Model Training.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = []\n",
    "ripeness_class_name = ['Early Ripe', 'Partially Ripe'\n",
    "                        ,'Ripe', 'Decay']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('data.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1>TCS34725 Data</h1>\n",
    "\n",
    "All ranges are in nm\n",
    "\n",
    "\n",
    "![alt text](nir-band-erf-community_636774556093727541.jpg \"Ranges\")\n",
    "\n",
    "\n",
    "\n",
    "We can infere:\n",
    "\n",
    "600nm Above is Red\n",
    "\n",
    "500-600nm is Green\n",
    "\n",
    "and <500nm is Blue"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "#we have to normalise bgr to 0-255\n",
    "#fetch max\n",
    "column = df[\"Blue\"]\n",
    "max_b = column.max()\n",
    "\n",
    "column = df[\"Red\"]\n",
    "max_r = column.max()\n",
    "\n",
    "column = df[\"Green\"]\n",
    "max_g = column.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "df[\"Blue\"] = (df[\"Blue\"]/max_b)*255\n",
    "\n",
    "df[\"Red\"] = (df[\"Red\"]/max_r)*255\n",
    "\n",
    "df[\"Green\"] = (df[\"Green\"]/max_g)*255\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NIR_1</th>\n",
       "      <th>NIR_2</th>\n",
       "      <th>Ripeness</th>\n",
       "      <th>Blue</th>\n",
       "      <th>Green</th>\n",
       "      <th>Red</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>54</td>\n",
       "      <td>66</td>\n",
       "      <td>0</td>\n",
       "      <td>7.391304</td>\n",
       "      <td>20.485</td>\n",
       "      <td>11.7300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>62</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "      <td>9.327122</td>\n",
       "      <td>30.005</td>\n",
       "      <td>14.0250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>68</td>\n",
       "      <td>54</td>\n",
       "      <td>0</td>\n",
       "      <td>10.734990</td>\n",
       "      <td>32.215</td>\n",
       "      <td>14.0250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>227</td>\n",
       "      <td>215</td>\n",
       "      <td>0</td>\n",
       "      <td>54.554865</td>\n",
       "      <td>220.065</td>\n",
       "      <td>204.5100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>126</td>\n",
       "      <td>129</td>\n",
       "      <td>0</td>\n",
       "      <td>27.453416</td>\n",
       "      <td>158.525</td>\n",
       "      <td>133.8750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>325</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0.351967</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1.5300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>326</th>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>1.583851</td>\n",
       "      <td>5.525</td>\n",
       "      <td>7.7775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>327</th>\n",
       "      <td>16</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>1.935818</td>\n",
       "      <td>0.510</td>\n",
       "      <td>0.8925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>328</th>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>1.407867</td>\n",
       "      <td>0.510</td>\n",
       "      <td>1.2750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>329</th>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>3</td>\n",
       "      <td>0.703934</td>\n",
       "      <td>0.680</td>\n",
       "      <td>1.1475</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>330 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     NIR_1  NIR_2  Ripeness       Blue    Green       Red\n",
       "0       54     66         0   7.391304   20.485   11.7300\n",
       "1       62     58         0   9.327122   30.005   14.0250\n",
       "2       68     54         0  10.734990   32.215   14.0250\n",
       "3      227    215         0  54.554865  220.065  204.5100\n",
       "4      126    129         0  27.453416  158.525  133.8750\n",
       "..     ...    ...       ...        ...      ...       ...\n",
       "325      4      0         3   0.351967    0.000    1.5300\n",
       "326     10     10         3   1.583851    5.525    7.7775\n",
       "327     16      8         3   1.935818    0.510    0.8925\n",
       "328     12     10         3   1.407867    0.510    1.2750\n",
       "329     10     12         3   0.703934    0.680    1.1475\n",
       "\n",
       "[330 rows x 6 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "some_frame = pd.read_csv('final.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = some_frame[['Red','Green','Blue','NIR_1','NIR_2','Ripeness']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Red</th>\n",
       "      <th>Green</th>\n",
       "      <th>Blue</th>\n",
       "      <th>NIR_1</th>\n",
       "      <th>NIR_2</th>\n",
       "      <th>Ripeness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11.7300</td>\n",
       "      <td>20.485</td>\n",
       "      <td>7.391304</td>\n",
       "      <td>54</td>\n",
       "      <td>66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>14.0250</td>\n",
       "      <td>30.005</td>\n",
       "      <td>9.327122</td>\n",
       "      <td>62</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>14.0250</td>\n",
       "      <td>32.215</td>\n",
       "      <td>10.734990</td>\n",
       "      <td>68</td>\n",
       "      <td>54</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>204.5100</td>\n",
       "      <td>220.065</td>\n",
       "      <td>54.554865</td>\n",
       "      <td>227</td>\n",
       "      <td>215</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>133.8750</td>\n",
       "      <td>158.525</td>\n",
       "      <td>27.453416</td>\n",
       "      <td>126</td>\n",
       "      <td>129</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>325</th>\n",
       "      <td>1.5300</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.351967</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>326</th>\n",
       "      <td>7.7775</td>\n",
       "      <td>5.525</td>\n",
       "      <td>1.583851</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>327</th>\n",
       "      <td>0.8925</td>\n",
       "      <td>0.510</td>\n",
       "      <td>1.935818</td>\n",
       "      <td>16</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>328</th>\n",
       "      <td>1.2750</td>\n",
       "      <td>0.510</td>\n",
       "      <td>1.407867</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>329</th>\n",
       "      <td>1.1475</td>\n",
       "      <td>0.680</td>\n",
       "      <td>0.703934</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>330 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          Red    Green       Blue  NIR_1  NIR_2  Ripeness\n",
       "0     11.7300   20.485   7.391304     54     66         0\n",
       "1     14.0250   30.005   9.327122     62     58         0\n",
       "2     14.0250   32.215  10.734990     68     54         0\n",
       "3    204.5100  220.065  54.554865    227    215         0\n",
       "4    133.8750  158.525  27.453416    126    129         0\n",
       "..        ...      ...        ...    ...    ...       ...\n",
