMzL98OJiOegGdbtYboPSGuY1d0D6d4kKxaL6vV6K8Ix1eRfM8h4ur29jYC93yPj7/DvLshZX7qf6bo5T4Svm43Sajqem89fE6KTFOZ4ei8OtIkZ7GEoZzJfyzr69bP29bFvqcBOaZy1+Vluk9GVyQFC67HQFGBZ50+5JpLWl56ln/e/c7NhXXhJUqQloh19o7YZKUrsBek+X0+I2DTvbcY6RkifnWpyh/tden/NSHNePX68zp1ZN/dtf+7PSv+9zsKStCKAsXiyrJUQDRYfwy2HTWv0NaUWkX8+ZXDnhfTv0t+5aS0tCyA2zWejhqXCYGdnJ+KQ3333nZ4/f77SRcJNBwAR8nkf8nNTCYQ5y7MwNcimSqUcZnGtVtPh4aFms5lGo5G63W4c2Lbj22+/1c7Ojp49e6bT09OIt5IuSAw4XQ+SmbWl8Tf/uR+KYwPp+qVlX15g//n8vqv8q1evVrKyBoPBxsqOh85Uuk+1fP36tV6+fKnnz59HrBc/HhfDrYzUF3RfLV2ra0hp9b4a0GFJX3SA5HfELLmYjBhuljNttVqRp4vLQ2RhPp+HFeWC2AEl5ppqTD7jvrh/3k1lnuu59qRcekiHkFq9XteLFy8eXFcuqw/0NJ7G0/jHG/+/S595Gk/jafz/fTwx7NN4Gr+g8cSwT+Np/ILGE8M+jafxCxpPDPs0nsYvaDwx7NN4Gr+g8cSwT+Np/ILGE8M+jafxCxobM53+xb/4FwsvLKZbAg3SqJ8kVZGWIs+ePdPR0VF0nnv9+nVcHUmvHDJGyDflHhWau11cXOjz5896//591OFS95rL5aJeU1I0wNrd3VWr1VK73VatVtPf//3fb5Un+O///b9ffPPNN2o0GlF8L903JiNVkY56ZP7QDZK0N7opkmxOBgytUqTlPbc0k6ZUjpsKcrlcFFlTq8nfcXVIPn/famUwGKjT6ej8/Fz/7t/9u63zMJvN5sILJTwjh+wjMnsKhYIajYbq9bpevXqlv/mbv9HLly/14sULvX79eqV5gaQ4F/KxO52OLi8vdXZ2pqurK71580Zv377Vx48f1e1249y8YyQtVth38m3pZvKXv/xlq7X+k3/yTxaU9vGMtGG756rTvuj4+FjPnj1TvV7Xs2fP9Ktf/UqVSkWNRiNqsXO5ZXsfsvqur691dXWljx8/6sOHD3r37p3evXunbrcbN62TEeW3s3PfjrflKZfL+k//6T+tXedGhq1Wq6rValFoy2ZCQCSqk6/ZaDR0eHio09NTtVqtaBtDnSC9bMhbTZuoQTi0Z2k2m5HqRYok+bQcoKTogLezsxNFCdScbjPy+XzM1etfISTafEDIvrleTZT26HUGIG/XO1rwHppms0a6Ffp+e/mXpwtmHV4q5yl463KmeQddN7y/EU3EYTSK2b01qacaSopzbTQaUarpXShInvfqF56PkN92IEDJEaekDWGQFopXKhW1Wi2dnJzo6OgoBDJphLnc/eVV0IinmPqVHuwV9xpDB14DzB08nAeVPxTTfHWbU3JradnIJtJNAiKHIFutlp4/f67T09N4MRpF0sr9lzAJBO6XRu/u7qrZbMb18/RUGgwGkYN7c3Ojer0ehIMFcHBwoMPDw8wVHt6QDMZ0pievVlIIBtrgSIr+tOSZInmlZTG1a1jW6XmszJnCf4QGA4GAxqc7Q5aBhVMsFoMYyVX2liaS4uf06aIzxe7ubrSrhQ4mk0kwGzXTtMalkwIWF3Og97AXXOzs7AQt+Jyprtl21Ot1nZychBaTlrTr2hVa4qI3OkNUKpU4L/KG+Sxzh3bJUWYP2u12WJxoZ2geBVCpVGK9tANqNpsqlUoba5w3Miyak26FHOjBwUGYGUgDft5oNFb62XoHgrTukeEFxRSOU1hQLBbDnIA55vN5dLigSRoWAK1cspTZIb1zuWUrUzRNOpyZXMtR/sc60/aWXnLHeyjS9lrZtCTLh1f2eDeQLIO1Ye1AjKVSKQo4KCSXFFJ/d3c3Ogu6tpdW28LwDh+Yj5VKJejGuy74lRzsu9eHuuWz7XBrBNM4l8uF1vUChb29veg5jeVE10wv9PCyQgRzeqZpNQ+F9H6RFwKTiiSa1LHPm6qSNjLs0dGRms2mcrmcms1mSEmuRmRD8NMODg6iise7+Hvxb1rsjsniVTmYbUjUdrsdEsl7H0n3pvTx8bHG47Hy+bxarVZ0qdt2+Ea7xoNAXMhAsGgXb6DlFT1OMN6Jzy8Y5jmpIHPhxv5KCmlMpQnNxbKMRqOhQqGwcsFXsVhUs9mMC6GY797enl68eBH3vqD9aIvCcAbg/4x0/X7tKGYjrWG8kQC3TKAsaA637aDVDbXauDWY49I9PcIstMUBl+BM0aRe5knZHtVNqcDxFkms07shouTm87mePXsWPML6Nt3St5Fh//qv/1ovXrwIvwXN4sTtUsd9BMyhm5ubACYwE2Bq72lErxsmT5c5NhMfCSKQlkXPlME5k9DRfpvhxcTexQLC8k54+DIwkjMO86XL4rrWMkhcLyfk+fhz6yQ6jMXALOaC7W3H3/zN36x0uYR4EK7FYnGlBzPrAFj0i6GwhiBQyisRulg6Lpj8zhtafXLbANdU8Hn8YMzSLOOf/tN/qtevXwcjsncwDf/3EkFfg7eZRehg1bmS4pm4QLhS1WpVBwcHIRzH4/FKJw1vXAc/wfRf3dNJWm2ryXDpIS1v6oKwKCYHPfPPpsXuPJeDWndvDOZG2mkBxnJwxw9h2+HgEr6qM866YmQYib9j3qlZlM6Ff2MtYJb581MACMnN+7xL/GAw2Hqd0rLBGAIWPw4AEUDQmwW4IJG0Yjm5Oexnlpqdfss770bIOgDp9IFm3rYFqA/Xiul5wIAMxztQOtSo4sf7cEsLQc86vaukKxDvyYWQdp5gcLYPjUdR4mq1uuJrITm9eD1tV8mlP0gpzCdnACdOfsei0cwUGKPhkc5svhORH1TqRzw2/G4f5oYZ5l0YEA5IVIquMXP94J0BMRE9XOGXPLl14IwiLbv2QRz8He0/N13rsG5wkRXvYj+xoHg/v/PrVljjupYpzBE3x+8KYt34zQh3rKf9/f24+sWReGiK2x6yCCe3IFz4emtW6A2Qi39z+zqmqfvmKYqeMjtdPinCB12n7aoLNs6U7wiITWMjwzabTdXr9bDppWUjbxaMlnAiGg6H8e/FYqFGo7Fy2L4Bvuj5fK7xeKzPnz+HSQxQUS6XA8hyyQRB+Kb5O7YZWAJIUt4ByOKHzmF6PJoYKghfyrjedQETejgc6u7uLgQRa/O+zek6+Hvau7LHWcaLFy90cnISDAFRttvt8PdA89HAnM1wOIxuFPhb7s8TosIicBT2+vo69hV62t/f1/Hx8Uo3Tte0aFWQ9SxnShwbIY8mRAHhzsEkfqvgaDSKvtCO+rtlyRkB2kG7l5eXcTkz0QEEUxoNkJYCGd5welk3NjKsx9fcqXaTjeGMAyMTQ/UYWgpOuCmK5GUjODwW4D6hI6TpIv1vthkQZ9pPx5/v6/W5wbiY6+lzfX6s1dHFVMC4heDvdC3te501Fkucmqs6eabfHePIN3NgrZ50kKLCEKPP0QUYloVbSPReAmF3dNhdj6xrTfte8zz/ma/D29kgQBxfcK3InvHdu0/y9w5GwrTuq6bPgDYeUzZbhXWwt910kxQZN8PhUGdnZ/rw4cNKJg9SFPQRcMYPALPy4uJCnz59iqsmQY7H47FGo5Fms5kajcbKlQpoPveNssZfWQd/T/gANyD1T7hxjEPCBNrd3Q2TmubVbhIixXu9nrrd7orGgZik5U1uqbkPPsDBYtFkjcO+ePFCp6enyufz8Yxc7j7cAcNhSWB+o4l6vV4kwHjDOjSxa53b21v1+/24rYD1o1E8gcMb+jkwA+NwxlkGigIQyN/DnmMtdbvduMbz+vpavV4v3DGiDdAudMHc7+7u1O12dXl5qW63G64AcflyuRwWSdqaV1JYG1g1jgOsG4/2JWbBTrye+XFxcaE3b97o+++/j87lSGy083w+D/OgXq8HNI7vcn19rQ8fPujPf/6zLi8vVSqVVghGujfZhsNh3DhweHgYjeEIhiO9s2hX6f6i5JOTE52ensblvh7SgYC52xSGY47s03Q6Vb1eD9PW/VDut726utKnT5/izlQym8bjsarVahC0ax2eQ2gA0AcGyjJ+85vfBMOCE6Q3JkCAXNoNsfJuQhDEDf1CLG9Fy98Ph8MISXFO3HRwdHQU5mOtVgtUms/xNZ1O9enTp63XiSDj1kRvZwrIdXNzo7OzM3369CluVfAUwvl8ruvra9VqNeVyuQht8Uwa/11dXendu3fqdDohzBDchUJB7XZbd3d3kQlIggQIPWEs9/UfGhsZFonuDODmBF83Nzfq9/tBxNLSfEWKeV6so2w8D0nDrXSppoRhXApLq36wAwJZNC3+ipsl68xSB5kgdLcYPDbtKK+b79I9MXHztpufDF9fOgf/LDHMLMN9Jf/iPPwmgMlksnJBFRqPPXOk1c1N5ohQHw6HK5lBHqpyU5LfY9G4S4TZuu3wtEgsAEf7md90Og3wjnPlvTAufrmbxb530ORwOFzp0ugm9mg0WnkO8/B1Oqj50HhUw3LAjmg57M1V8Kh0j5G6WeI+nh8CCyak4Oixh0k82O8J3Mxxnb+17eCZpJehsd30B5Dwm9OdoFirm4aOrKcxWQfhXFAgZZG0fgaY5U4MWRMnvOeyM6u7PX72uAvOXL4uR4hdUHk4yMMy7lZhaiMMYNRcLhfzdKbLIoT9nc5sklbe7V+sz7EYF0QpTuPIc+pfp6FGhAFmMsBXeoOdC8V1YyPDptzvwBGoWKfTiWoEUsAkreTZ+oKc8DhkN5HRtGhdEv4xT3Z2dgJyZ7FI7xQI2nbgb5Cby8Hhh2LO4q/7BsNUzqBuxkrLazgQPnzGzUySEvAR/XN+kI/F6bZZqwNWzJN3olnccnBNgvBOiZnPuQ8OaAkDumDjPePxOM4Z4YGW8SQNhOa2wwVgqhWlpcWGGeyM6BeTpbFc6AwFxnoA7TCDvagBC5QIC/4s83QlAFM/NDbWw65zflMJzKYC17tTLy2dfzYgNacZbBi+n7+HjcSnc82azi2rJJa0QkA8ByHl5hu/T/1kCBcNtY6h3OxEQKFpPVjv0L4Tnb/HzaZN/s664Qzmz3Sz090LB0hcU/laHSBzc0/6Mvc6fZ4LLvYgVRBOO9sO4sFOC460IwA88sFnfJ3+HF9jSrt8FsswBRzxm9MrZXydKIGvBp24yh5GdBMFwMLTsZg4h87PubbQLybytEY2oVwuq91uh5Th9/l8PgLQHp9D87BB0+lUw+FQw+EwwKptB76ptEzIcJgec9nzUCFYiNvvE/KDcFScfFY+g4QlT5f8XogbAejxWfylrGuUlkkiLlQdSGSwB3weYprP5yuMnPp1zJHnU9k0m83iTNHe+HvMwV0o6EeShsNh5sQJ8gEAtHADeLYLCWkpPBz/IPGBOXk1F+cKU1Iy54k0vI+qNneFUvN6Op1G7TC3V6wbGxn27du3Yea6j+LB+1wuF3WDFxcXcaC5XC4uy8VkoDjXL9CS7qsx2u22ZrP7RG/ygNlk0EcIBtMhrbeEkM/Pz3VxcbH14V5dXalWq+n4+Ph+U/6/phySHUCCumD3xUBNXfhAwB6qoBrj2bNn2t3dVafT0efPn1cyrNyn9BI0mLVer4dZ6L5uloFp72ZrCtqVy+W457Tf78c5LBaL0CLUsqZ+MDjEzc2Nms1mYBu4Ep4cgdlLUgy+HQQMfQyHQ338+DFTfvibN29CeHimnbQ0Q2FCaYkqo1HTqi0wFs8T5/P1el1HR0fK5XLqdDorSDgJHKmQTGllOp3q4uJC33///Uba3WgSE/9MtYVLKJjHpQqbwsTpEsEmwPz+b88lJpbFJmN6c/mvB7ZTFBXi21TxkI7BYLAS0Oe9rNcBHhBTH27aunnlgIwzMXN0SYtfjD+TBtEduGLgW2cZpEMi5Na5F77nmGhuugKgpG5BivzDfMyTkAnvcwTa9zklbOgoS1ZXv99fASb9ee5+ODLMAAMBiJRWk0NSUBE3jr/x9RBFobMEwsJBMObF+WyyJB4N6zjjeFwRzZailqm/k8vlwixxreqmI1KQL2pvIQpHptFqmOMOMHm2ShY/Fonq4AdE689z38PRXsxWTB/303ytHC6SlyJpX4ejhggn5pSGTXw+247JZBImOATnc3SAy7WuMzmmIl0xHJThWcViMbQm1oKnlPI3AH0p47IHMHlWIDEVAszLQUkstJRuPDzHOblpnNIuliRf7CHnxLn6d3jLz9wtxYfGRoZ1TZqGWBy99PCGHz4ToPLewzagsJLCz63VauE3wwwgZmgkJDzS+fr6esWMYmxy3NPhPlh6wMwDwmG97ANzw3SHEaQv7xbFkiiVSqpWq/EsJDlSGontmWZercT+pQDSNoNMtHVWAAQEGu535LJP+NOVSiXaqDAvB5wwI+v1emSHUTKH4JUUvildGWAemBhG/hqG5Rn8ndMEKDGa3y0s9hb/G1fO86rZO6IbtVpNzWYzzrnb7a6E/3K5nEajkSqVSghNGNXDlI+NjQyb9pdB26Vwf4oapmVvzqg8Z53kg+jJL3UCxSR0FDMdDlJlyTulPtTXAiTvSJ9reyce/u2a1TUOw5nEGTBlCket3bRyIMOFRJbhhQzpXD0s5meNqe9Ir1tL0tIC4XmeMEItKQNBwFrclEzPln14DD1NBwzmLhN0loI+7qLwDqfdlJZS+uezHt4hxMPzU3fP5+FuEc95aGxkWJBRpBTpZR4DReKSToh2xGSq1Wpf3NDOv33DkMi1Wi1AHpA+Dx2wGWhnr+Qfj8fRcbHf7297trGZ3q4Ev8ZjZJKCkTFryGv2W8sfexcEz76RpggOwHeKLtAwnAfVPqkfuc04Pz+PcwE4Qzi6RYWbUKvV4j3sPS1UPEzFl4MpEDD3DKM5Ae7SVEY3TYlLk/56dnaWCSVmHfTP8iQKd+lQEOwvwBopl34PsmMwvu/sIVYTZ0o+fS637F6S4hbs+WQy0dXVlS4vLzfGYTcy7A8//BDmLOVX8/lcg8Eg2lSCErtUyOfv24AeHx9HCAMTw5P3YxLFYpTxYU54OR1MRJYI5X0wFgJlNBrp4uJCP/30kz58+LD14XY6HdXr9UBQaUFZKBQ0Go0CwEDYwMD4o41GQ7VaLX4mLc3hlIGB90ksv7u77xzR6XTU7/e1WCw0Ho9VqVRib9H+nlnmWinL+B//43/o/fv3evXqlX71q1/F/lMayF4z1+Pj46hPLhbvW8nQAXCxWBZLgJ56iIsz3d3djQSJfD4fgpHoQGpBYHLDWFdXV9Hudtvx8ePHqEMdjUZRobRYLCKlM5/PR56wx4DL5bIODg4iL5zQDKFFMAX2jsuwi8WixuNx+PYkhkjLkJ5HOfb391fMcOjgq8M6SEL3c/Bd0YBISg9EIzUcXWXSDji437jOH3PTCsYk3otEllZTKNGAWWKUpFbyDDR5ChjgszvS6QH91FRKzWLfQz7nUhugDT/XkWLMMnweR+CzjKurqxW/ElNzNlvmSbtv5eCX77Obeykazs89BdFNZrQnFkOhUPhCALH3AEBZUeLhcBhFC36mWCh+5g4EscfMgXCPr+uhdTLYD5QS6yTE5xlcklZo15l83XgUJWY4XM/vMBfL5fJKRgcDkAUp7IFnR4v9wJHONMRCQw+Hw5XEAc+McRQRvykLSsxh8UwO2KF9iNsBiXUZXetyiVmnr5e9wXUol8tf5PlymJ6cwM9SP2zb4fFkzEUXRp6IIn1Zr+vrdFPR/eL0PHFhWCOAI+E+admjCqGM2+VdGLJYE5777fPHUqNsE0GJKewYiPdzcszBecHPmUgBTe6vr69Xzg+actcLdwyB+BjtPooS82AnZAgHqccBO6CBTyQpzGAHV3wRKRjjHS0chebvAUJc8vucsyYUYGp6yRzvBZUeDofxfwiU9ULUKbMyUoDH43fz+TyEE1aExwlZT5pvnT5324EWpeEday6Xy18AWhBTGseUVq0lH6l1galIySLEv7u7q9FoFARKMg4ugIeX0PpZhBN/5z4se8m6YVi0Ha6HKxNvm+PWQgpoOarsyRgIKKddz5yjlts1/SZc4tFcYh6eajE/NGmZT8nPib1xYH7Ym4jMCdr/noN2s8zRQ979NYiimyIOfrhkZo1IQf7OUUQXRpvW6YwG47p/jyR2wAntgsT3PcoyPMU0fTZzc63Jetyq8PemCRgP7a9rIYQdVhd/j7vl6Z7sBSZjlnUiCBxsSs8GgcG78/l8xFxRIMzP/y4dacTE0xk539QywpriHSiJTet8VMOCSF5fX4f/6P1w3Y9DK3LYqHoQ0ZSJPFjNgRUK972UICz+1vvLkt9M/u1wONRisYjOFxcXF5kqOyAahI6/B2HlwW6XtGx0lvd5CEta3qAAsIFgJIeW9FBvcgcQlFXDOvrulpMzscfeAUskhQ8GMENut7RaJ52uEyaQFNaCm9Ygx7gHMDPPpnNFFn+d/aKFqifbpOEbZxB+TvXZwcHBWvR/3ToR4ERJ0haxmMNu7qNph8OhPn36pMvLy6+vhz0/P9fR0VEQDkhit9tVp9MJk8JNA0zf6fS+40A+n9fr169XIPyHEFTp3nym8TJ+DgztKZDF4n3/3MlkErmXtF/p9XqZQwCSvkA7Kb8aj8craJ5rYDZ/sVjEFQ0IMQeYXPsi0SuVShwO6ZG0VuGZaCFAmH6/H9A/EjrLOD8/jyslAHzy+fxKuZn35c3lcoGwYqZOJpMQoh6ycy3lzEpZGZoGIsWXGw6H8bcIwdPT00ioKRaLMb9tR7/fV7FYXAGfPOmGDDI/K0kRBSCUdXp6+gXtujXBOlFUhHYKhUKY47yDMx0Oh1EokM/n4xI479H80NjIsByeE6k7zmgfh7ldQyE98TNTZ53v7p/yDD90GANfRtKKbwsDsJE3NzeZcmy9T5PnTnu2jef6crAg2NTsenpZyqjrGJfvDqqwTjSR7x3rhsm/JjXx5uZGpVIpNA7xVd9niJtQDr62VxJ5vrebi+kZM/yzjvzjQ/Ldh8cqszZNd2DHM53cfyTu6rSNtYEQY52pb55+pQAgNOlxdc9sS/Eg3vEYGv4oSuzADmqfBcK4joB59oajgm72pgebaon0/+tCQWwKYSU365yxthl+iB4GQOMBSnk4Aysil8vFNYTu8/paWUMqmX2v/HfuC/F3mK78jftlWQZ/42vz1D9nZN9nT4THNPYEhDSkxRp9L3xt7iOn1gvn52iuJ69su05oIUX+Hf132uEMcMV8fc6MacSE4S6e0ztA3rq5pWfxGGD6aIsYHgqMnxIiGz6bzcIv4mfSvX/oyQ0pcfIeN4fWHTrPkpYgBmZGuVxeAakeg8bTgeDxO16kZUjDY5JIXUnxPtaFRPfE89TndSbl/ykognDCz3MrItXQXzM4V+9jhDZPLQUIinViMrp/71lE7g/6mSLoXfD6v3kG/qwLDw89ZVmjg6Tus7qwTLUgZ+OmcdoSKHXnfJ3+bo+wOKKOdYHfzmfTENq6sRFK9YuJnYEcMEhRSibOzx0mz7LhbhZLq4g0KKMjj7yLkRVRlLSWAF3qM3/Xlm66e1H3Q8M1UWpGsRYE0zot6witI4zbDpiEZznxuubzz0F4HrLzFE0f63xqaCAFb3x9rhB8Ppix0F2WdUrLe29cy/oafDgY6LSVIszpSK0LlIbTCO8CLJW0wpys0+e+bmyFEvd6PZ2fn4fU6/V60VeJuCRgER3ayZdtNpuBEEL4SFsHKpDsXsUwHA4D9EHaItld0xDymU6n0RE/SxwWoUSP2qurq/CD0UIANJRSgdCCaNbr9QBgUvjeGcNNPs8oAitwpgc93NnZiXWj3SGgLOEr9gsAh/TSy8vLWDtAibQsFEBQQcjpZdapkHOh5WamC8FcLhdXOvIOcAPWjUkMQpzVqgDp7/f76na7mk7vuzrwbIrYQXHd5N/d3VW73Y62vC688cFdi3vVDUkgnk21WCzi9377He9F4z5m9m9k2E6no3K5rFarpe+//16SgnAoS8rlclEzSIM06V5KNBoNHR8fh8TApHS/zsEdSut6vZ4mk4m63a7Ozs7Cj+n1eur1elos7nvnck/MdDqNLgk8JwtAgUk4GAz04cOHWM/FxUUgzjc3N5Ez7HWg3nHAY5MghW6ep0IHYAShABFcX19Hf+fhcKjLy8tIuCcxHCGVdUCoi8VCb9680cHBQczZEVXql5k3RFytVqOvLud6e3sbVheE6JpmNptFMwRMz1KppMViocvLyxAIECtM5uAhzLHtYE6j0Ujn5+d69+6d9vb29OHDBw0Gg2BOkOjd3d3Y20KhoGazqcPDw5WKNZ7pqYiOPnu4jR7WnDGhKc4XoBNh5eV9m9y5jSeO74LmQEo6XM2h8RLvjO4lWL5QDtOD/6QtOijhJjeSDF/KtZX7P8Qve73e1oebljW55PO4nZvoXqvqMWY381I0fD6fh0/vpi0hEtee6xBWN7WwCrKCTmgytB7P5Uy9mB7t5zct+OVYnJuDh+6bOqbgP8Nf5udoWujItTTaCabIsk43ozk/tNg6ExuLwhNi3AX0s2Q9WEW+TuiCvWEuWIGOW3C+LrQ2xZsfvaqDG+wovCWJma4Je3t7ajabajQaK+AP2U7cM0rVfpotA2F6tg9