       "325    1.5300    0.000   0.351967      4      0         3\n",
       "326    7.7775    5.525   1.583851     10     10         3\n",
       "327    0.8925    0.510   1.935818     16      8         3\n",
       "328    1.2750    0.510   1.407867     12     10         3\n",
       "329    1.1475    0.680   0.703934     10     12         3\n",
       "\n",
       "[330 rows x 6 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv('Final1.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('Final1.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>Red</th>\n",
       "      <th>Green</th>\n",
       "      <th>Blue</th>\n",
       "      <th>NIR_1</th>\n",
       "      <th>NIR_2</th>\n",
       "      <th>Ripeness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>11.7300</td>\n",
       "      <td>20.485</td>\n",
       "      <td>7.391304</td>\n",
       "      <td>54</td>\n",
       "      <td>66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>14.0250</td>\n",
       "      <td>30.005</td>\n",
       "      <td>9.327122</td>\n",
       "      <td>62</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>14.0250</td>\n",
       "      <td>32.215</td>\n",
       "      <td>10.734990</td>\n",
       "      <td>68</td>\n",
       "      <td>54</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>204.5100</td>\n",
       "      <td>220.065</td>\n",
       "      <td>54.554865</td>\n",
       "      <td>227</td>\n",
       "      <td>215</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>133.8750</td>\n",
       "      <td>158.525</td>\n",
       "      <td>27.453416</td>\n",
       "      <td>126</td>\n",
       "      <td>129</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>325</th>\n",
       "      <td>325</td>\n",
       "      <td>1.5300</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.351967</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>326</th>\n",
       "      <td>326</td>\n",
       "      <td>7.7775</td>\n",
       "      <td>5.525</td>\n",
       "      <td>1.583851</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>327</th>\n",
       "      <td>327</td>\n",
       "      <td>0.8925</td>\n",
       "      <td>0.510</td>\n",
       "      <td>1.935818</td>\n",
       "      <td>16</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>328</th>\n",
       "      <td>328</td>\n",
       "      <td>1.2750</td>\n",
       "      <td>0.510</td>\n",
       "      <td>1.407867</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>329</th>\n",
       "      <td>329</td>\n",
       "      <td>1.1475</td>\n",
       "      <td>0.680</td>\n",
       "      <td>0.703934</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>330 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Unnamed: 0       Red    Green       Blue  NIR_1  NIR_2  Ripeness\n",
       "0             0   11.7300   20.485   7.391304     54     66         0\n",
       "1             1   14.0250   30.005   9.327122     62     58         0\n",
       "2             2   14.0250   32.215  10.734990     68     54         0\n",
       "3             3  204.5100  220.065  54.554865    227    215         0\n",
       "4             4  133.8750  158.525  27.453416    126    129         0\n",
       "..          ...       ...      ...        ...    ...    ...       ...\n",
       "325         325    1.5300    0.000   0.351967      4      0         3\n",
       "326         326    7.7775    5.525   1.583851     10     10         3\n",
       "327         327    0.8925    0.510   1.935818     16      8         3\n",
       "328         328    1.2750    0.510   1.407867     12     10         3\n",
       "329         329    1.1475    0.680   0.703934     10     12         3\n",
       "\n",
       "[330 rows x 7 columns]"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = []\n",
    "for i in range(len(df)):\n",
    "    inputs.append([int(df['Red'][i]),int(df['Blue'][i]),int(df['Green'][i])])\n",
    "\n",
    "inputs = np.asarray(inputs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 11,   7,  20],\n",
       "       [ 14,   9,  30],\n",
       "       [ 14,  10,  32],\n",
       "       [204,  54, 220],\n",
       "       [133,  27, 158],\n",
       "       [203,  50, 217],\n",
       "       [255, 107, 255],\n",
       "       [255, 114, 255],\n",
       "       [255,  87, 238],\n",
       "       [255,  53, 155],\n",
       "       [255,  65, 181],\n",
       "       [255,  53, 146],\n",
       "       [124,  17,  64],\n",
       "       [ 43,   7,  15],\n",
       "       [129,  18,  60],\n",
       "       [114,  12,  24],\n",
       "       [139,  14,  50],\n",
       "       [199,  77, 155],\n",
       "       [255,  44, 120],\n",
       "       [255,  54, 135],\n",
       "       [255,  60, 179],\n",
       "       [255,  67, 184],\n",
       "       [225,  26,  74],\n",
       "       [255,  67, 184],\n",
       "       [255,  67, 184],\n",
       "       [255,  42,  57],\n",
       "       [255,  35,  28],\n",
       "       [255,  45,  62],\n",
       "       [243,  43, 191],\n",
       "       [224,  31, 143],\n",
       "       [  2,   1,   0],\n",
       "       [  8,   6,   6],\n",
       "       [  1,   1,   0],\n",
       "       [ 52,  27,  90],\n",
       "       [  9,   6,  19],\n",
       "       [ 14,  11,  27],\n",
       "       [227,  75, 246],\n",
       "       [191,  42, 210],\n",
       "       [126,  27, 152],\n",
       "       [255,  87, 242],\n",
       "       [255,  71, 220],\n",
       "       [255,  77, 226],\n",
       "       [255,  41, 130],\n",
       "       [255,  46, 145],\n",
       "       [255,  58, 174],\n",
       "       [ 90,  13,  31],\n",
       "       [ 89,  14,  37],\n",
       "       [119,  15,  54],\n",
       "       [187,  20,  55],\n",
       "       [240,  54, 183],\n",
       "       [142,  14,  39],\n",
       "       [255,  47, 126],\n",
       "       [255,  64, 148],\n",
       "       [255,  42, 117],\n",
       "       [ 68,   6,   9],\n",
       "       [ 75,   8,  13],\n",
       "       [143,  16,  73],\n",
       "       [255,  25,  26],\n",
       "       [255,  25,  34],\n",
       "       [255,  25,  27],\n",
       "       [217,  35, 177],\n",
       "       [180,  26, 134],\n",
       "       [ 36,   8,  20],\n",
       "       [  0,   2,   0],\n",
       "       [  1,   1,   0],\n",
       "       [  1,   2,   0],\n",
       "       [ 25,  17,  46],\n",
       "       [ 24,  21,  41],\n",
       "       [ 17,  15,  25],\n",
       "       [242,  81, 242],\n",
       "       [253,  88, 254],\n",
       "       [255, 224, 255],\n",
       "       [255, 143, 255],\n",
       "       [255, 161, 255],\n",
       "       [255, 156, 255],\n",
       "       [255,  69, 160],\n",
       "       [255,  66, 154],\n",
       "       [255,  82, 181],\n",
       "       [213,  34, 143],\n",
       "       [118,  18,  46],\n",
       "       [156,  25,  49],\n",
       "       [234,  28,  81],\n",
       "       [255,  38, 122],\n",
       "       [255,  74, 184],\n",
       "       [255,  67, 139],\n",
       "       [255,  75, 148],\n",
       "       [255,  77, 152],\n",
       "       [122,  16,  29],\n",
       "       [147,  19,  42],\n",
       "       [157,  16,  24],\n",
       "       [255,  38,  41],\n",
       "       [255,  41,  52],\n",
       "       [240,  27,  18],\n",
       "       [255,  47, 196],\n",
       "       [123,  16,  71],\n",
       "       [125,  16,  74],\n",
       "       [  1,   1,   0],\n",
       "       [  2,   1,   1],\n",
       "       [  2,   0,   1],\n",
       "       [ 22,  12,  41],\n",
       "       [ 20,  11,  37],\n",
       "       [ 30,  19,  58],\n",
       "       [235,  86, 249],\n",
       "       [150,  31, 190],\n",
       "       [149,  26, 152],\n",
       "       [255,  74, 223],\n",
       "       [255,  87, 236],\n",
       "       [255,  70, 227],\n",
       "       [255,  77, 190],\n",
       "       [255,  60, 175],\n",
       "       [231,  41, 142],\n",
       "       [221,  73, 196],\n",
       "       [222,  93, 215],\n",
       "       [114,  22,  47],\n",
       "       [210,  33, 159],\n",
       "       [116,  12,  43],\n",
       "       [185,  26, 131],\n",
       "       [210,  40, 104],\n",
       "       [131, 232,  85],\n",
       "       [136, 204,  80],\n",
       "       [192,  20,  70],\n",
       "       [148,  16,  26],\n",
       "       [165,  43, 105],\n",
       "       [227,  26,  27],\n",
       "       [177,  17,  11],\n",
       "       [203,  20,  18],\n",
       "       [114,  15,  74],\n",
       "       [ 67,  12,  38],\n",
       "       [ 66,   7,  33],\n",
       "       [  1,   1,   0],\n",
       "       [  0,   1,   0],\n",
       "       [  1,   1,   0],\n",
       "       [ 91,  39, 125],\n",
       "       [111,  83, 174],\n",
       "       [ 48,  33,  79],\n",
       "       [255, 225, 255],\n",
       "       [255, 255, 255],\n",
       "       [255, 177, 255],\n",
       "       [255,  87, 236],\n",
       "       [255,  91, 238],\n",
       "       [255,  85, 229],\n",
       "       [255,  74, 188],\n",
       "       [255, 118, 205],\n",
       "       [255,  94, 193],\n",
       "       [255, 124, 186],\n",
       "       [255, 159, 236],\n",
       "       [255, 164, 211],\n",
       "       [226,  24,  60],\n",
       "       [240,  36, 125],\n",
       "       [255,  58, 181],\n",
       "       [255,  83, 160],\n",
       "       [255, 146, 208],\n",
       "       [255,  83, 163],\n",
       "       [255,  96, 181],\n",
       "       [255,  50,  89],\n",
       "       [211,  29, 127],\n",
       "       [255,  39,  43],\n",
       "       [255,  36,  37],\n",
       "       [247,  29,  19],\n",
       "       [ 84,  11,  42],\n",
       "       [186,  28, 119],\n",
       "       [224,  33, 155],\n",
       "       [  1,   1,   0],\n",
       "       [  1,   1,   0],\n",
       "       [  2,   1,   1],\n",
       "       [ 33,  23,  52],\n",
       "       [ 23,  20,  34],\n",
       "       [ 64,  43,  97],\n",
       "       [255, 241, 255],\n",
       "       [255, 183, 255],\n",
       "       [255, 168, 255],\n",
       "       [255, 217, 255],\n",
       "       [255, 135, 255],\n",
       "       [255, 183, 255],\n",
       "       [255, 102, 190],\n",
       "       [255, 127, 198],\n",
       "       [255, 164, 216],\n",
       "       [255, 165, 239],\n",
       "       [255, 103, 211],\n",
       "       [213,  28,  80],\n",
       "       [255,  81, 184],\n",
       "       [255,  41, 121],\n",
       "       [249,  28,  61],\n",
       "       [255,  51, 117],\n",
       "       [255,  78, 138],\n",
       "       [255,  61, 126],\n",
       "       [255,  33,  69],\n",
       "       [255,  46, 132],\n",
       "       [255,  35, 102],\n",
       "       [255,  36,  24],\n",
       "       [255,  34,  23],\n",
       "       [255,  42,  40],\n",
       "       [ 56,   7,  21],\n",
       "       [131,  17,  69],\n",
       "       [ 82,  11,  31],\n",
       "       [  2,   3,   1],\n",
       "       [  2,   1,   1],\n",
       "       [  2,   3,   1],\n",
       "       [ 30,  21,  48],\n",
       "       [ 26,  19,  47],\n",
       "       [ 19,  16,  29],\n",
       "       [255, 222, 255],\n",
       "       [255, 233, 255],\n",
       "       [253, 110, 243],\n",
       "       [255, 136, 255],\n",
       "       [255, 124, 255],\n",
       "       [255, 125, 255],\n",
       "       [255, 129, 207],\n",
       "       [255, 100, 195],\n",
       "       [255,  67, 170],\n",
       "       [228,  54, 195],\n",
       "       [255, 108, 216],\n",
       "       [222,  52, 150],\n",
       "       [196,  20,  47],\n",
       "       [181,  17,  40],\n",
       "       [242,  48, 163],\n",
       "       [254,  37, 101],\n",
       "       [255, 126, 198],\n",
       "       [255,  48, 122],\n",
       "       [ 79,  14,  21],\n",
       "       [243,  39, 140],\n",
       "       [148,  13,  33],\n",
       "       [185,  19,  12],\n",
       "       [196,  18,  12],\n",
       "       [198,  19,  12],\n",
       "       [ 18,   0,  12],\n",
       "       [ 21,   3,  17],\n",
       "       [ 19,   2,  14],\n",
       "       [  0,   0,   0],\n",
       "       [  0,   0,   0],\n",
       "       [  0,   0,   0],\n",
       "       [ 29,  21,  46],\n",
       "       [ 36,  24,  57],\n",
       "       [ 31,  21,  50],\n",
       "       [251, 136, 255],\n",
       "       [219,  86, 221],\n",
       "       [255, 126, 249],\n",
       "       [255,  60, 209],\n",
       "       [227,  41, 189],\n",
       "       [255,  58, 213],\n",
       "       [255,  52, 158],\n",
       "       [255,  80, 192],\n",
       "       [255,  71, 188],\n",
       "       [209,  42, 106],\n",
       "       [226,  63, 204],\n",
       "       [ 94,  12,  27],\n",
       "       [138,  12,  29],\n",
       "       [214,  35, 142],\n",
       "       [141,  15,  33],\n",
       "       [255,  50, 126],\n",
       "       [255,  45, 116],\n",
       "       [212,  21,  72],\n",
       "       [222,  35, 115],\n",
       "       [182,  24,  50],\n",
       "       [147,  37,  65],\n",
       "       [236,  20,  21],\n",
       "       [237,  21,  20],\n",
       "       [201,  19,  13],\n",
       "       [ 19,   2,   6],\n",
       "       [ 10,   0,   2],\n",
       "       [  9,   0,   2],\n",
       "       [  3,   1,   1],\n",
       "       [  1,   1,   0],\n",
       "       [  2,   2,   1],\n",
       "       [ 47,  34,  74],\n",
       "       [ 53,  35,  82],\n",
       "       [ 18,  16,  23],\n",
       "       [255, 213, 255],\n",
       "       [250,  98, 255],\n",
       "       [255, 250, 255],\n",
       "       [255, 109, 251],\n",
       "       [255, 179, 255],\n",
       "       [255, 197, 255],\n",
       "       [255, 129, 208],\n",
       "       [255,  75, 180],\n",
       "       [255, 138, 215],\n",
       "       [242,  65, 123],\n",
       "       [225,  35,  77],\n",
       "       [210,  39, 157],\n",
       "       [190,  19,  40],\n",
       "       [245,  44, 151],\n",
       "       [222,  23,  52],\n",
       "       [ 99,  18,  46],\n",
       "       [232,  30,  84],\n",
       "       [255,  35,  85],\n",
       "       [224,  35,  74],\n",
       "       [247,  92, 166],\n",