ZUnd3d2q322F27u/vh7YjP9nBJ0d6s5jEHCib7LebEe9Fk9I9w7OvqAphT/h5SrgwLMgkJqX7N3yOliIQD3gCX54ql2WkIQoHuDwPnHacWE5+B46XBnqDPU9XdNCMvWKupO/xPNJCXYD4HN29yDJYE3SXWgC4bLgzrii8WwZldmnqKdqRdVBlBFMeHh7GfjUajbAc2Tv+3vPYU0A2HRsZloRtukXM5/NoppbL5cJvPTw8DKaGYJFM+Xw+noOv4Itm4WwuB5nP35foFQr3zccwiweDQZguHCaBdz/QdeDHQ8OJly9JK/5No9FQq9UK4YMAcrCGpAuvFfW1OvHCvPjgfovd7u7uSp0vjInF46BIVh/W1yotQRkXCpTGIawhXNaAdvZmel4MkYanmDOaE2F+e3sbuIdfTyot45hpQUXWdSIsmRM0BRMdHByEQOb8pCU4l8vl4loS15opWMYZYolCD9AuSS/cM8Q+eXfQbQTSoxoWZxlzEwaaz+8vuoVwkSQkeLvpljLpuhgWwzU0TOi2Pf4UJgXmJZ9DK2XpxIDp46ABVRdkxLBGCNu18Lp5P7ZOFwqY9FgSEA2MDbOijd10z4Kc8l7mhAsiLSMClPZBpHT987xfCMsFUYqMpmGtNBvKi0PYX6+Wcq1KlCLLWj03AA22t7enWq2mfD4fRfysj3V7q9p0z/xs05i6nyl04GEaD4+mfZ18rY+t81HU4vr6Wufn53Hxbq1W08HBQWgGN51ADj1zBjTQTUhHeN1ngSCQ4jAnZgnv8LguubH4rldXV5nvEmWDvMNGvV7X6empptNpWAhYG6TZ+ToxXd339lCYhyq8cZm0DHe5pvKMIxBskHkv98vKsNIyEWI4HKrT6ajZbOrVq1dxnlxK5Y3WsH7YV9bqJZMIKCdm1uuMyM9xe/CT+ZlntQEiZkma4PngLQ4achFYuVxWvV4Phk1L6Jx5HK9wV8TP1ENWbg1hkd7e3qpUKgW9Aj6SINLv91dayj40Hi1gPz8/197eng4PD/WrX/1KL1680IsXL+JwkbxIKZeiKRjhWtcZGjMszToBPWMTWODt7W10XRgOh/r+++9j0WTFZAFjHMDCRKlUKjo4OFhhVtewaQJDiny79PV1uqZij3A1+PzBwUHMhTYjg8Egcogdzcxi+ksK4G4ymejs7CzW9fz5c+3u7urg4CCEM9oGYmWdri1Zq/uufLlPzn6gZdA+nkzTaDQCaf3w4UMwrjPItoPznEwm6vV6ms/nqtVqOjk5UblcDszFLaXUqksBMz9HF0hgOT7H2WymarUaP0Op3NzchC9Lwn+3242w0/+TD+tF29671RMI3BT0Bfrwg1xH4HzGY7NuBns80/0+/o4vCIeD2nZ4lgnxPgAh94N4j5v26XpcGPn/fU0ItBRZ9iwbUEW3OtyP89hgluEIrCPsoLY83//tZjjzTdflPjXr9D30v3UNlq43PX/mBbC47XBzFNr19/p81yH6DDeDfQ3ska/NM6BYj2d68Znd3d2wHt3iZH5fnZrIH3tCw2KxWDEPUlMh9Wd8YUwo/RkLZKMhjlRL4ee5/8fCpeXNbFkJ2U1zT6j3/lOpOe9hHp9nuq70566RnIhwA1w4cQYecE8zkb4GJfb3eOqea1Ffhwsqf5+v0xnN587fOGHDoP6uNGoApuDN6bKsNT1PwnPOJCnDuuBPgbNUmLFmX0P6ebeu3HXwHGL2xzt9fLWGdR+KRUM4Dry4v8ZYJ3nSTU+1k2+CMzfP4OA9S4RqITY+7UCxzUiJEEBg3TqdoFJm9f+7YFoHULDG1CLBenCUHX+WLKu0pO1rBwxL3C8VuMyRL0eB1wlENwlTMzZ9HgIq1aj83d3dXSQ4fA3DuhD2bDKYMrUKU6shFWyunXm+z5eRWpOs0XEKt2rQwlg8j61xI8NiiuAIswEu4TcRjfus6eI2/Y0nWzjSuI6QmBdCxc2fbYebK2wc0LwjhOtMJp/3Qz/btFZfj/tIqfZyUxgf0OeeZbgA8bTDdfN97N/pz1KC9s+4VZH6valmSSuI+Py2g3c4HoKZ+hjtpkyzaV/WCeD0836W4Bb8H83KWnnOQ2Mjw/72t7+NS6JevHihVqsVDddIMQSpdGJeJ434nU/YCdQlth8OJi6Lg0G9gqJcLuv29lYHBwd69epVmO3bjm+//TZaer548WKlJaabLTCuH7avwf0jB90gUl+nC7N0/Z71A46AhgXpfPXqlSRlWqek6E6/v7+vly9f6uXLl3r9+rXa7XbE2mEQB/7cumC+rNXPNz0711ROpK79/EwxW7lZ4Pj4WCcnJ2HZbTuePXumQqGgdrutb7/9VgcHB6pUKqHhHKvwDDs3+9OzS8/amdAVE8/g5w6egq94aWaz2dTp6ak+fvwYZ/PQyGU1H5/G03ga/3gjW5rM03gaT+MfdTwx7NN4Gr+g8cSwT+Np/ILGE8M+jafxCxpPDPs0nsYvaDwx7NN4Gr+g8cSwT+Np/ILGE8M+jafxCxpPDPs0nsYvaGxMTfzDH/6wkO7vZvUyp1wuF6lW1Ezm83m12221Wi09f/5cL1++VLlc1tHRkX7961+rVCrFlYOkd3GZFS02qWfN5XL6y1/+oh9++EFnZ2fq9Xq6uLhQt9uN91JgLWml/SddMFqtlv77f//vW2XG12q1BfnI3r4jTdqm3O309FSHh4f67rvv9Nvf/lZHR0c6Pj7W6elpdN7w4oVyuRzFE4PBQL1eT5eXlzo/P9dPP/2kn376SR8/ftTV1VXUSpKmRwHCYrGIDnykXnL7+5///OetKwDa7faChH9vyEYhBQUArPXFixc6OTnRd999p2+//VbValXtdlsvX76MYgRqefP5fKyVGt5utxttZ9+9e6cffvhBb968ifXTWZBKLO7w8fttKAI/ODjQH//4x63W+nd/93cL7hZmL0nv9KZ7VJwdHx/r2bNnevnypQ4PD1UoFFStVtVqtaL7CEUmNCtgUFjAuXU6HZ2dnennn39Wr9fT+/fv9eOPP8aVpZwp3RS5w4j2sdVqVf/tv/23tevcyLB+2REXX1EADTF6XWqr1dKzZ8+CoL2HL5/36hcvHUubUHGBUr1ej/aY3lZjZ2cn8po9j9W7JGw7yCWlUodcUc+f9cP1NiqskYu7/H4eDtd7LlN5w7wpkK/VavF7GIaWPF4czXfvKJllwBQudCEi8qQp42OPa7VaNNqjIZlf30KhBHedetdFr9ahRcvh4WEwJdVWrJV/e4cKhFOW9ZZKJR0cHCiXy2k8HkepXblcXqEh1vvs2TOdnJzo6OhIjUZjpVke7U0p/STnWtJKzjC5xNBHo9HQfD5Xs9nU0dFRdJ6Yz+exhzx/Pp+v3N73IK1uWnSj0VCz2VS5XFav14uiYCQXG8xmvHjxQq9evdLr169Vq9WiSRkd8ai+8UuWaJXBZUe3t7fBBCSkw+Q8D61XKpU0n8+jD/FisVCtVot3Zzlcbv32Ui4ON5/Pr/QB5mb4ZrMZEhGhRtE2B+z9ouhXS09n6b474eHh4UrnDoQY+7a7uxuCBIb1tiZZBh0CaXoGsdXr9bAq6H4JDRweHurZs2fRbC+Xy+n6+nql5SllmFhhWE60Q6H75MHBwcr1mvTN8gvA6Kgh3TMEbWtKpVKmdZ6cnKw0YZ/NZmq1WpG4T9XZbDbT6empTk5O9OzZs+ieQiNyqn6kZXcSL73zq1y8eOHg4GClGR97Ap3c3t6qWCxG72Va1mwqmdx42khxOvohkZvN5kq9Jiq93W6HFKbLHMSbliOtKzqgt49L9lwut3K3ihfRc2uctKxw8aZw2w60J1oFgVCpVKI7vNfgtlqtaLRNuxgvAJeWFRzSsmTQy60QdrTFhAm8cNuZ37URGhGhl2Xwd/QCZt9qtdpKtz/Oo9FoRItPGNTL4qTVLhZeHuj7iyWBZq7VamGR0LYWrUVdM3vImWYRwl6BAxMsFgs1m82owPIKMwRC2vXSq6uk1R7RacE5z/Omct7SiHY38AX04grosfPcyLAnJyeqVqshadB+lDsxUa4w/NWvfhWbC6MtFouQjJiY3roSBknL88rlsnK5XJgJzWZT7XY7GqzhJ2KmdzodTafTMCmylJ21Wq2oMfVa2FarFTcZQOjValXfffdddBT0fj3VanXl4NKDplzP6zu55gMGqFarKxcRuyl8d3enbrcb55DP51WpVLZep6S4oeDm5iaYdH9/X8+ePdP19XVYE9I9AWEtIRxhKs4Yy8mbl6GFELTeCcR9b5jJm6xhUo/HY3U6Hd3d3YUZXi6Xt15nu90OWmS+XL+BkMI64OLmcrkc++xXQnqJoLtfDNboDIyQxyLxBvwuKPH1p9Np+MVfrWH/9m//VkdHR9Gb1xmNB2MmFwqFYMzpdBr9gx0gARDicL1jHsxCVwDADbQRz/QrFtLrJCB06f7O123H7373u6iBxTRFSF1fX6tQKKhWq61YD8zr7u5Ol5eXYWlwSNReFgqFlUu2nGlh0kqloqOjI0mKJl3j8Tj21huWuz+ctc+RJP3zf/7P43pMB1MkhaWERvR+UsViMa4UoQZ5Z2cnaqTxMQGzIErMQNrRUtPbaDSCObnKU9LKFSDuKkFX246/+qu/0osXL0KweR2rtLy5gXV6If1gMAiGxT0oFAorjfBdeFHjiiuD0EPZ0GTe79nBapGWNwTiomyi3Uc7TqT9bxxMwf7HiWdD0m4BqPq0DYd3S2DDvLE1woFnY8p54TGtRqQleOJgx7bD+/B4wbH3CvZWNO5PeqO6tACaZyN0+Blax1uvSKu3A3oTANbPvksK5sky+DxzhoAQCG7WecF6qjk5w3VmMN+ZJyahm5qslXelrV4R9Fg7WW9zYE2YxpwV9EvLn7QHWUq/Nzc3K5rPi9yxKKBd9saf6Z/FivP/s37owwHadWMjw4IMOlqKZGbBgEjuS/plP+4DONH5AbMANo1uFhAz0rDRaKw0DUOyO7KMv+fAxWMDFM/nwxqZsyPm3jcZTcfcWav7euxf2lWDJuVuoiHx9/f3v7gdEGCIHreE27KMZrOpWq0WwAvXRgCcMSBAzsK1iKPc3n2Dc4W5ec58Po/zgOCho2q1uoKsY01gPrN/d3d36nQ6W68ztRJgrPSeIBiKOQAOcq7euTJdr7TU+t41A0EoLa+gqdVqK0ISReA4UD6fD1DqwXVtWvSzZ8+i1QZxLNpaAEAQP/P7Ubh17Pr6euUAYTSXUi6N8Xu5he7u7i78lr29PbXb7dgItCnSGO2MFs5ye92rV690enoaUg5B02w2V0wlfu9tUTGDWGcKQMAUMCdmkF9U7AAEvYEhGMxFBBbENR6P44CzjBcvXsQ+OpMcHBzEHDCz/dIzQmsQpIc2UoAJTYXAu7u7v6KUfUPw7e/vB2rLmbq2Auz0vdp2OLjGuUn3CL8zK+vlmhDoz8FN/t79VncL3ZLo9XohLGgzVCwW1Ww2Q2ATPXAXE2uL83hobGRY/E5sdp/sOsc4laouofzzKeLmZgiD/7u0ciCJZyM0YA6en6VhF6GZYrEY1z4SOkGwoJF8PswDQZFqKLcaWG8an+Sg/AvtDpH5Z3inC8AsgxCJpBA2EHDajI61M//UVeCzDAdn/P+SAhzEl8e0J1SH5eDr9Ktisp4pjOZumKQvzoiR0h7KKTXffa2pGe3gICav066348XK4O9dqW0ajyZOEJ5xgIPNBiHlwtxutxuNw+jAv7u7G3C2x9rYSK6h+PTpkz5//qzBYPBF/JZAf7PZXDGpUqQZTcU7th3Hx8c6PDxULpcLxBtLggOk3abfkMclvLe3tythEqSxN/iC8Nir4XAYe4hUJ96Mj8X/IXBAJ9BzTLAs4/DwUEdHR5E5RVIAa0ar++XW+PW4OdADIRmAQjQGwNjV1ZU6nY4Gg8FKeA5/Np/Px/NwcZy2cDnIiMuiYf2eJw+jpCEbzhCQD20LKIiwhL7QqDyTfcGidMbFKmVPWCeKwNdaKBQCpUagrhuP3l6H74akwR8hbtjv9/X+/Xv99NNP+vnnnyPhgJDHZDLRu3fvVKlUNJ/PV64vLBaLcTXDjz/+qLdv3+rm5kYvXrwIqY90HQwGcQDckocFACH7Ddr9fn/rw3316pWOj48lLTv9IWXR4KTQcfes+yp89vLyMvxqQlWAZzAsaWsQrWeOIRzReuwV2g//nrt/CGNlGWT04De5eQ9TIVR+/vlnvX//PjoWAuhJ0sXFRdAFgo34+fX1tTqdTlyk3O/3I0EAJueMiPOSoNNoNFYyu5jbfD7Xhw8ftl4nyibdw9vb21AA/X5fw+Ew0l5hNNBttC4uILRBEgjJLQi3bre7EhkZjUbxd+wNiSuOIiMo6vX6yoXn68bWasgRT//CXB6NRhHsZ4OQMGgPaQkwIeVgDicWrnV0xNUROEcjUzMJzZ0FPXVTxE0ntCS/w3f0S4HdvB2Px+G3uL/uphPhrMFgEOalm09oFeKczAnt525C6kZsM9KYooOBWDySwvLxi8Uwa7kJzv3V1AVg/dPp/f1HCF+/BRFwkGSQNCkhPd9NhJyONHHFUVqsL9YL0OTnDIo/mUxW8ob9TNlPt0yIFKRRB4DClI98OPL+0HiUYf2PPRzjKt8D49LymgT3sRxpZvAZRxg9HQxAC8ZN+9w60XrQmt9tO1JAjLVKy6wb1snF0czFE+Z9jxxR9P1zxNiFlVswnl/N3rl2dYLKihJzNi4M0qwlj6MiEFzbOR2kIT9nupSppdVwi/Rl4jx05P4v83zMv/PhSgFUGMGAqcs6cUn4O9aZAqM+UqHHz3yf3f/lfW4xpELXafqhsZWGTTcK04KrAEHVnPAgdGdaJsnkHURiY2ezmXq9nq6vryNZAc2DvwMTe8Cbw5aWoMG2AyJ07e2Mj2bFL4dRfG0pIAEzgpamfrek8D89lfLu7k7D4TA+T9gDDIDgOn+f1Ydlrz0+6NrQNavHQPk7Z0gXcC58PRzn8Wt8ZY8ng+YTXrq+vo7fw+TMLYsQZq+9GASFICkSOsgzdnwmFT6sD789ZSjXjLhCfs4IBL/0GdeHtUErLqjXjUcTJ/yL4cSMrZ+CBSyMyadmIpvgRE8IhFAAmsaTJ5wZ10ngFMXdZvgaeZcjnS71PNaYmtGuTVJGdmEG7I80d+Jkna7RMOOYnwMbWYcTpDOuS32PdcOwLmghOGk1FONEzhrQtCktuZnKs90yc3rzuG6WdTIvXy9rBc1dZ0F4qJCzSWPoPtfUzUstEVwrGH4dIuy0+9Umca/XCw3nt6p7qRgS3kEJh+LJAfW4mGfR+MYSSCe8glOPww6ROBMxD/4PCJAlZe8hYeK/c43t72fuHpT3OFyqbUB7SV4ACcdEdoCP/SEs5oTMOrNcqykptKjH+jhTN4FJ1u92u6EFmIMnv0hasXrckpAUqZo823PQXeP4l+MICOh1mm3TwOpzIMnPLq2wYW2sEySedETW6K6d0wvoL376eDz+wqJKBY9jIPCMg13rxkaG/fOf/6zpdKqDg4NIYmaS+B35fD5uJSdUgUTzahaQNUd4YYZGo6Hj4+Mw/cj0AE1E42JCuDYdj8fq9XqxOV7Ctu1Yd52jS8hcbnldoDMZhOVJJOwHt8xh7uJGHB8fx+cuLi7CnOfwKDnkb3nObDZTv9+P+UwmE11eXuri4mLrdUoKF0ZaDc95WWG1Wo11eiYXoZHFYhGRAjKW3PTM5/ORC4559/nz5/D/WT9Cg6gBf8+ZQ2ucfRbz/8OHD2q328FwCLtUo0tL7eqWpOcYozBKpVLQLoKFM6pWq7q+vo4IAmEcFA37Ja0yOrnq0LT70+vGoxc6s1B3smEWtILXMzIpguCgyPybCWN6woAAAl6pAhOwEJjKwSVpmTnC5nuIZ5uxycT29/BcDpd1ctAQFGtyouMLSwXi9aoQ1j+ZTOL2br7cvGNkXaevlTWmAI+byZ4yiJZFsOzt7UURxDq/FkYmfEQBAO9HIEwmkxVN6jSRz69eqpZFwxIecTOVM3a3CdCJ/fC/le4Zl7RNnsW/WTNnmmIbnFXql/Jud5d8vV/tw0pLKew+lKt1n6RPiM9QkkaJlAM7MYliMTo4ADx4qCBFlxluvnqqIKls2440VLLuuV4Wx2G5z4Gp61qCA5aWaCzpdnwOExXpj3BaZwaiGVKQIstIy8DYRw/gOxP6eaGVvEsG4bfUV08tKi8icWJ3tNz9O+awDk3dZrhPmeIJ7ua4puV9abzVq5HcZGetHlOtVCor2ERqLTBSn9pxi02A6UaGdTCBibk0lLSWuDAjIMx6vR4B8hSBw/erVCpqNpvq9/vBNOuACBbl//d8Xpg1i/mE74AJBEFxMMSZvQ+SpPDTqOahrI6kEPdvYMj9/X1Vq1XV6/Xwc0aj0YoJTvYN7+Odjh1AbFkZlr9PE9qJm3tIB6JleHiEM6WM0EM3vlaQ/lqtpsViEVlv7J+DWZylE6zPKcta09Cihw3ZS1Bi9tPNZnIJXNm4MHOQkrXSZWI0GkX82oWFg7Xpv6Ul6r+JdjcyLPZ6Kq1SiJ2Jw8yYuPwcAnap4hvJ75ygSQdkY9JSKdcQ0tJcco217UgZ1E1/NtSJ0c1UhBjr5Fk8JwW0mDN+EcTqIStH2/1A/d/83UPWx6a1PvQ3KVrs++FhNHxO37cUNXctBOGDGsOQnCka1t/n8/ka5B+6dTpxd+4hReDAEhbRujOVvkx+AFRF+KNEHCtI0WYGFsZj4ciNDEu6FVIVInVTAtie7gVUkYCSlkqlFaTNQSk0JZ+nvxH+Eygxkp4UL4gHQAInfzKZqNPp6PLyUr1eb+vDPT8/Dw3AQTgkD+MgUNDIIIM03KpUKitVIBCgxxMxs2hVglYByPLEijRtkLzi29tbXV1d6eLiQoPBYOt1StLZ2Vk0A6MBAEzoDIKb0mw24wwQuICJaWIChM9aEWLNZjOsCUnq9/thfrv74Any7BsVMOQ1bztwLcAJyFbyBBhCOKS2Qo8IEtwXF8qp4HVXB0UD3aIpCWs5LbDX+K3ebbHb7T64ro0M+8c//lF3d3f69a9/rfl8HpXzgCagmYQsHBzY399XvV5fSWRmI7zsikm32+3weUhlm8/nQZAkk0vSwcFBoKiz2X2LjdFopF6vpzdv3ujNmzeZ6mH/1//6XxoOh/r222/14sWLIEqEBM8qlUohqPr9vvL5+xYnx8fHarVaX9RZcsAexqBt5v7+foSfMInxkUEUfY8hCBiq1+vp48ePmQSTJP3P//k/1el09OrVK7169SpaqZD/DCPRdVCSOp1OdB4kTIeGYr0QNsKlWCyq0WiE4PZzpCWKn7/XWCPwEE6dTkdv3rzR+/fvt17nxcWFLi8vVavVYn0Iwuvr65hPrVaLJAZckMViEaCam7yOHGPdkNxCKG88HoclyPniGrhJ7bF1FN3l5aXevHmjz58/P7iujfbUYDAI+961ITmWaAUHZVyCMTxTxn1iD5tIqwkZjhqyyaDWkr4wZ3gnGjELw15dXUVLG0dtYdbhcLjSMAxEnPWmsVqksIMNLp2Zs/vjSFrqih1U8rCKxw1dmGw7+v1+rJP58yxye/k5n0mLuh31TNFTP1sHozhH1oobQAzdTU00sQMy0FzWdTq2Au3SYAGBwdrBCxCcHhlxq8lpl59JS7/ZaZPQo3eScCvEXS9oYNM6N2pYHGDMCkcXnTGdAEmwZpPQqJiwKcLmQADmhcflIA43pd2vgUHTGF8WH5byKIAlDpln8uW+nQMtSGH3QZ2Q0zW7GQmyiDnsgo73w0Ce/8qBZ/XtYBKYkLOCmGFU1uqBfc6JtbrWSdvn+Hc3L6nI4dwxh6Ep5sNc+JKyhXVIKvHzZA3QNEUpDpa6n+kYiwtdVzQOQvFZEkJokOdgGWvz/XWf2gHZdePROOxwOAwN40gajIM0mE6XFfMcABuVEq607F3EdxabZsYUCoXQIg5suNmUFh+4RNxmjEajMKtdc7IufEzfVE/2d83hRMs6+bejxrQN8WwYTCqPx2FZ4GOyD2i/r2FYt44gaGKJrNmFIev038G0Hupw68k1E2t1HxUB4QUCXnrHvkLUWethOVMqqwCCYFpiw57dxbz5/UPgZcpQfAZXQlqa8+w5P/PwYQqmuVXz0Hg0rOO2vYcYIFqYjolzQPiAPIeFurbxhaamoyeOe0zQpb4vGMnkYNi2w01C39AUhHKBRGKHo5FpeINn+3CmZr18sW9pNwefi7sAWdfJ+z00hBBieHogz2ZtbkGkINO6PeU7z/K1crYICjcNEdR+po4ibzMAR93kTtfpQhUaxXLwqIa0LBDxNfk60zN1IUzUJEWoEUB+po+Frh5NnHCTFGfbe+yAIKYZHm6uef5wSgipn7CzsxO9lGazWfgaziij0Uj1ej2yo9A4o9Eo7mvJQsj4VIvFskE5TAwg4o2uSSHks1gS6TpTAktR1EqlEu8hJRFBhd8+Ho+j1zICYbFYxP08Wat1JpOJer1etFJF4zIPD6lwLn7DAGcEkLOubNL/nr8BwJlOp9FgTlIkj5Azzt55B49+v6+rq6tMa/W98rAKlpL7j9CWt5OF4VhfyqDp2UL7JE7c3d2t3OPDOgFUPUSJJQVQ9tU+bK/Xi8upuOdGkn7++eeo0qd5FWYFPp20hNCRyN7qlN/zNyy4VCrp9PQ0+hcThHZfulgsqlarrVwpQZeDq6urKE/bdlDOl8/f58CCHF5dXUX4hrAV5iQmnrR0