       "       [255, 139, 212],\n",
       "       [204,  22,  13],\n",
       "       [213,  23,  15],\n",
       "       [255,  26,  26],\n",
       "       [  2,   0,   0],\n",
       "       [  2,   0,   0],\n",
       "       [  3,   0,   0],\n",
       "       [  1,   1,   0],\n",
       "       [  1,   1,   0],\n",
       "       [  1,   0,   0],\n",
       "       [ 28,  18,  53],\n",
       "       [ 64,  39, 113],\n",
       "       [ 39,  26,  75],\n",
       "       [220,  67, 232],\n",
       "       [230,  81, 235],\n",
       "       [209,  49, 206],\n",
       "       [255,  73, 224],\n",
       "       [255,  85, 231],\n",
       "       [255,  68, 216],\n",
       "       [255,  76, 183],\n",
       "       [255,  74, 189],\n",
       "       [255,  68, 186],\n",
       "       [177,  23,  63],\n",
       "       [208,  39, 150],\n",
       "       [152,  22,  85],\n",
       "       [144,  14,  29],\n",
       "       [158,  16,  37],\n",
       "       [210,  34, 143],\n",
       "       [237,  27,  65],\n",
       "       [255,  31,  77],\n",
       "       [255,  29,  78],\n",
       "       [222,  21,  53],\n",
       "       [188,  20,  53],\n",
       "       [168,  21,  72],\n",
       "       [174,  17,  10],\n",
       "       [171,  17,  10],\n",
       "       [211,  20,  16],\n",
       "       [  2,   0,   0],\n",
       "       [  1,   0,   0],\n",
       "       [  7,   1,   5],\n",
       "       [  0,   1,   0],\n",
       "       [  1,   1,   0],\n",
       "       [  1,   0,   0]])"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = df.pop('Ripeness')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = np.asarray(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3],\n",
       "      dtype=int64)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "330"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "330"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(inputs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(330, 3)"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inputs.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(inputs, labels, test_size=0.05)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 23,  20,  34],\n",
       "       [255,  45,  62],\n",
       "       [231,  41, 142],\n",
       "       [242,  81, 242],\n",
       "       [142,  14,  39],\n",
       "       [255, 102, 190],\n",
       "       [255,  76, 183],\n",
       "       [133,  27, 158],\n",
       "       [255,  38,  41],\n",
       "       [222,  52, 150],\n",
       "       [255,  47, 196],\n",
       "       [182,  24,  50],\n",
       "       [255,  74, 188],\n",
       "       [143,  16,  73],\n",
       "       [232,  30,  84],\n",
       "       [255,  71, 188],\n",
       "       [ 17,  15,  25],\n",
       "       [217,  35, 177],\n",
       "       [224,  33, 155],\n",
       "       [199,  77, 155],\n",
       "       [  0,   0,   0],\n",
       "       [242,  48, 163],\n",
       "       [  2,   0,   0],\n",
       "       [255,  35, 102],\n",
       "       [ 43,   7,  15],\n",
       "       [224,  31, 143],\n",
       "       [198,  19,  12],\n",
       "       [230,  81, 235],\n",
       "       [  9,   0,   2],\n",
       "       [236,  20,  21],\n",
       "       [148,  13,  33],\n",
       "       [119,  15,  54],\n",
       "       [114,  12,  24],\n",
       "       [177,  23,  63],\n",
       "       [  1,   1,   0],\n",
       "       [ 25,  17,  46],\n",
       "       [  1,   1,   0],\n",
       "       [255,  83, 160],\n",
       "       [222,  93, 215],\n",
       "       [ 91,  39, 125],\n",
       "       [255, 107, 255],\n",
       "       [255,  35,  85],\n",
       "       [255, 126, 198],\n",
       "       [224,  35,  74],\n",
       "       [255, 108, 216],\n",
       "       [203,  20,  18],\n",
       "       [  3,   0,   0],\n",
       "       [255, 183, 255],\n",
       "       [ 26,  19,  47],\n",
       "       [255,  87, 236],\n",
       "       [ 67,  12,  38],\n",
       "       [168,  21,  72],\n",
       "       [ 14,   9,  30],\n",
       "       [ 28,  18,  53],\n",
       "       [204,  22,  13],\n",
       "       [255,  51, 117],\n",
       "       [255,  60, 175],\n",
       "       [255,  67, 184],\n",
       "       [196,  20,  47],\n",
       "       [  2,   0,   0],\n",
       "       [255, 164, 211],\n",
       "       [ 68,   6,   9],\n",
       "       [  1,   0,   0],\n",
       "       [255,  67, 139],\n",
       "       [255,  42, 117],\n",
       "       [255, 179, 255],\n",
       "       [  8,   6,   6],\n",
       "       [255, 129, 207],\n",
       "       [255,  54, 135],\n",
       "       [222,  21,  53],\n",
       "       [255, 159, 236],\n",
       "       [196,  18,  12],\n",
       "       [255,  82, 181],\n",
       "       [ 99,  18,  46],\n",
       "       [  0,   0,   0],\n",
       "       [255,  46, 132],\n",
       "       [  1,   2,   0],\n",
       "       [255,  74, 223],\n",
       "       [255, 135, 255],\n",
       "       [  2,   2,   1],\n",
       "       [255, 129, 208],\n",
       "       [180,  26, 134],\n",
       "       [243,  39, 140],\n",
       "       [227,  41, 189],\n",
       "       [114,  22,  47],\n",
       "       [255,  25,  34],\n",
       "       [255, 225, 255],\n",
       "       [250,  98, 255],\n",
       "       [255,  83, 163],\n",
       "       [177,  17,  11],\n",
       "       [255, 177, 255],\n",
       "       [213,  28,  80],\n",
       "       [255,  41,  52],\n",
       "       [  1,   1,   0],\n",
       "       [255, 156, 255],\n",
       "       [255,  81, 184],\n",
       "       [203,  50, 217],\n",
       "       [187,  20,  55],\n",
       "       [228,  54, 195],\n",
       "       [111,  83, 174],\n",
       "       [210,  34, 143],\n",
       "       [  1,   1,   0],\n",
       "       [253,  88, 254],\n",
       "       [255,  87, 238],\n",
       "       [  0,   0,   0],\n",
       "       [  2,   1,   0],\n",
       "       [  2,   1,   1],\n",
       "       [214,  35, 142],\n",
       "       [255, 217, 255],\n",
       "       [190,  19,  40],\n",
       "       [114,  15,  74],\n",
       "       [255,  42,  57],\n",
       "       [255,  68, 186],\n",
       "       [191,  42, 210],\n",
       "       [255, 224, 255],\n",
       "       [174,  17,  10],\n",
       "       [ 64,  39, 113],\n",
       "       [255,  60, 209],\n",
       "       [255,  39,  43],\n",
       "       [209,  42, 106],\n",
       "       [255,  42,  40],\n",
       "       [255,  25,  26],\n",
       "       [  1,   0,   0],\n",
       "       [251, 136, 255],\n",
       "       [255, 164, 216],\n",
       "       [181,  17,  40],\n",
       "       [  1,   1,   0],\n",
       "       [211,  20,  16],\n",
       "       [255,  25,  27],\n",
       "       [ 47,  34,  74],\n",
       "       [255,  85, 231],\n",
       "       [ 66,   7,  33],\n",
       "       [247,  29,  19],\n",
       "       [255, 109, 251],\n",
       "       [ 22,  12,  41],\n",
       "       [255,  33,  69],\n",
       "       [ 11,   7,  20],\n",
       "       [255, 213, 255],\n",
       "       [226,  63, 204],\n",
       "       [  2,   3,   1],\n",
       "       [255,  58, 174],\n",
       "       [255,  74, 184],\n",
       "       [255,  58, 213],\n",
       "       [  1,   1,   0],\n",
       "       [ 79,  14,  21],\n",
       "       [226,  24,  60],\n",
       "       [ 82,  11,  31],\n",
       "       [255,  96, 181],\n",
       "       [139,  14,  50],\n",
       "       [255,  67, 184],\n",
       "       [ 19,   2,  14],\n",
       "       [ 30,  21,  48],\n",
       "       [186,  28, 119],\n",
       "       [255, 126, 249],\n",
       "       [220,  67, 232],\n",
       "       [253, 110, 243],\n",
       "       [221,  73, 196],\n",
       "       [150,  31, 190],\n",
       "       [255, 124, 255],\n",
       "       [255,  53, 146],\n",
       "       [255,  74, 189],\n",
       "       [  1,   1,   0],\n",
       "       [210,  33, 159],\n",
       "       [  1,   1,   0],\n",
       "       [213,  34, 143],\n",
       "       [255, 114, 255],\n",
       "       [210,  39, 157],\n",
       "       [  2,   1,   1],\n",
       "       [147,  37,  65],\n",
       "       [ 24,  21,  41],\n",
       "       [219,  86, 221],\n",
       "       [165,  43, 105],\n",
       "       [255,  35,  28],\n",
       "       [255, 146, 208],\n",
       "       [255,  75, 148],\n",
       "       [255,  67, 184],\n",
       "       [255,  78, 138],\n",
       "       [255, 103, 211],\n",
       "       [185,  26, 131],\n",
       "       [255,  69, 160],\n",
       "       [123,  16,  71],\n",
       "       [ 56,   7,  21],\n",
       "       [ 53,  35,  82],\n",
       "       [255, 100, 195],\n",
       "       [255,  61, 126],\n",
       "       [245,  44, 151],\n",
       "       [255,  47, 126],\n",
       "       [ 89,  14,  37],\n",
       "       [ 52,  27,  90],\n",
       "       [ 21,   3,  17],\n",
       "       [227,  26,  27],\n",
       "       [201,  19,  13],\n",
       "       [227,  75, 246],\n",
       "       [ 19,  16,  29],\n",
       "       [255,  44, 120],\n",
       "       [185,  19,  12],\n",
       "       [255,  73, 224],\n",
       "       [222,  23,  52],\n",
       "       [255,  67, 170],\n",
       "       [  2,   1,   1],\n",
       "       [255, 124, 186],\n",
       "       [141,  15,  33],\n",
       "       [ 20,  11,  37],\n",
       "       [255,  68, 216],\n",
       "       [255,  34,  23],\n",
       "       [208,  39, 150],\n",
       "       [171,  17,  10],\n",
       "       [156,  25,  49],\n",
       "       [149,  26, 152],\n",
       "       [255, 183, 255],\n",
       "       [255,  65, 181],\n",
       "       [  0,   1,   0],\n",
       "       [255,  77, 226],\n",
       "       [242,  65, 123],\n",
       "       [152,  22,  85],\n",
       "       [  1,   1,   0],\n",
       "       [188,  20,  53],\n",
       "       [209,  49, 206],\n",
       "       [  1,   0,   0],\n",
       "       [210,  40, 104],\n",
       "       [222,  35, 115],\n",
       "       [237,  21,  20],\n",
       "       [255,  36,  37],\n",
       "       [148,  16,  26],\n",
       "       [255, 250, 255],\n",
       "       [255,  53, 155],\n",
       "       [136, 204,  80],\n",
       "       [255,  41, 121],\n",
       "       [255,  31,  77],\n",
       "       [243,  43, 191],\n",
       "       [255, 136, 255],\n",
       "       [249,  28,  61],\n",
       "       [255, 233, 255],\n",
       "       [213,  23,  15],\n",
       "       [255,  91, 238],\n",
       "       [  0,   1,   0],\n",
       "       [255,  71, 220],\n",
       "       [ 19,   2,   6],\n",
       "       [255, 222, 255],\n",
       "       [131, 232,  85],\n",
       "       [235,  86, 249],\n",
       "       [255, 125, 255],\n",
       "       [240,  54, 183],\n",
       "       [255, 139, 212],\n",
       "       [255,  85, 229],\n",
       "       [255,  64, 148],\n",
       "       [211,  29, 127],\n",
       "       [255,  50,  89],\n",
       "       [ 29,  21,  46],\n",
       "       [ 36,  24,  57],\n",
       "       [ 31,  21,  50],\n",
       "       [157,  16,  24],\n",
       "       [118,  18,  46],\n",
       "       [255,  46, 145],\n",
       "       [255, 127, 198],\n",
       "       [255,  58, 181],\n",
       "       [255, 138, 215],\n",
       "       [255,  41, 130],\n",
       "       [255, 118, 205],\n",
       "       [255,  26,  26],\n",
       "       [247,  92, 166],\n",
       "       [ 36,   8,  20],\n",
       "       [ 39,  26,  75],\n",
       "       [ 14,  11,  27],\n",
       "       [255,  38, 122],\n",
       "       [255,  87, 236],\n",
       "       [255,  45, 116],\n",
       "       [138,  12,  29],\n",
       "       [225,  26,  74],\n",
       "       [  9,   6,  19],\n",
       "       [ 64,  43,  97],\n",
       "       [234,  28,  81],\n",
       "       [116,  12,  43],\n",
       "       [131,  17,  69],\n",
       "       [255, 143, 255],\n",
       "       [  2,   0,   1],\n",
       "       [  1,   1,   0],\n",
       "       [255,  75, 180],\n",
       "       [ 30,  19,  58],\n",
       "       [255,  66, 154],\n",
       "       [ 14,  10,  32],\n",
       "       [129,  18,  60],\n",
       "       [255, 197, 255],\n",
       "       [255,  60, 179],\n",
       "       [  2,   0,   0],\n",
       "       [ 84,  11,  42],\n",
       "       [255,  80, 192],\n",
       "       [ 18,   0,  12],\n",
       "       [ 48,  33,  79],\n",
       "       [255,  77, 190],\n",
       "       [204,  54, 220],\n",
       "       [255, 241, 255],\n",
       "       [255,  48, 122],\n",
       "       [122,  16,  29],\n",
       "       [255,  77, 152],\n",
       "       [ 94,  12,  27],\n",
       "       [ 90,  13,  31],\n",
       "       [  3,   1,   1],\n",
       "       [  1,   1,   0],\n",
       "       [ 18,  16,  23],\n",
       "       [254,  37, 101],\n",
       "       [  0,   2,   0],\n",
       "       [225,  35,  77],\n",
       "       [240,  27,  18],\n",
       "       [255,  50, 126],\n",
       "       [ 33,  23,  52],\n",
       "       [147,  19,  42],\n",
       "       [237,  27,  65],\n",
       "       [255,  87, 242],\n",
       "       [124,  17,  64],\n",
       "       [144,  14,  29],\n",
       "       [158,  16,  37],\n",
       "       [125,  16,  74]])"
      ]
     },
     "execution_count": 207,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 2, 1, 0, 1, 1, 1, 0, 2, 1, 3, 2, 1, 2, 2, 1, 0, 3, 3, 1, 3, 1,\n",
       "       3, 2, 1, 2, 2, 0, 3, 2, 2, 1, 1, 1, 3, 0, 3, 2, 1, 0, 0, 2, 2, 2,\n",
       "       1, 2, 3, 0, 0, 0, 3, 2, 0, 0, 2, 2, 1, 2, 1, 3, 1, 2, 3, 2, 2, 0,\n",
       "       3, 1, 2, 2, 1, 2, 1, 2, 3, 2, 3, 0, 0, 3, 1, 3, 2, 0, 1, 2, 0, 0,\n",
       "       2, 2, 0, 1, 2, 3, 0, 1, 0, 1, 1, 0, 1, 3, 0, 0, 3, 3, 3, 1, 0, 1,\n",
       "       3, 2, 1, 0, 0, 2, 0, 0, 2, 1, 2, 2, 3, 0, 1, 1, 3, 2, 2, 0, 0, 3,\n",
       "       2, 0, 0, 2, 0, 0, 1, 3, 1, 1, 0, 3, 2, 1, 3, 2, 1, 2, 3, 0, 3, 0,\n",
       "       0, 0, 1, 0, 0, 1, 1, 3, 1, 3, 1, 0, 1, 3, 2, 0, 0, 2, 2, 2, 2, 2,\n",
       "       2, 1, 1, 1, 3, 3, 0, 1, 2, 1, 2, 1, 0, 3, 2, 2, 0, 0, 2, 2, 0, 1,\n",
       "       1, 3, 1, 1, 0, 0, 2, 1, 2, 1, 0, 0, 1, 3, 0, 1, 1, 3, 2, 0, 3, 2,\n",