HQhP+W0ALoHZG4QAzatp0uYAE2EBZ6BWqxUSHAQ9S860JL1//17z+Vx7e3s6PT0N5H86narX62kwGISbwbmSqYRWYJ0kxjjglCYDSIpEi7u7+04anCVWAs0PAPPm83k0nEMge8/pbQYNAkB79/f3NZ1O48Ix6MjfgZ+N4GWf8cM5hzSu67F2LnwrFArhbuBqASjCxDyHkkry4jch/xsZFt81NQ9ZJAzpqDBEyXdX8y6VUg3ki+c5CIHb29tYEMAMhM27HHQCNNl20KDcgTUHt/i5x0ZhVtaGxZH6N0h6NwtZX3rYIKdIYwfkXCNA6NTHZhkIWAdzOFP3332t/nv2I82/5TluQaWayV0gSV9EIDhvZ3p+TyuhbQfNwJ25crlltpifF/9Gm6bnnyLD64BT5s13R/45U9abXqYFLxHS2xRb38iwjtKlwArE4z4tL+bznu7mxMlYZ17471OUzsMb0hJGJ2nCEdAsDOtmviOKjso6cWNieTyN/XJYP/Xt0ywX92mkVf+Rv3WhiE9GgbkzzraDd/LFmXIGHtZgENhn//0ZadYb6/XvvIOfufmZ7oWXmHkIBu207UgRdV+L7wXak7i++564RumXu3tuXfhz/bvvg68R6xOcCEvyq6t1PGTi4E5aG4jN7ofs6j5FciFyP0QncK+b9bpSjxk6EXsc1kMD246U6SmNSsMLvl7m7eYSYI6j406svCsVWqkZCQOxTvwev0DKazmzDLdgHPH3dblg9GQK9sD3Cs3A8zyNz9fk74BpHYy8u7uv6CIjqtlsrpzlNoBMuk4YI/WnPSUS0xem9Y6Znv2V7nOqtPz5jja7YHRtCj7hDcYJo24SwhsZlgnjS6VBYXfmOVB+5+1bUpQ4ZdR0I3i35zF7SIDnS18mefu/tx0uEHhmesBObGy6x5jd5/EDxAdaxyhuQTAcofXO9R56wLT7GpTYe2S5CZiu1c3CdE+Z9zptLK2CiU4rXszNz1zoQW+u+RwQzLJWd1lSZeA1z7wn9bkdPF3nl/s6GeuEH/vlPycrMP17+MfpIR2PXjdJLufHjx/DVMAPQuqwCaVSKeobMTGazWYUKiNJ3I9LfWFiUcQ+iaNBILPZTN1uV+fn55IURcOYicxt06LTARJ5fn6us7Mz1Wq18CXS60ZoS+lIcbFYVL1eD6CMdXk4ISV+99kYJMm7BkDT3t7exnwmk0mAKllCHdI9kdGWs9frqd/vB4H2ej1dXV2p2+2GAIZhiL0WCoWVW+HQCGk81rUz2gNzEE3pKD+f82w51ubVL9sOeltPJhN1u90VDQfSvVgsvggdAhqicWEup103cd3vd2vScR2PcTPYk1wuFyAc2Vmb1rmRqvEfzs/P9Q//8A9BzDs7O+EvSoqaQWkZ8iCZH2SZhfA35Om6n4L93u124xY3kFrKoQaDgXZ2dtTr9XR8fKz5fK5Op6NOp6Ner7dSFL7tuLi4iHmdnp5qMpmoXq9Hc2hieayzXC7H4REs936zrn3cP0yD5KCH+PsUBNBBg2eAGl9eXsZeon2yJBNIiiszJOnNmzfa29sLf/jq6kpXV1caj8cr/ZVZK2Z5o9EIDYXJ7ozq8WiAtMvLy0DZSY6Yz+cBJvKs6fS+wOP6+jp8yp2dncypidDBZDLRx48fIz8XV4dYcL1ej+s+3WLBHePcENIel2VtrAkkm7kiyACcHHxFCHARud84scmdexR0QoVj8klauQpRWs35JPjv/XowZ1wCYx54pQbvQwK6uYgEY7Mc3ILR8akcJNpmwOBuruXz+fBnPePG47WejshBOnC0zkRy89rX72ETmMNNdE/Pwy97rJ3IQ2tFO7rl4IAdGsJNc7RNmijicUrPWGOPwB1SdJXvXpLmtOQADcJtUxVLOlAcWAPsIe/xZB3pS9/Wzfd1tMv8oQOnXT/TdK2pP8u8sOT8es1141EfttFoqN1uq9lsRqsMDpo8WPrdwNxUbnhyQdqJIQUnUnQWomm320HApJNBLEh0T+RI0722HbzDywHddyZG65UZaFMOmk4MjjY6c6aHCbw/n89Vq9Ui2L63t7cC6kA0ng2DD5sFXJOW9am4LN6fCFcmn1+WDXK2aFz2hr0ijus51BC9/5/zZo8wh3O53Epzdk9L9FzurKmJaE8uAse6wwxF0LJOB1LZJ0kRh/ZyQsdliN+6Wc+Z1Ot1FYvFsN5QKo7BeLNBVzgPjY0MW61W1Wq1ogkbktmD59VqNa5XkBQHzOKQJBCHaxgHo9Byi8UiajS5pbtWq4XphPnm6LT/rQMd2w6eBQMyN4iSg6Xmlf/j3zgzOROkcD8Smt/t7Nz3YXbzE7OQUAYHismVgk9ZfVgES61WW0nwKBQKkcSRz+eDGWu1mur1eghrP9N1V2E4UJOGPNBclUol/LSdnZ1InKBKiXikxytTIOexgetSqVQiNx2GdGHjSSFerI7i4K6ldcoG4Ig6ZVwbYtf5/H2GFxaR93tGAWESg/inQNQX57dp0TChm4X4bEhqv+NGWnabQ8oiaV3LrCPkFPLn0Fmsx/34fZoW5/mfWQ7X/UEPQyA4qLwgswftCJFKy3xfN7E2MZNLaZ+rgx9+kIBPDs6gKbIM1yL+tzCvnzGaFReHObFWF27pGhwVZ40wxjr0mGwr/3sEA2ZslrV6Y3bOlXW5wIUR8WtdwzuSniZJpMNRf+jPmdD9ZucB94+3WefGHcCp7vf7cW8MaWZIKqQUktkTxNEGXtmxjjil1W4MHCzPnEwmKxtMYN3jp45Kuj+67QDs6XQ6Ojg40O7urhqNRrRyIXeWA07DA24au+nk2oX1or09+cJRcEoWMZvIf6XwmwynLHFJ3/PJZKKrqyudnZ0F6k2xOeE0TFfW7TfpuZXgRd0uKD2MQjMDJ05PfvEwBwAf30ejkTqdThT5bzvI2+10Ojo8PAxGxcrDdfPzSktA3a91f9T30v1a0GAEraTINYcZd3Z2wiJE6Dqg+hgm8WiLGLTncDhUo9FQtVrV4eFh9F5yIvbY2jpT5iFmRfKkuZ13d3dRK3p7e6tutxvgCFc2krYGEMbIcrj4Fbu7u3r79q1qtZpqtZqOj4/D5EfbIgg8Bi19WdiMtk3jcQghDyV4GiDAQ6lU0mJxfxMgCfukIQLorItjPzbYJ5h9MBjo9evX+tu//VuVy2W1223VarWVG+c9HoxWd7OQtaWhHdaKzwwhe4E+pYO4Uvhznz590s7OTtx4h2Dednz+/DmshMPDQ7169Uq1Wi0sBfdJEZwe93WL0M+U4T9jnZ5Z5Zl+IOl+jaoLZfKL0xj+uvFoWIf8RyQek3QCdX801Sx8dzM49b1gdI+T8SyXeCxwPp+vJE44msr3LITs6K+vk+9p4NsBFczIFBX2PeLv+Tu0FMOf77Ft5kZ4iH2B2TiXLANgZ7FY9s2C0NwdWQeWpfP1c0x/lu6Xf3c/2NeKoPYzJ/HhsZS9dOArSktNmKK+KUO6qernx99gSaVa2K1J6Ji1QhvQi7t3Lsw9i++ruyaCWPmtaUxq04I5EPcHUzicwe88I8ill2td1+TcQ+r+sqeQZWFYT3BwKbfJ3/ZQDPPmANmjh5h4HZETGnCXQlqGcfg/NxugpbKixO5bpnnfoL18ecgtNQlTn85pIRW4CCqGP8uf7WYxa+33+2H+Z0GJXYiiOaHbdULJmSodnKXT70Pn6OfMfHO53IrZ7+uEdr0hwVfHYTFDvLoDaZ/6pKlGceAgRW39/6nWTf0+TD9J4W/lcsub0kE2XTJmHT5P8oHRtOtMaz9kNjwNJ7kfnSJ/KSO7CQ2ayb562uNsNosuCvi1X7tWadUU92bYngCSCuJ160mf7+vku2toT1tFKPuZYyKyTm7sy3q2Hit2obCOWX3Obu3xHGc2H6nAkpZ3AnGOLrx5Bu6Ag7NeXfTQ2MiwEBD2dor6+oQfGilj8u+HfILU5+Mr9R9Tieib62jtNgMBg4bzn/uBpGt6aK3bvHudlHaT0ffbiRnhwOE+tv/r3osg8phjGtJK57dunemZ8j0lOv+du08e+krP31P8/CaIbYe7FP6OVEumc/SR0vhD+/HQOt3ycovpoWcBxm0CTDcy7K9//WvVajU9f/5cJycn0b1fWt6dg8b1ALn7Yi5VXJO4GSwpzAEnUswGR38dBd7f39fR0ZG+++676LxOS00aYG0zXr58qZ2dHR0eHurbb7/V8fFxFLGDlHsmlwsQJ073eVMfiMNJTTsnUNZPcB+0lFTEer2um5sbnZyc6OTkRLPZbOV28G3Gq1evVCzetyB9+fKl2u22Xr58GevlnV6d5GEkFx7pmbopLOkLNJjPet2x5wyzN4VCIdrjtlotnZ6eRvH7tuN3v/udTk5OdHh4GLiHx+ux1HxNfqapVoWmU0xF0krNuAtRrBfO14UtArNer+v29lanp6d69eqVJEWIdN3IfY0J+TSextP4xxnZarOextN4Gv+o44lhn8bT+AWNJ4Z9Gk/jFzSeGPZpPI1f0Hhi2KfxNH5B44lhn8bT+AWNJ4Z9Gk/jFzSeGPZpPI1f0Hhi2KfxNH5BY2Nq4u9///uFp4ORW5zL5SKlzAuvm82mSqWS2u22jo+P1Ww29fz5c/3ud7+L/jpeNEy+pzdSGw6H6vV6+vnnn/X999/rhx9+0KdPn3R2dqZerxdpfdRrMid6J5dKJdVqNTWbTf3X//pft0o+/cMf/rDY39+PnjpUztBTiCJrUh+5Ufvw8FBHR0eq1+s6OjrSt99+G+9nbmmZFQXonU4n7ig6OzvTp0+f4oIrqnEWi0UUf+fz+Ujp29/fV7vdVqPRULlc1t///d9vnWT7b/7Nv1kcHx9HrSt1t1dXV1FOSd0paXaNRkMHBwdR5N5ut/XNN99E+xU6UXgeLPO9vr5Wv9+PC63Ozs50dnYWRSX5fD6K5SnkICU0bYczmUz0b//tv91qrX/3d3+3oKJrNBqpUChEd33f22q1qkqlEt0gT05O4kypo+XsvQaZ1FFSLUejkS4vL/Xhwwf1ej11u12dnZ1pOBzq8vJSnU4nOirSqYUURn7WbDbjzqj//J//89p1bmTYXC6narW6ksA8nU5XLsEln3g2m6lWq6larardbkf7SDri0aPH8009R9hrUiUFoR4cHKxUbDzEsOTBlstl1Wq1yHneZtTrdTUajbjoio6FfnUHrVK4ua3VaunZs2dqtVpx6BSfQ6wQn+ehen4trUrIJ5Xuc25LpVIwZ7FYVLPZVC6Xi66BtFUl/zfLgNm5SIx9KhaL0RkBIUU/6Ha7rcPDw2hAt7+/v3K3jvfY8mby7CFF3uVyWc1mU5KirS0NEuhrBVNQ+5rmLG87EDL0rZYUnUsQSOVyOZowwKCnp