       "       2, 2, 2, 2, 0, 1, 2, 1, 2, 2, 0, 1, 0, 2, 0, 3, 0, 3, 0, 2, 0, 0,\n",
       "       1, 2, 0, 2, 2, 2, 0, 0, 0, 2, 1, 1, 1, 1, 1, 1, 1, 2, 2, 3, 0, 0,\n",
       "       1, 0, 2, 1, 2, 0, 0, 1, 1, 3, 0, 3, 3, 1, 0, 1, 0, 1, 0, 2, 3, 3,\n",
       "       1, 3, 0, 1, 0, 0, 2, 2, 2, 1, 1, 3, 3, 0, 2, 3, 1, 2, 2, 0, 2, 2,\n",
       "       0, 1, 1, 1, 3], dtype=int64)"
      ]
     },
     "execution_count": 208,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = keras.Sequential([\n",
    "        keras.Input(shape=(3,)),\n",
    "        keras.layers.Dense(16),\n",
    "        keras.layers.Dense(32),\n",
    "        keras.layers.Dense(64),\n",
    "        keras.layers.Dense(128),\n",
    "        keras.layers.Dense(256),\n",
    "        keras.layers.Dense(4,activation='softmax')\n",
    "        ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer='adam',loss='sparse_categorical_crossentropy',metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_11\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense_66 (Dense)            (None, 16)                64        \n",
      "                                                                 \n",
      " dense_67 (Dense)            (None, 32)                544       \n",
      "                                                                 \n",
      " dense_68 (Dense)            (None, 64)                2112      \n",
      "                                                                 \n",
      " dense_69 (Dense)            (None, 128)               8320      \n",
      "                                                                 \n",
      " dense_70 (Dense)            (None, 256)               33024     \n",
      "                                                                 \n",
      " dense_71 (Dense)            (None, 4)                 1028      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 45,092\n",
      "Trainable params: 45,092\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/44\n",
      "157/157 [==============================] - 0s 2ms/step - loss: 0.4972 - accuracy: 0.7955 - val_loss: 0.4435 - val_accuracy: 0.8235\n",
      "Epoch 2/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4906 - accuracy: 0.7859 - val_loss: 0.3900 - val_accuracy: 0.8235\n",
      "Epoch 3/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.7259 - accuracy: 0.7604 - val_loss: 0.4624 - val_accuracy: 0.8235\n",
      "Epoch 4/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.5326 - accuracy: 0.7764 - val_loss: 0.4168 - val_accuracy: 0.8824\n",
      "Epoch 5/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4725 - accuracy: 0.7859 - val_loss: 0.5694 - val_accuracy: 0.7647\n",
      "Epoch 6/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.5016 - accuracy: 0.7732 - val_loss: 0.5100 - val_accuracy: 0.8824\n",
      "Epoch 7/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4892 - accuracy: 0.7764 - val_loss: 0.4831 - val_accuracy: 0.8235\n",
      "Epoch 8/44\n",
      "157/157 [==============================] - 0s 2ms/step - loss: 0.4864 - accuracy: 0.7668 - val_loss: 0.7048 - val_accuracy: 0.7647\n",
      "Epoch 9/44\n",
      "157/157 [==============================] - 0s 2ms/step - loss: 0.4745 - accuracy: 0.7891 - val_loss: 0.5024 - val_accuracy: 0.8824\n",
      "Epoch 10/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4806 - accuracy: 0.7955 - val_loss: 0.4075 - val_accuracy: 0.8235\n",
      "Epoch 11/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4748 - accuracy: 0.7955 - val_loss: 0.6760 - val_accuracy: 0.8824\n",
      "Epoch 12/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4828 - accuracy: 0.8083 - val_loss: 0.4744 - val_accuracy: 0.8235\n",
      "Epoch 13/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4690 - accuracy: 0.7764 - val_loss: 0.4949 - val_accuracy: 0.7647\n",
      "Epoch 14/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4779 - accuracy: 0.7732 - val_loss: 0.4304 - val_accuracy: 0.8235\n",
      "Epoch 15/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4700 - accuracy: 0.7955 - val_loss: 0.5027 - val_accuracy: 0.8235\n",
      "Epoch 16/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4861 - accuracy: 0.7827 - val_loss: 0.4661 - val_accuracy: 0.8235\n",
      "Epoch 17/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.5061 - accuracy: 0.7700 - val_loss: 0.4955 - val_accuracy: 0.8824\n",
      "Epoch 18/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4859 - accuracy: 0.7891 - val_loss: 0.4593 - val_accuracy: 0.8235\n",
      "Epoch 19/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4693 - accuracy: 0.7859 - val_loss: 0.4959 - val_accuracy: 0.8824\n",
      "Epoch 20/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4534 - accuracy: 0.8019 - val_loss: 0.4211 - val_accuracy: 0.8235\n",
      "Epoch 21/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.5040 - accuracy: 0.7732 - val_loss: 0.4765 - val_accuracy: 0.8235\n",
      "Epoch 22/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.5122 - accuracy: 0.7604 - val_loss: 0.4437 - val_accuracy: 0.8235\n",
      "Epoch 23/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4848 - accuracy: 0.7827 - val_loss: 0.5151 - val_accuracy: 0.8824\n",
      "Epoch 24/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4684 - accuracy: 0.7827 - val_loss: 0.5939 - val_accuracy: 0.8824\n",
      "Epoch 25/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4673 - accuracy: 0.8019 - val_loss: 0.4915 - val_accuracy: 0.8235\n",
      "Epoch 26/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4807 - accuracy: 0.7923 - val_loss: 0.4582 - val_accuracy: 0.8824\n",
      "Epoch 27/44\n",
      "157/157 [==============================] - 0s 2ms/step - loss: 0.5098 - accuracy: 0.8019 - val_loss: 0.4004 - val_accuracy: 0.8235\n",
      "Epoch 28/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.5394 - accuracy: 0.7700 - val_loss: 0.4805 - val_accuracy: 0.6471\n",
      "Epoch 29/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.5831 - accuracy: 0.7700 - val_loss: 0.4640 - val_accuracy: 0.7647\n",
      "Epoch 30/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4803 - accuracy: 0.7987 - val_loss: 0.4119 - val_accuracy: 0.8235\n",
      "Epoch 31/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.5055 - accuracy: 0.7987 - val_loss: 0.5217 - val_accuracy: 0.8824\n",
      "Epoch 32/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4827 - accuracy: 0.7923 - val_loss: 0.6559 - val_accuracy: 0.8235\n",
      "Epoch 33/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4865 - accuracy: 0.8051 - val_loss: 0.4128 - val_accuracy: 0.8824\n",
      "Epoch 34/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4667 - accuracy: 0.7987 - val_loss: 0.5355 - val_accuracy: 0.8235\n",
      "Epoch 35/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4873 - accuracy: 0.8019 - val_loss: 0.4065 - val_accuracy: 0.8235\n",
      "Epoch 36/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4850 - accuracy: 0.7891 - val_loss: 0.3790 - val_accuracy: 0.8824\n",
      "Epoch 37/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4785 - accuracy: 0.7796 - val_loss: 0.3792 - val_accuracy: 0.8824\n",