6fRwwzaZ33ef9mbF3gtLGdarVZXzrRcLkdfaem+XxpCiPpu+Oerk//ZSDaLxP6Tk5OVSdNEq1KprBwujdvG47EkBQHDoDQhS1uVkmRfq9V0enoan0Hb5/P5aFsCYdGNoV6vq91uZ6rWOT091cuXL3Vzc6PLy8uoSUTS07oGQVWtVnV8fKyTk5PoGum9h9IWIHSXyOVysVakarvdjgZvzWZT/X4/tBUChCZizAem40a4LAPLoF6vR/ki2obhLTdzuZyOjo6izUqhUIiySxLo/SpOysVgWC6O3tvbi44PjUYjiJembAgOmN6tEs4yS7H+y5cv9fr1a00mE3369Cm6L0J7FBjQEI4zff78uer1egg0mNHvNmLN0rLlDiWBaEp6cmO55PP5aADf6/WiuwZNBXd3d9VqtaKE9KHxaOd/L+6lrpHGy65dIQY0ETWVFGRLy44C6zpCuBZn4aVSKVprILGocqBVJgxABT/maJbhRfDlcjm0oDMK0l9StH3lwL3/rrR674pXd/iai8ViNLCjbQmESqkfFUjeB+j29nallWzW8jrvmOB1qV4KJin2E7PZ753xSisv10PLep0utJP2ffLuDWnfa2cU/u/lgNsM6A6LhbVDX+w/jNloNIJ+oWHOzDtmePeKtEaXVjdU5SDIaEdzfX290ugNC4Y2OfDMV7c5PT09jYZrvKRWq+n169fBdHd3d3E9Hk3L0DoQeq1WW+kY5wTNoaZF0TR4Y/LValX9fj+q9KvV6grBdLvdFebKaj5x4TAMu7Nzf7G0S1S+1+v1YBg0obde4VC9fzFWAlIe7YGJtVgswrzGPE3LCtFsu7u7Ojw81O7ublgv2452ux0+G+eB5uc9dDCEMGnE5gIa4QUTsA72HvfFy85gTn8uz0g7bUL40rIVTpZSQs4Ts5YrZLhOJpfLhYuDhUOjPz4/n8+jD5Mzqjdv8LJQhvMLl3rt7u5qMpmo2WyG1cHeYGm02+24VeOhsZFh/+W//Jc6OTmJhXjbSyQ7C5OW/oZ3THDthFZ0Z9sLi/k5hF6v13VychKahTY1EDqSn9/3ej19+PAhc0f8Fy9e6OXLl+Fze82rF8MjZd03GY1GYd4hrSEUtCN/60TsVzmgSQF1/JJkzE7qZPP5/IrvxRUU247nz5/r5cuXAQDyDr9n1rv+e0cIv13NbxbkOa6tvdGbg0vQEtrPG3u7AJOW/ZHY4yz3w+KTVqvVoDmUhVsP/j7WOxqNAiBqNpshXBBcuIJYCJwLLoCb+d5hFFr3Gywmk0lYhs1mU/v7+zo7O3twXRsZ1iUHJoYXqTtw5H1cIVIkkxezpyacm5GSgqi9S59vNpvv88EnYWO8/9Q2g3kzh7T5moMAgCreEQEBgdR2y4FnuBmKLzoej7/QGo6g8zdYEQ/1JMoy3NT0eaaF3P5zPpt2u/B9cV+T//N7Opbws3QNaFh+7/6rF8BnvQvX73h1FNtp2DtBQMuAZbQhQuumdMGXt7sFXPUzTHnGrcmHWtg8NB69qsN7/bi/BUPRd9UZVloSNofNSDswIEV9oi4o+Lf7JGwSoQn+tlQqqdVqPeoHpIP+u245IFW99xDakTaVABkgnuyXr9fb2Xi3xNlspl6vt9IrF5AOcw7CdZMSAuA5Wfs6QXjeKdGtIv7NHAGX0B6EfiQFIbub48MjAN7wjflDV+7X8YVQwBXIej9so9GIGymYP5osFUQIXITwZDKJdrAegWBApy7Y2K9+vx9Kg3AQSLhbEewNvru3Xd3kzj16VQfXNOAzYgLRA8gPws2ofr8fhOExK98sN0cctPF2KUiufD4fiB7PoB8sn8OHzoqeNhqNuH0On4TnuJbgbh8aq0O89Bza29sLwnIpCjPws+l0qsFgoM+fPwfhVqvVgP89Vu0Mlu4bzJBl8Cz/W/chnWC4QRBCHg6HcZ0noRrvUeSuAwKbv5tMJtrb24vrP3gnPrsLbm+rQk+nwWCwYrY/Ng4ODkJ4uytFBAFznOdz1aP3QAazQYB4bNh9dbT/YDDQxcWFpGVYEtALdxDBDersl5+5oHxoPHpVB137nNHSvjZ+WA4m+WGmoFI6UjPSkUdpSVT+PNdeCAX+Hth9m+Hd/CWt9AFeZ564iZ92kmSkFgMM4r2A+fv0Eq801pcKOp9HVg3rc+K7a8j0PQ5G+SVcCDKfm5+zWyQOHrkwlvSFqehzRFilz9lmeIQiNYd9vm6psAeciSsUn6vvvWt/6N3PFMGV9jt7zCp5aDzKsH4REOYPL2dxSBhv20hsCqQNDYof6v1oiUchwek3jM8C0ELGD8+BQTBDgewdxcxyuC4AvIsgcd7BYKCzszNdXl6uXCA9n9/3TAamR3O5j4ZU7fV6uri4ULfbjZCUgzn5fF4HBwcB8bNP6fUnbmpmGWhIYqNONMwVv3w4HGowGITfyv27AFCEKnAdHKC7u7tTt9uNzC3wjtFotKJF3SVwhnBGurm5Cbdj2+EYgAOfaEZ81MFgoMvLS3W73TD5Qd7L5XIIY+jLLRGsleFwqG63G9GS6XS6Yg3s7+/r8vJy5T4mD2ehaGDwTb76o4GtnZ2dMNGk5b0wMBuHORqN9Pbt24DJPaGCDWu1Wrq7u4sb0Uql0oo5cX5+rqurKw2Hw0CBeR+ZU+k1lg4CIEWHw2GmEIADHw4CuLne6/X05s0b/fjjjzo/P1en0wl0ED9fUoRLnCAJknPH6fv37/Xp06cwsWGg3d1dffjwIbo+npyc6NmzZxGKIagOM5HmmGX8/PPPEdhn7i71iQv2+/1ImSTRA0bGjKSzIb4/YRMXTh8+fFC32w3Tcm9vL66AqVarmkwmqlQqgVr7HU0Ihvl8rn6/r06ns/U63TJJL21D2JFU8dNPP+ni4kJXV1exn8wP05asMp6BYCPZ5/LyUufn5xqNRnE5c6/Xi1vuh8OhKpWKjo+Pw1zHHUDZIZQ2heoe7fzvcTRJK9rN4Xi0hQNPXJPoGi9FJh2JRQL3er0ggvTOHWn9VRGpWZZF83jMk2d5sge+i/s6+LQeU0ZQOLE4GolUBU1cN/AbSQpB+qbmlKObWQZXWLjZ5nN0VBpCImkDRuQmgnK5HHskaWWOnC3CBUHrIUC0J4kN6Zk6rWQVTh6/Zj4etUAAcQkV6/TbLVAkMDlzcTPeQSRy0Z1WJcVzPOMNHiJigAv02Jk+entdiibyIgga88DjjPg5noKXmjtuu/NcwAryhmOShqr5YaZIcnpI2451f8MBA0BgHsJojtim101A8Cmy6LFjDgeTSFoiyumlzevQduadNTXR75olBugZTzwTgYSpnu65Cw3fMz9LCNL31/vzYpryPreY/LaJFCTbZqRnCqMgLGFUrgJxAYtWBgdZhx84L/idwF7Qwj5Jyx7NrnTcZXJM4KsznVwaIU2w6fk9C2WTgP/RliQUuCZMmc0XgvlDOIHFgNJ5d36eSfCdjXJtuc2AeNCsznz4KOfn52HykFeMCejJEfg7LsXJ6CHsQ/IJaDJ+MnPxqhLQREeJ+T++b5YBqo2VgO/tOd746/i7WE4esnKC40zR/mkICr/eMQxCRoPBYMUd4VxBcDEvs/rrruUdA2CPe71eVA4Nh8OV4o69vb3I1JMUAgb6I4qAduVeIq5C5WesxQXguggC9IdVs+lMN0bdPazAxBkebnDTxc0mGP0hpM2fxwZD1B4KeojBU/Oa32dBE6Uvr+FYhyBiFiJAPD2NdaEtIc507h5894yvNP7mpXluxvmzYNqsA/cC0y2dH+eXJlg4sUKozqS+935WxBjXodP83sMc0JMj9C7Yth2uINjfVFNiwbg1kApszFjPTkrpC0DK3cOUxv1M+bnTsbteXx2Hvbq6UqVSCS1BPJESNEcXYTRPvfMMJD7rhJ5qQeJy7Xb7i1ImpBYmi7QMOYA8Et8cDAYaDoeblrYysAQ8xW6xWISWw3z0uLCbeIVCIcAJCh/SG9r9cKm24c5XtCc+O6mJEDKadH9/P0Anakyz5hJz1yylix5y4tm4MuARCBeIizN0hvXYsPvd5XJZrVZrRRCkhRL+O2d26Oj8/FyXl5eZrAkUBuYttMtZA/Iw/9SnXyzua6sB6LCqAMXc9M/lciqVSjo4ONBisQgTG/pEaDk6TSREWropnOmmtNqNDPunP/1JnU5HR0dHOjk5UbPZDKIlwO9MDHLrFxix8fhBMB8mCgRQKpVUrVZXMk1IlgDhw1yhOmY6nWo8Huvjx49BdBRKX15ebn247969i4wUki/m83kgfJ1OZ6W+dTgcBvHArBwqxexcfg3CDvDWaDT04sUL7e/v6927d9rZ2dF4PA6hhkQH3h+NRiEEarVa+Ji9Xk+fPn3S+fn51uuUpDdv3sSec2G1tCxdS29QY3253GrygJuooKcg81gb7XZbs9kscp4hVn4PQ7ql5v4cAvj8/FyfPn2KsMk24/z8XMfHx1EqR1ILBeYXFxcrIJbjE4vFIqp3aLxwdHQUEQCA0Hw+H8KXfZLuFY8LA0nhgoCS8x54x8OGm9DwjTZGr9dbScPzdEPMIk/dw8SQVrtIpMnk7v/whbQh1EEW0UN3Zrp5tA79y4IoOsjlZpOb9wih8Xi8YhlgJjqK6eZXuk4Po6Sot8evXRLDGA64oZmyath+v78WREpBQASD7zvn4LFCzjRdKxaRa1DfK/fB0UTrzGG0OjHTbQd+d+prO13yfTwer2AxzBXNx+/SM3Wf3LELjwr4mSLk4CWGx8IRYg+NR++H9URoTDc0KRNiwz08wkG7BkpNWvfhPFbmh+UoYYoWuibgINicLNU6qV/Cvx0sQLswX/c5YDxP21yXCIBpRKpapVJZCZdIiv1eLBZBTAguT2dzAssyUuSa7x6O8bW5sOC9gG3kX6cYAM8EvKHSxdfpQA604z5sivhnRcRTQekAJczqZ+qgmwNrHvdH+zuzOqLMWj1agiID2PNid1+/uxFf7cOmiCWE5EkFSEnvFiEtGTWfz4d56FX8HtdCSnkSPovGnEbL4ec5qAOqOBqNNB6PY1O2Ha45fAMhKiwM1zYclsP1mNXuu7p0JqmEetR2ux2m5mAwWPHb0GAkjZABts5vzDqcCfw5vBuBB2GjqfAJC4VCtDPxEBXPm8/nEe6oVqtqNBoRs+73+yuCdzqdroSZcD14DkTN87cdaSTDBSNajpg0n/eaXM6X8/TMKUkrZ0pyTLlcjlxl3DPeTwIIyomc6kqlsgLmrQuX+djIsO6juuPuML7H1JAObmqlYFEaFGfxfIdx2VyXaE5waAL+75k4WcuwHMJ34MMFAtKVEJPnDzMPQj2+BymS7c9zqc3vOSyk8kPai/97/vI2gzNNTVjWiybzvQX15LyxFJhrit76MzlTCN/zwR28gil9z/g/Wi9LumkqSFLEnXmlGpxzYa+x+NZFLNzNY92pJmbeCHcH3PzvPBlj05k+ahJTkEuQuVKprMQDYZD5fB5xKKQiTjmSC+Z1c5rhkqfZbAaih+mCBOazPINY3mg0inQ6uvJtO1gHfobHI9EymD60ScE9IBHCASYO0wWUMwjWB1IZEAOCASBJc5kRJlTA0IEjy4BwQEtxd8jA8nAR4CDWB8TkZn8a3kpDgGjacrkc2pv4LmAcf+vRA1DeyWSiq6srdbvdTGtF4OFO5PP5APGwwmBa5osG5Cyx+FKFs86f5dzShnLsOdlS0AMo9dHRUcw51frrxkaG/fnnnzUcDsOkGQwGmk6nev36dRCT+x100+Nw8XM82AzKhgnGAmBscoGlZRI1QoB6152dnUg4p2XocDjU1dWVPn78qDdv3mQ63Ldv30auMtrl7u4u2nJeXFxoNpt9UaeKuU9LVywBhElKzMViMdL5FouFBoOBarVaACqj0WglFIE/dHV1FaVsmJDj8VhXV1eZ8mul+1Dd/v6+ms2mLi8vo7Wo53WjFfb391fanhaLxSi3xKxNCzncfGUNuVwucoYHg0G4UNK9dqlWq2q1WrFf7keTPnh5eZnJzXn79q3a7bZqtVoIh9vbW3U6HX3+/FkXFxeaTqeBm7jWbTabOjw81MHBwUpnDRgT5nVXDuuJnGFvNDebzaKxwf7+fviw+Xxe3377raRl4waKYB4aGxmW3FgOzoEBN0sdEHIpTYjC07TcTndJ5Y43w01TzNA0OO+IJv7WYwnU6RiNRgFoMDc0HhU50jIrhR5OJIAzp/QZDnik607NKkfJvZMG2t19ZXxEEMgsA83qKaSY8Z7wwrt5L8zD3vtafR3rkFT/PXvA866vr6NgwF0rP1dJcRbbjtQ/dXcGLYuVx1rpjIGCcCSf56RrczcmdfM4o7u7u5VaV/epnT/4zCZl86hJDKg0Ho+/YBhHuZgEw+Ft4nQeBvKDXffZUqmkUqm0UnbmxO6S3DfdD3zb4ei2zw3mJ6YMOowG5D2Yjgg3NxedoP3QcQHQ3I1GYyXQDpPSdA7GScMwWcvrPOyVhuE8z5jfeR65o+Sk4/k6naA5Uw/x0JGwUqmsgDA8OwX9/Gcpff2/nCkuFi4be8qZAjqxTs8bSGkrpWWPeDQajficCwdo1MFXp42vNonZMBBgDttRLaSxa1nPVnFU2ZFdD0v492KxGMF8j1v55/zwfD7OQFnAmDTTxhFs16jSKrECxvHztL2ra1L/DsPi8yLVIU5cCgQimgGt7Rrxa5L/8ZXwz/29HnP3c/Y83MVisRLqSDVrqk3dFWg0Gur1esGwDlix125yOriYVQjjo6dnOp/PV96Pyctc2V8Ec6ps3N/1s+XzzNMtEubBvmKdoVU5c49xrxuP9iVmIDWZmIcrUuTY/TgYyJkyBWR88DwHqGB8h/odoXYU18GObQebh7/hZj3vlZaACsgwQINnKaXmbrpGRwZT84o153LLIgv8SicQBKUDNlnWSqgqdWfSBAHWzLs9gd6thxQVXge2cUb4vDBCGtN209uFE/uz7YD43az39EOn1+l0Gm1VYWDWt86FeexMfc1eQuf7nCoZzjwVYul41CTGb/CF83DpHtnCbGbBKVTvqVzu16Tahw1Ba93d3QVIAeFSW4npzGEPh0N9/vxZb9++1cXFRSZpDGN4WIg4JP4ipiHEhuZDO5J25u0s07X5/wHl0Oz0yyVBBW0D0dzc3Ojq6ioQezpfZNWwkgLc8IoY73rhWheGQftA9E5Ujimk65VWz5SUVor8QU/BCrBciKv3er2okspiEvMOr7jhmQg6Fz7U6rJOF1TrQovr1p7L5SL3YD6/7+Xk4UavPgJ/4Ey5g+fq6mrjOjcybLfbVaFQiGsHjo+Pw3RCGxH6wMnncByoYDFu/7s2llaLoIHFc7lcJER79wrMQxIPFouFPn36pD//+c/6+eef4/KjbQfdGwqFQmwwYSJMtGazudLbygm41+uFj4LWZY3+XVoCIAgdRxAhakcXeedkMtFPP/0U4ZF+v6/BYPBVyf/FYjHcHPxWumF0u90QWGTnuJXE3nBPjQNhnCF74aao14zSkZC9pDEAOAAuACVw5G574spjI5e7b6BGux1CgJ1OR6PRSPl8Pgpa2HciGOw3KYuuhPwceQ9fnOnd3V0UcJCH3el0IjQJ7QwGA/3lL3+JBnw0q/tqk5jMIjKekEx+GB5sRzJLS//V079cKrvJ5P6sDzZrNlvmzJJUweYhBDzvFr9224EWT31wL+B2U0paFp6jhdBQKcjh5hHz9bW5Tyut+tNYNKCoPBtLBQskyyDmilnsAIhrWbRB6ouzLyTtp4TrZqELY08+4EyJhUKgDoLha0I7aKltx3A4XAGemD9CivWm/q2b4GAEqavg7gwYTeomuoJyc5hzhbHdrUFQfXVYxwEiD6ivA38cqWViTIrJ+N+m/+YZjHQTHVVLzU1Pa0sZa5vh6Y9ussIwHBrzcIAKczYNeTH39HBT/873GU2Gac1z8LMcafQQWpbhlo9rRPcjEbypH+9+us+Nsc5/ZX3rNDFEzp5DzP7lxJ7FJPYceBiE+TKv6XQa/YJZP/Trtc0PobnrzjQdjk2QWOTNGJiTn+NXo8TSaqE6L4GJPFE6JQS+PPPJ/QJnTtfWqT8LkaRZMf5ODsZRuSyDeWMW8z4fbDz+VopyIzTQUG6FuHnu0t4/41YBt+FJy0J1kPIUGMk6EDK8y5MVAAqZpwshzDhHk/1s1+27W0IO8KwTbOwLWj4FGHnGtoMYL7FVButAM5K55skhzN3DmtAW80iFUTo3X6uklRAlmtublztAtWlshBidOdapfR+p5kyTJNLPpd/T93q1SxpKcRMThuVZbo5sOwCKPASQAg5IU97Dvz0+hxXghMpI3Yj0dxAnn8N/xafji+HxvCyDM4WQUpOQn/H/1I1JkfJ1fmW6vhRl9WdDoB7jdHwDFJl3bjvAEdwslpbn5RrS17ZYLHN/PYEjXacLqHVCmDN0SwiFk56pW7H/Tygxk/HYHTnFmA90+M/lcisXD7Eh7XY7Ojm4acffpOYS0nuxWGbeIOHR6oPBIIqAu91ugBL8bdZwB+BZv9/XxcVF/NszXhAg5ADjU5PCd3BwoFqtttJt0Q8UX8YlLOvl5nnMetfWSF6QXYQCBc9ZgBhpNbZMsT9pnvTnZW8lRVG2A22tVivafqYJD8zHtTBrZt7D4TCErPvs+/v7UbXkXSM5hyxuDnXV/X5f5+fnWiwWOjs7C5AJQIhbLRDYuVwuEmBoRerhOndfYMrUHeJcWVcq+JwmAKiw3LDeHhqPXobF4unicHFxobdv38aEbm5uoi9vq9WSpEic9msnnGAZqfnFwQ6HQ81ms0gU98Pu9/u6vLxUv9/XwcFB+ByeoZP1cM/OzlQqlXR0dKS3b99qOBwGqkdaYrF4fxP6wcGBisViIJ3SfTIAFx6viy0yPBHEC/RZEz+nywWIIpUn5+fnK9YDaG6WgUszmUz08eNHHRwcBLERVhgOh0HIMA5J/OTZMif23kv9ABvZP86RjDmadjtTLRaLON9Wq6Wjo6NwUbwoYNvx/v17HR0dBU1+/PhRl5eX2tnZiYb09Xpdz549izMFA5nP5ystf9z9cO3v7oUXGXDODuqxDmnZ62qxWKjb7QavkFyySQg/munkpgPSA4QwLcKVlp3c07pQTNs0BzOXy8XBoD3cp/J0MWlpmrsZh8ZI/eVtB3nBmMZstrf5SE1DvwoE4IL5OcDm68Zv8qA6h7ezsxOoJc+Rlj5nmt86n88jTplloPkhONLu/EYA3sseAphwrh7C8qQYd1fQkmkqpte78g4sCH8nfwcNZkX+SWdlDlgoWEDgFKnbRV8yLx90c521+Lp4DnTM+fEsN//5PcqOM0VQjMfjjWf6qEmcppj5pnLYlBRx+zklZ56ix2YwafcNpWV7UuKYEG6j0QgCQxjMZrOVXF5PT1znI26zRsqhmCf5oMRHaVrGBUvMHdPKy7G88siZ1rO3nGBms5na7fZKvyByqSFY12YIKAc1sq4XNJp1M0/eS8OBZrMZn6Uyxb+835ZrQ3xBiJYzlpbXe3LZt5fb+R3CKZiVRQj7nIkBkzpK+muj0VCr1Yp7eV3YwOxYjw+lYcLoWCLEpheLRTyX2C7v2NnZiaQYrBG+P4b8b2RYDsBRUyQ+L+biYr5z4L64xWLxxUXJKTCVEvZisYiO8KCilUolgvYOz/u83MfYdjBXDseLDyjlazab0bTMs2gcBc7lcitmcQqQsT4P1kPgmGrX19dqNptqtVqRQEGSQ7/fXzFFeWeWwT4zB0+MQNhwayHrprMECDVEioDylkH+PNdYxF75OZcdY9Z7MQDayte3KXSybnhqoSsUrgSp1+tqtVpqtVpRYsgaECxoPxdq0peZeawb5nQzmEIHLuGu1WrR3A88iJ9vA5g+yrApI607lPTeWLfXPYbnzLkpbuVmuBP5dLq86BlTCono5llqwm5zuJLCDyZuRycKNhurge++sQiJdYzqn3GTb51/BlpIHalbIP4dofW1KDHvQ8j5hVZuJSHAvC2PpBWN42mnqbB0AsSi8HANa0dDpQg1n/8a5F9amp68o16vx72tXmie5g87UJgmg6SKwmP30BKuFXuNdVipVCJLjDI6r6B6DPnfCifncLmyAhOBsjDMQS738XietPQBXACwWL4jAPz3vnFurhJ3nc/nUbiOr+DP23bk8/dpayCkxeJ9C9Dj4+PQmlzexVrdgnCzlfX7750ZUwKXFNKbpmVoBO5F9U6Ke3t7urm5CVQ5i2BirdKymmUwGARohg/m50nbV9bEZzAt/XeskfWiiTH7Ja3Eusmf3t3dDeTckzcAq0jTzLJWfPxOpxNdI168eKHnz5+vCGGEkd/kl+YFEB/196fCl8+gUNDULpQ5X1ysZrMZ2nexWEQ/7U2m/6OgE0R2e3sbBJ3P399kTd9Xb42RapZUOsGA/m9pGcDnvUhev22Mxdzc3Ojy8jLasF5dXUX70aymk3QPUIDMjsdj/fVf/7VevHihw8PDMIkRGAijlBn9vRBoqs1gSgr7+Tnr9PkQLiC4PplM9PnzZy0WC52fn4ebkrWA3WOEg8FAvV5PR0dHQbxcbo1/ignu2tMBl/RMXdvgHhCiY634aswBRHk4HIYv9+nTp2DY1P3ZZnQ6Hb17907lclnPnj3Tq1evol80JrkrEgcD16G00KODZXzez9Q/76GgarUa/+Z8OdNCoaBerxdWwVf7sGgOzLQ0zpmauikiKn3pg0DE63wT/gZCYOHp3/Mej4tJq822ssQnMYMdAXeT3NfF3HyOjBQB52fuFvhcUyvEiZ3Pon0xJ32PCItkGUh9BzmcSB0UdBPR55qeO2Pdma4zK10bo20BEnFHPEkCZslypgg6FxYudFLsxNfFgDEdIfbnrDtTBtahYyr8LD1Tnk8q7/9zLrGnWaVq3pko9T998HPXNr55qTmb+qIpeOO/c0nGM7KMNBvFW6c4ozrh+prXgRCptve9SueXSuMUOPMMGExZbyiQZbhA8/xp2q+mITlfv491BO+/8y8XdBAmf+vAl7Q01Xm3tBTgWc4VhmI+Pud1X+7C8Pfp36VKKD3T1BVz/vHkC96JIEPbbnOmW91e50TkEtA3Ov07Hw4uOFOlTJb6tGyOfzEn9x9SICur+ZQKJICK1GRaBwb4Abv24stNqIc0sq/X46T4hl7VwXM9cSPL8H3mfbPZLJp9g/SnjJSavA8JSGeUdYJM0gpzkEKIiU+2EX6htzbNeq6SVtawDgB1bZ/SrZ9beobrPp9ahdCCm9NOyyRaAMBRorpJMG1kWCciX7CDSusO56HnpD/LIjFTxvXnEvBPEbttBxvqaZPSkrBcKGyaX/qZx4SZryf1hVOTXFpteA4Qk5WInfikZXGBN8xGOPlep3//0LMf+nlqjXlCiv/b1+qWj5TtTDGpp9PpSow1tXoemvu6z6RjncByGvBnphbhOkFIeeimd25k2MPDw4hDHR4eql6v6ze/+Y1qtZry+dUWHO6Iu6/AgpwwH/q3b4SbgzwbZJjKCdIFaRW5t7enTqej6+vruJN1m/Hb3/5WtVpNL168iFgrYSQG2VApI7nUdSmaro2fp4ACh+PaEsnrAAWJBYXCfYH/ycmJhsNhZoZ99uyZ8vm8ms2m/uqv/kqnp6c6PDyM2LdXuaRhsvQcH1or//Y1uc+Oic/6/HzxO7n869mzZzo5OQmhsu34/e9/r5OTEx0fH6/cOIGgA4mWVsOXvgbOxU3YdS5PWuzBOt20dRASlwvAkBDlwcGBXr9+rdPT0wfXlcvq7z2Np/E0/vFGtiDe03gaT+MfdTwx7NN4Gr+g8cSwT+Np/ILGE8M+jafxCxpPDPs0nsYvaDwx7NN4Gr+g8f8BmPzfh9aG38MAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 288x288 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for epoch in range(1, epochs + 1):\n",
    "    \n",
    "    # tic\n",
    "    start_time = time.time()\n",
    "    \n",
    "    for train_x, train_y in train_ds: \n",
    "        l = train_step(train_x, train_y, optimizer)\n",
    "        losses.append(l)\n",
    "\n",
    "    print(\"Epoch {}.  Loss: {}\".format(epoch, l))\n",
    "    \n",
    "    # toc \n",
    "    end_time = time.time()\n",
    "\n",
    "    display.clear_output(wait=False)\n",
    "    print('Epoch: {}, time elapse for current epoch: {}'\n",
    "        .format(epoch, end_time - start_time))\n",
    "    generate_and_save_images(epoch, test_sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2020-11-21T17:23:18.768Z"
    }
   },
   "outputs": [],
   "source": [
    "plt.plot(np.arange(len(losses)), losses)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "+ [x] training loop appears to work\n",
    "+ [ ] build the random number seed thing in "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save generated as GIF "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2020-11-21T17:23:19.566Z"
    }
   },
   "outputs": [],
   "source": [
    "import imageio, glob\n",
    "\n",
    "anim_file = 'cvae.gif'\n",
    "\n",
    "with imageio.get_writer(anim_file, mode='I') as writer:\n",
    "    filenames = glob.glob('image*.png')\n",
    "    filenames = sorted(filenames)\n",
    "    for filename in filenames:\n",
    "        image = imageio.imread(filename)\n",
    "        writer.append_data(image)\n",
    "    image = imageio.imread(filename)\n",
    "    writer.append_data(image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Testing \n",
    "Take and image, pass through network."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2020-11-21T17:23:20.109Z"
    }
   },
   "outputs": [],
   "source": [
    "for i in train_ds.take(1): \n",
    "    ims = i[0]\n",
    "    \n",
    "    \n",
    "fig, ax = plt.subplots(4, 4, figsize=(6, 6))\n",
    "ax = ax.reshape(-1) \n",
    "\n",
    "for i, ax in enumerate(ax): \n",
    "    ax.imshow(np.squeeze(ims[i]))\n",
    "    ax.axis(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pass through network "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2020-11-21T17:23:20.514Z"
    }
   },
   "outputs": [],
   "source": [
    "mean, logvar = encode(ims)\n",
    "z = reparameterize(mean, logvar)\n",
    "x_logit = decode(z)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2020-11-21T17:23:20.728Z"
    }
   },
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(4, 4, figsize=(6, 6))\n",
    "ax = ax.reshape(-1)\n",
    "\n",
    "for i, ax in enumerate(ax):\n",
    "    ax.imshow(np.squeeze(x_logit[i]))\n",
    "    ax.axis(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SAve model \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2020-11-21T17:23:21.100Z"
    },
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [],
   "source": [
    "from datetime import datetime\n",
    "import os\n",
    "\n",
    "now = datetime.now().date()\n",
    "os.makedirs('saved_model/' + str(now) + '/encoder/')\n",
    "os.makedirs('saved_model/' + str(now) + '/decoder/')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2020-11-21T17:23:21.304Z"
    }
   },
   "outputs": [],
   "source": [
    "encoder.save('saved_model/' + str(now) + '/encoder')\n",
    "decoder.save('saved_model/' + str(now) + '/decoder')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2020-11-21T17:23:21.508Z"
    }
   },
   "outputs": [],
   "source": [
    "losses_lst = [l.numpy() for l in losses]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2020-11-21T17:23:21.688Z"
    }
   },
   "outputs": [],
   "source": [
    "with open('saved_model/' + str(now) + '/losses.txt', 'w') as f:\n",
    "    for item in losses_lst:\n",
    "        f.write(\"%s\\n\" % item)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# TODO \n",
    "+ Research autoencoder architecture and loss functions that could help with overblurring "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.8"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}