      "Epoch 38/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4827 - accuracy: 0.7891 - val_loss: 0.6061 - val_accuracy: 0.8824\n",
      "Epoch 39/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4891 - accuracy: 0.7923 - val_loss: 0.3933 - val_accuracy: 0.8235\n",
      "Epoch 40/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4828 - accuracy: 0.7923 - val_loss: 0.4701 - val_accuracy: 0.8235\n",
      "Epoch 41/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4770 - accuracy: 0.7732 - val_loss: 0.4443 - val_accuracy: 0.8235\n",
      "Epoch 42/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4878 - accuracy: 0.7987 - val_loss: 0.5077 - val_accuracy: 0.8235\n",
      "Epoch 43/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4883 - accuracy: 0.7987 - val_loss: 0.4527 - val_accuracy: 0.8824\n",
      "Epoch 44/44\n",
      "157/157 [==============================] - 0s 1ms/step - loss: 0.4625 - accuracy: 0.7827 - val_loss: 0.4676 - val_accuracy: 0.8824\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train,y_train,epochs=44,batch_size=2,validation_data=(X_test,y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAABchklEQVR4nO2dd3xUxfbAvycJIUBCSegkNBOaUkVAUEFQHwqCPpGiqMCzV/T3fKJPfVjw2cX2sAM2UFQUERFBiqKAVKUIG3roJPSWNr8/7l642Wy5u3s3G8h8P5/9JHvvzNxz596dM3PmzBlRSqHRaDQajScx0RZAo9FoNKUTrSA0Go1G4xWtIDQajUbjFa0gNBqNRuMVrSA0Go1G4xWtIDQajUbjFa0gNFFFRL4XkZucThtNRGSTiFwSgXLniMjNTper0fgiLtoCaE4/ROSw5WtF4ARQ4P5+m1LqE7tlKaUuj0Ta0oiIPAxcrpS6yON4dWA70E4ptTIqwmk0XtAjCE3QKKUSzQ+wBbjScuykchAR3QEpykdAZxFp5HF8IPDn6awc9LM+M9EKQuMYItJNRLJE5CER2QmMFZFqIjJVRPaIyD73/6mWPCfNJiIyRER+EZEX3Wk3isjlIaZtJCLzROSQiMwUkTdF5GMfctuR8SkRme8ub4a712+ev0FENotItoj821f9KKWygJ+AGzxO3QiMDyRHgLrvICK/ich+EdkhIm+ISLzl/Nki8qOI5IjILhF5xH08VkQeEZH17ntbIiJpItJQRJS14fdS//NF5BURyQFGishZIvKTux72isgnIlLVkj9NRL5y31+2W8bybplaWtLVFJFjIlLDzr1rIodWEBqnqQ0kAw2AWzHesbHu7/WBY8AbfvJ3BNYC1YHngfdFREJI+ymwCEgBRlK8UbZiR8brgKFATSAe+CeAiLQAxrjLr+u+nr9GfbxVFhFpCrQBJtiUwxcFwP0YdXE+0AO4032NJGAmMN0tYzowy53vAWAQcAVQGRgGHLV5zY7ABow6GQUI8F/3NZoDaRh1j4jEAlOBzUBDoB4wUSl1ApgIDLaUOwiYqZTaY1MOTaRQSumP/oT8ATYBl7j/7wbkAgl+0rcB9lm+zwFudv8/BMi0nKsIKKB2MGkxGtd8oKLl/MfAxzbvyZuMj1q+3wlMd///OEZDZ56r5K6DS3yUXRE4CHR2fx8FfBNsXdm4h+HAZPf/g4BlPtKtBfp6Od7QXZ9xfp7VlgAyXGVeF0Np7bGWZ0nXEdgKxLi/Lwb6R/vd1h+lJ6k1jrNHKXXc/CIiFYFXgJ5ANffhJBGJVUoVeMm/0/xHKXXUPSBI9HEtX2mrAzlKKWtPeCtGj7YYNmXcacly1CJTXXfZphxHRCTbh7ymnJOAG0XkN+B6jF58KHVlvYcmwMtAewwlFAcscZ9OA9b7yOrvXCC2Wr+ISE3gNeBCIAljRLTPcp3NSql8z0KUUgtF5AjQVUR2YIxwpoQok8ZBtIlJ4zSe4YH/D2gKdFRKVQZMDx5fZiMn2AEkuxtcE6/KwU04Mu6wlu2+ZkqAPOOB/sClGA3pVAfkGAP8BWS48z5iybcVOMtHPl/njrj/Wuuwtkcaz2f9X/exVm4ZBnvIUF98T2aPd6e/AfjC2snQRA+tIDSRJgnDlr5fRJKB/0T6gkqpzRhmipEiEi8i5wNXRkjGL4DeInKBe1L4SQL/rn4G9gPvYJinch2QIwnDdHVYRJoBd1jOTQVqi8hw96Rwkoh0dJ97D3hKRDLEoJWIpCjD/r8NGOyeyB6GbyVjleGwW/56wIOWc4swlOmzIlJJRBJEpIvl/EfA1RhK4sMg7lsTQbSC0ESa0UAFYC+wAGOitCS4HsPunQ08DXyGsV7DG6MJUUal1CrgLoxJ8R0YJpWsAHkURiPYgKKNYchyYEyaXwccAt7FuF/zeocwRitXYpjKXMDF7tMvA58DMzAUzPtuGQBuwWjks4GzgV8DyPAE0A44AHwHfGWRocB9/XQM1+gsYIDlfBawFGME8nMQ962JIOKeFNJozmhE5DPgL6VUxEcwmtAQkQ+A7UqpR6Mti8ZAKwjNGYmInAfkABuBy4CvgfOVUsuiKZfGOyLSEFgOtFVKbYyuNBoTbWLSnKnUxnDLPIzhWXOHVg6lExF5ClgJvKCVQ+lCjyA0Go1G4xU9gtBoNBqNV86ohXLVq1dXDRs2jLYYGo1Gc9qwZMmSvUopr3GvzigF0bBhQxYvXhxtMTQajea0QUQ2+zqnTUwajUaj8YpWEBqNRqPxilYQGo1Go/GKVhAajUaj8YpWEBqNRqPxilYQGo1Go/GKVhAajUaj8coZtQ6iNLFkCeTlQadOweXbsAHWroXLL3dOlqNH4fPP4aabwOfuzl4oLIRx42DQIKhQIWDyqLJsmXGfXboETmtl0yZYuRJ6946IWMXIzYXRo+HwYe/ny5eHu++GKlWCK3fyZOjQAerVC1vEsFm7FrZsgUsvDS7f7t0wZw707++cLHl5Rn0fOhR83rg4uPVWqO25TVIUWLIEvvnG9/nERPjXvyJw4Wjveerk59xzz1WlhS5dlDrvvODz3XWXUgkJSuXnOyfLhAlKgVKLFgWX77ffjHwffuicLJGia1elmjcPPt9ttylVrpxSubmOi+SV6dONOgWlRIp+zOPvvRdcmQcPGvkefDAyMgfLwIFKVaumVGFhcPlGjjTuY+dO52SZOdN3fQf6gFJPP+2cLOHQs6f/e6hdO/SygcXKR5uqTUwRYt062LMn+Hx79sDx45Dld8uZ4Ms0ZQoGM32w+aLBunWwfj0U+N252Xu+vDzY7HMtqbOYdblzpzFCs37y8qBcOXC5giszM7No2dFm3TrYtw+yfe7M7TsfBH//dsrMyipe34E+9eo5K0s4rFsHAwb4lnXHjshcVyuICHDggNEo5+QEn9fMY/7onSDUMs30TsoSCY4cMX4gubmwdWtweUv6Hl0uwxxQs2bxc3Fx0KjR6f2clApdnkjch8tlmEfr1Ak+b3p66ajT3FzDFJqeXvLX1goiApgv1cGDRq8wGMzG3MmeS6hlmulLSy/KF9YfcTCyHjt2SqGU1D1mZkJGhu+5oIyM0J/T+vVGbzKa7NljvPdQOt63zEyjYY0JoaUL5VlEgk2bjOeakVHy19YKIgJYX6p9+4LLaw7LnXwxQy3T+oMtzduGWO8rmHtcvz60fOHgcvn/oWdkGI1aMPVtyu60aTIUQn0W2dmnfitOPotA9e2PjAxj4txUeNHCrA+tIM4QrD3aYM1MpcXEZJoKypc3fiB79zonj9OY9xUfH9w9mmnLly8ZU0JeHmzc6N9UkJ5ueGMFY1M2n5P5fzQJtU4j8SwKCgyvwFBNM2a+0lKn2sR0hmDtAQWjIPLyTrnjRcLElJNjX569e425lK5dnZfHaVwuwxWxadPg5DTTXnRRydzf5s1GoxVoBGGVzQ4uV+l5Ti4XxMZC586hPYuuXZ0bsW7datjvwxlBWGWLFi4XVK4MNbzu2BBZtIKIAC4XJCUZ/wfjyWE23klJoXnk+CI7+5Q8dl92M525HiPaPxJ/uFxG7yo9PfhGKSUFzjvP6NkHO18ULHZMBcE2SgcPwq5dRsOakBD95+RyQcOG0Lx5cA29y2XME1x6qbFGZNcuZ2SB0BXEWWcVLSdamGayYNYwOYVWEBEgM9NodCC4EYSZ9rzzjJ6PU/bknJxT8gSrIC691OgRRnuY7Q9z4jcjwzAp2FWs5g8vI8PIE2lXV7NO/ZkK6tcPztXVfC5NmxoNWrSfk/VZHDhg3zTpchn3fs45p8oJFzv17Y+KFQ1X19JSp9FAKwiHMV1czRXUoSgIM68TPZfCQmPyr317o4dm92XPzDQUQ0aG0SOMdi/KF4cPG/Z6s1HKyzNW8drB/OGZDUik7zEz0xjJeXNxNYmNhcaNg3tOcOo+ovmclCqqdK3yBcKqWMCZ+8jMNBr5unVDLyPankzRdHEFrSAcx3yZzj3XaJCDMTGZaTt2LFpWOOzfb/xw69Y1emjBjCAaNDAmfqPd8PjDOoEXTEN/9KgxQktPLzlbs11TQTCNkpnurLOMfNF0dd2925hDC/ZZmIolPd145+LinHkWZpnhmGairSA2boyeiytoBeE4ZoPVpAlUqxbaCOLss43FPU4Mbc0yk5ODa+it7oGhuF6WFNYedDC9VtPFNSPD6NEnJUXelGA2WIEw69tOQ+9yGcq/UiUj34kT0XN1tT6LRo2MDpKd9y072+jIZGSEvljQG3br2x/p6YZF4MCB8OUJBWudRgOtIBzG2qNLTg5NQVSv7lyv3aogzIYnEKaLq1VBHDwYWuiQSGO1M9etayhWO/Vm/eGJRH6UlJdnmArs/NDT041FfHZcXa3PqaRMZb6wTgrHxxumSTvvm2cj6ESv3XRxDbdhDdZU5jThzqOEi1YQDuNyQVqa0VClpARvYoqNNVzanBramtdPSTHKtBMjx1wNa76U0W54/GG6uCYmBtfQe/7wIm1K2LQpsIurSTAmL2svOdpumaaLa4MGxvdQn4WZL5wR65YthlJ2SkFEs06rVDE6jdFAKwiHMZf2Q2gjiOTkUw1dMB45/so0ZbHb0Hu6B0a7F+UPTw8Pu6Mkl8v40VWteirfpk2Rc3UNpidot1E6eNCw+5vp69UzXF2j9ZwyMw3zULlyxndT6QZq6E0X18aNT+U7ciQ8V1enet6mTNGs02i5uIJWEI5jtd2HqiDAKCOU4HPeyjRlsdvQew75GzY0eoaldQThqSA2bID8fP/5PBVLerqhjDdtioiYQdmS09LsrQr3LDMmxjBtRrO36/ks7KzCz8w85RBh5jPLCxWnbPcVK0JqanTrNFrmJdAKwlH27zd+DOZLGYqJKSXF+N+poa15/WrV7E8cmqaChg2N7+XKlU5X10OHjLDZ1h9QeroxCgikWD1/eJE2JQSzGtZ0dbU70vO8j2g8J6snkkkwI9ZQ8gUqs2LF0KK4ehKtOs3NNdbmRGuCGrSCcBTPmCnJyUYjZtdsYR1BOBUHJifHMKPExhpxbuy4upqrYU1TAdg33ZQkVk8kEzsN/dGjsG2b93yRusdgXS7tNEq+FISTq/Dtsnu3sSYl2GdhXTthYrq6hvMsnHBxNYlW2O9ou7iCVhCO4mm7Nxt7uxFdrQqibl2jBxRuz8VapimbHdOF50tp155ckngLpWCnUfKmWGrUMFxdI9VTDHY1bHp64DUNmZnGvEPFikXzObkK3y7enoVpmvT3vmVnGy6k1nxxcfZGUP5wcvVxRkZ0XF2j7cEEWkE4istl9FjMGC6muciumclqYnLK9dJaJgRu6L2ZCsD4fuiQ0VMsLVhdik3q1DEaTH+NkrcfnkjkTAnBuLiaZGQYrq7bt/tO4+05RcvrxludxscbowF/svhqBMN59/PznXFxNYl2neoRxBlCZqYxoZWQYHw3e+52Jqpzc40hurW378TQ1nMEkZ5uzJX4UlrmalhvIwgoXWamzExDISQmnjpmR7H6+uFFyoy2caNh8gmmJ2inUfI0zVjzlfRzysw0ev7mvJVVnnCeRSgjVtPF1amed7TCfmdmGuZhawevpNEKwkE8f7DBKAjTDOVpDgrX1dWbiQl8v+y+vD+i7Q/uDW8NJARulDIzDZNSlSpFj6enR8bVNRSPmkCNkhnzy7PMunWjE9XV5TKcIOLiih4P1NBnZhqOE40aFc935IjhhBAsTq8+jlZUVyfnUUJFKwgH8WywgjExWRe0mZiurnaDz/kq17NMU1Zv+BryN2hQ+lxdfbkApqcbvXZfrq6+8mVkGMp440bn5TTLt4vp6uqrvn1tIhMTE53YWf6ehb9V+C5XURdXaz7zfCiygHMKokIF43lEo06jaV4CrSAcw1yhbP2RBDOCsK5XMAl3aFtQYJiTrGUGcnX1dHE1KVfOuRg5TnDokLGQytcIwl9UV38jD3D+Hk0X12BWw8bG+l/T4K8RLGmPM8/QLJ6ygP/7cPpZuFxGbKratYPP64uS9mQ6ccJ4f89oBSEiPUVkrYhkisgIL+eriMi3IrJCRFaJyFDLuU0i8qeILBeRxZGU0wm8DWsrVzYa41AVRLhmnQMHjB+vtUxz4tCficm6GtZKtPzBveHPjOCv3o4cMSZ+veWLVEiRUFfD+muUzOPWCXprvpJ0dd21q7iLq4m/ht6fYgl2XwwrZjQDJ00zJf3umy6u0fRggggqCBGJBd4ELgdaAINEpIVHsruA1Uqp1kA34CURsQ42L1ZKtVFKtY+UnE7hzTQTE2M0zqGamEyPnFBfTG9lmjL669H5eimdiJHjFP5cAP2NvEwXV2/5atQwlLrTDUGopgJzJODN1dXlKu7ias3nxCp8u/h7Fv5W4Zvb2nrLZ0Z1DdXE5HTPOyPDkHf/fmfL9UVp8GCCyI4gOgCZSqkNSqlcYCLQ1yONApJERIBEIAcIECShdJKZWdTF1cRuuA1vIwjTIyfUoa23MsG3q6u3RUue+Q4fLh2urv42cq9TxzAxeGtc/P3wwq1vb4Sz4UtGBhw/bizq8yTQc4KSM4n4G835W4UfqBEMxVRmurg63fMuaU+maIf5NomkgqgHWPswWe5jVt4AmgPbgT+B+5RSZn9JATNEZImI3OrrIiJyq4gsFpHFe6IYj9rlMiayTBdXk2AURFzcqb2jTcIZ2vpTEAcOFB/ZeFsN65kPSoeZyeU6tQ+CJ/5cXf0pFnDelBDOalh/jZK/hWAlHX3X5TLeXTOKqye+GvpAjWAorq6bNxtKIhIjCCjZOq1atfhvt6SJpILwZgH0fNR/A5YDdYE2wBsiUtl9rotSqh2GieouEbnI20WUUu8opdorpdrXsBPoJkL4Ms2kpNhTENnZpyK5WjGjugYKPuerTFMGzzJNma0EWrlZmsJ++zOFge+RgMvl3cXVJCPD6PHn5joiZlg9QV+Nkuni6uv+g9kXwwl8ubia+DJNulyGGdbTIcKa78gRe/timESq533WWcZvsyTrNJR5K6eJpILIAtIs31MxRgpWhgJfKYNMYCPQDEAptd39dzcwGcNkVWrx1aOzOwfhuV7BxPTICcWe7G8EAb4VhK8fV8OG4cfIcYpAoRTMNSSeijWQfTo93ejxOxXVNZxwCWlpRvysYJ+T6epakuaQQM/C2yp8M+aXp4urNZ9Zvl0iFZ4iIcFYBFta6rSkiKSC+B3IEJFG7onngcAUjzRbgB4AIlILaApsEJFKIpLkPl4JuAxYGUFZw8J0cfWlIIIZQXgSztA2J8fogZh7HpiYrq6eL7uv1bAm4UwcOsnBg75dXE0yMgzlsHlz0eN2GjNw7h5drtA3fDH3SPD2nCCwoiuJ5xRo3gp8N/SReBaRcHG1ylMSdWq6uEbbgwkiqCCUUvnA3cAPwBrgc6XUKhG5XURudyd7CugsIn8Cs4CHlFJ7gVrALyKyAlgEfKeUmh4pWcPFX68lJcVeRNecHO9L6sMx62Rnn4rkasVXjByzR+fLVGDKE20FEWgewXrO2iiZLq7+8jmtIMJ1ufTWKJnfzc1sfOVzYsOpQOzcadSrnWdhvQ9fMb+spKUF7+oaCRdXk5J69zdsiH4UVxM/TUH4KKWmAdM8jr1l+X87xujAM98GoHUkZXMSfz0662K5WrV8l5GTA6293LHpkRPK0NaX2cqU1VvDE+ilzMiAn382fuDRso/a6UFbG/q//c1+vurVDVdXp0wJLhd07Bh6/owMmDHDaDBiYk6VmZrq3cXVxIzqunWr7xGhE9ipU2+urua2tv7ymVFdgzUxefsdOUFGhtHp2rfP2F8lUpQWDybQK6kdweUyGktvPTq7q6l9mZjCieoaSEFYPUT8LVryzHf4cHjbQYaLWRfeFomZ1K5dXLHa+eE5GdXViQ1f0tOLu7rafU4Q+R6vHX99b6vw7TaCwTyL/HzDayxSDWtJuQ9Hah4lFLSCcABfLq5wymzkT0GcOGEM031FbQxVQXjGYfIs88CBU9tBmqthA72UpcGTyZ+Lq4k3xWr3h+eUgnBiNay3hj6QacZXvkhgurjWr+8/XajPwpxst+Pqarq4RqphLal33+UyRijRjOJqohWEA/jr0Zk9eH+eTN4iuVrJyPAffM4XgUYQcOplt7tys6QXYXnDroeHZ0PvckHNmoYJyR/p6UZjE66rqxOrYT2fk+e2tr6oU8dwdY30c8rMNEbO/uatoPjiTDPml2cUV2/5jh615+oa6dXHpqtrSdRpaTAvgVYQjuDPdm/HxGQqD3+Nub/gc76woyDMl93ukN/cDjLaIwi7CsKqWINRLIWF4Ud1daLBSk01XF2DfU4lFdU1mGdhXYWfmVl8W1tf+czr2JHFmsdpEhJKJqqrnRFiSaEVRJjk5BgfXw/UjonJPOfPHATBvZhmJFdfZTZsWDSqa6DVsCbRdnU9eNBoZOz8gNLTi7q62v3hOWWeycw0XFzDMRXExBSN6hqMfTrSbpnmvJXdZwFF7yOUfP7IzDQ2j/LnDBIukVa6x4+XjiiuJlpBhEmgHl3lysZQ2p+JydeCNpNQzDr79xeP5GolPr5ojJxAq2E95YmWiSkYDw9rQ3/4sGGmsJPPqbg7Tq2GtTb0diboTcxV+JFydTVdXIN9FnbWTpjUr2+8q3aehal0IuldF+l3f+NGo360gjhDCDSsFQm8WC6Qicn0yAmm5xJI6UDRlz0Yu6enPbkkCcaMYFWsZhRXO/mqVzd6/uH2FO02goHIyDDkLyw07iUtzZhfsJMvFNOkXYJ5FtZV+Hv2eN/W1huxscYch10TU6QbVquraySItJksWLSCCBPTxdXfZFsgBRHIxBSKq6uvOExWrDFygrF7mjFyQtkOMlyC6UHXqmWYHFyu4Ewz4bgWmzi5GjY93SgvKyu45xRpT6Zg6tRcoR/sszDTBbqHvLzQo+YGQ6Q9mUqTiytoBRE2mZnGMNibi6tJoHhMZiTXxETfaYId2todQRw8CH/+ad9UYOaD6JiZMjN974PgibWhD7ZnFq4pIZwort5kgVP3YbfMSIeozsw0JpkDubiamCPPUJ+FvxFrpKK4epMFIlunycnRj+JqohVEmNj5wQaK6OorkqsVX8HnfGFXQQB8/33R74EoKR97bwRrRjAbl8xMY0ThGU7dX75wXF2dNBWYZfz+u++YX96oWze8DacC4XLZc3E1MZ+F6eJqd4V3RgYcO2aESfEni5k2kjRuHNmoriVhJgsGrSDCxI4Xhx0TU6Aeg6dHTiACzWuYZcIpBWF3WFu/fvRcXe16zZikpxu9+TVrgs8XjqurnXhRdqlXzxihTp8eXJlOmMr8Eaw7Znq64Swwf749F1drPvDfa3eyvv1hurpGcgRRWsxLoBVEWJguroE0vh0TUyBXyGCHtr4iuVpp1Mjoyc2fb8/F1SSUGDlOYLq4BjuCyM83et/B5oPQG1eXy6h7J1bDmq6u8+cXlc0OkQr7bTc0ixUz7fz5zj8LlyvyLq5WeSKhdEubiytoBREWdoe1KSlGz8mXucJXHCYrwTZYOTneI7laMbeDzM8PzlRgylPSI4hQzAhm2mDt004oCCdcXK3y5Od739Y2UL5QN5zyx44dxgrnkngWaWmGq2sgBeFkffsjUu/+hg2ly8UVtIIIC7vDWrPx9+UaZ8fEZPXIsYMdpQOnZA92WBtMjBynCMWMYE0bTL6UFMPVNdTet9OmArOs1FT/DhGeRMrVNRRvG3MVfrD5TFfXQCamkjLNpKefsh44SUmZyYJBK4gwcLlOberij0DxmOyYmEx7cjAmJjvmDbO3EmyvJSOj5F1dg3FxNTEVKwR3j+FEdTVdXJ3sCYb6nCLlyRRKSGpzFX6w+cz0vp5FXl5ko7h6kwWcr9PStgYCIrwfxOlCqKGrV60yJmzLl/efzl+4DTOSq53efkYGLF9uT7aSUBBgvNR16gSX9/hxI5JssKxcGXgfBE/Mhn7ZsuB7ZhkZhr082PcjM9P5DV+ceE6XFdt5JXRcruBcXK3yhOKpk5EBM2caHRJPM9KmTcZq8ZJWEEuW2J+3s8Offxq/2UjuNREsWkFgjACOHg0tr7kZjT/8Beyz445qkp4OkycbPaZAHiB23SGbNSv61y7WBUMXXRRc3pYtQ+999egRfJ5mzYzJbbsuriZNm8KECaFvX9m0aWj5/JUV7HOqU8dQqOZKcqcwXVz9zXF5o1kzmDUr+Ia1aVPD1dVfZ8TJ+vaHGZLmzjuNj5N07uxseeGiFQQwenTok3jduwdO48/EFGgVtRVzonLz5sC9YbsjiEsugW+/tXcfVurXN5RUsA19To6Rp39/6NYtuLwAF18cfJ7nnju170Uw3H230SCFEsuoalVo3z74fL6oW9foQXfqFFw+M9RLKCM2f4QaknrECOjXz76Lq8l11xnKyJejR+XK4e3cFwwJCTBtWmS8wy680Pkyw0ErCOCWWyJbvj8Tk531CiZW26c/BWFGcrVTZkwM9O4dOJ0npqtrsDZ6M/3110OfPsFfNxTS0oxPsKSkwK23Oi9PqIQyegJjDubQIefkMF1cQ5GnRg3jEyyJifCPfwSfL1JceqnxOdPRk9QlQFKS0ftxwsQEgRvlQBsQOUUoi7BK40TcmU5iouFm7RTbtxsm2dLkbaOJDFpBlADmMD9cE5NdV9dgygwHOzFyPLET3FDjLE4riFA8mDSnJ1pBlBC+4jEFY2IyPXIC2T6DGZWEQ0aG/e0gTewEN9Q4S1KSswpCjwLLDlpBlBC+4jHZieRqxY5ZJxilEw6hhD4ubcHIygJOjyBcLmNlcyjzOprTC60gSgh/JqaUFPshAjIyjEVBeXm+05SkiQnsK4hgdhLTOIfTk9SZmaG5uGpOP7SCKCH8mZiC6elnZBheSv6iupaUiSlYV9ecHMO7Sk9uliyRGEFoJV820AqihPBnYgqmIbdj1snONkYkVaoEJ2OwBLMdJGjbdbRISjJW6xcWhl+Wue2pVvJlA60gSojkZO8RXe0uaDOxY9bJyTGW65eECSCYeEVaQUSHxETDvHfsWPhl7dhhlKOfYdlAK4gSwtdiuWBNTDVrGj1Cf2adYEcl4RCMq2tmprEwT7u4liymA4QT8xBayZcttIIoIXzFYwq2MbezS1iwSidUVuxcQcPG+QG3gzRxuewFN9Q4i6kgnJiHCCXMt+b0RSuIEsJbPKbjx411BMF6GwUy6wRrtgqF6ZnTafN2G/4q/BawZ2bSk5vRwUkFkZmpXVzLElpBlBDeTEyhehtlZBghjn25ukbaxHTwxEFu+dYIYLVGTQECezKZLq6651nymFFsnRpBnHWWdnEtK2gFUUJ4MzGFqiDS0w1X102bvJ+PtInpwRkPsv3QdrqkdWHRoS+Jj1e2Fu8dOKBHENHAaRNTpJT8X3v/YuSckRw64eCiDU1YaAVRQngzMYW6oM2fJ1N+vtEQR8rENGvDLN5Z+g4PdHqAB85/gKMFh6iTdiyggtCTm9HDqUnqwkJjX4lIPMPcglz6T+rPE3OfoP277flz15/OX0QTNBFVECLSU0TWikimiIzwcr6KiHwrIitEZJWIDLWb93QjKckIqWEdQYQaEsPflof794dWph0O5x7m5m9vpklKE568+Em6NeyGIJSvtTWgiak07rdbVnBqBLF9e+RcXEfNG8Wfu//kiW5PcPDEQTq814Gxy8Y6fyFNUERMQYhILPAmcDnQAhgkIi08kt0FrFZKtQa6AS+JSLzNvKcVZkRXJ0xMNWoYCsdbrz2ScZhGzBzB5v2b+aDPB1QoV4HkCsmcW/dcDicuP7nNpi/s7t9dmvhs5Wec//755Bb42KWmBNl/fD9dPujC+OXjg8q379g+hn7fDwhfQUTKg2n5zuU888sz3NDqBh7v+jjLb1tO57TODJsyjKHfDOVonvftHgtVIXM3zWXI10No+kZTVu1e5axgmoiOIDoAmUqpDUqpXGAi0NcjjQKSRESARCAHyLeZ97TDMx5TqCYmM6qrNwURqThM8zbP483f3+TejvfSpX6Xk8d7NOrBzvifA7q6ulzGNpPx8c7KFSmO5R3jgRkPsCBrAT9v/jna4vDe0vf4deuv/GPKP/je9b2tPCfyT3D1Z1czZ7vhaRaugohEmO+8gjyGfjOUlAopjO45GoBaibWYMXgGj130GOOXj6fjex35a+9fJ/Ns3r+ZJ+c+ScbrGXQb342v1nzFzsM7GfrNUPILQ9waUuOVSO4oVw/YavmeBXhuCvgGMAXYDiQBA5RShSJiJy8AInIrcCtA/WB3UC9hPOMxZWcbsYwqVQq+rIwMWLy4+PFIxGE6mneUYd8Mo3G1xozqPqrIuUsaX8Jz1Z4FjAYkNdV7GadbeIY3Fr3B9kPbiZEYvnN9R4/GIW7n5gD5hfm8vuh1zk89n+P5x7l20rXMHTKXc+ue6zNPoSpk6DdDmbt5Lh3qd2RRTB579xUAocdZd7mMNSxpacbI5I9df/hM27BqQxpUDbzx9HPzn2P5zuV81f8rkiucemljY2J58uIn6ZLWhcGTB3Peu+fxYOcH+XnLz8zaMAuFonuj7jzR7Qn+3vzvTFk7hUFfDuKV317hwS4PhnyP0WTPkT2s3rPa5/n05HTqVa5XghJFVkF4i0/qud72b8ByoDtwFvCjiPxsM69xUKl3gHcA2rdvH8TWNSVPcjJstag90x3VbiRXK+np8MUXhqurdX/fSJiYHv3pUdbvW8/sm2ZTKb6oNuuS1oX4GlvJxWhAvO0zbbq4Xn+9czJFkgPHD/Ds/Ge5PP1yClQB37m+4+W/vRw1eSavmcyWA1t4redrdKjXgU7vd6LXp71YcPMCGlZt6DXPI7MeYcLKCfy3x385P/V8usUfZu2OfUDoNj6XyzARxsTADZNv4DvXdz7TxsfG82rPV7nt3NsQHy/4yt0reXLukww4ewBXN7/aa5q/pf+NZbctY+AXA/nPnP/QsGpD/tP1P9zU5qYi9z7g7AF8tuozHpv9GH2a9qFp9aYh32e06DuxL79l/ebzfL2kerjucVGhXIUSkymSCiILsC6nScUYKVgZCjyrlFJApohsBJrZzHvakZwMy5ef+h7OgjYzquvGjdCkSdEywTkT069bf2X0gtHc0f4OujXsVux8hXIV6NKyPnPicnG5vNuP9u49vVxcX/z1RXKO5TCq+yh+3vIz902/j8ycTNKTozMEGr1wNI2rNaZ3k97ExsTy/fXf0+WDLlz+yeXMHza/SM8bYMzvY3hu/nPc0f4OHuryELkFuUj5PWzYtZtwFERmpvEMD+ce5scNP3J9y+v5R9viG0UXqkJe/O1F7vjuDuZtnsfbvd8mqXxSkTT5hfkM+2YYVROq8vrlr/u9bmrlVOYMmcP6nPVkpGQQI8Ut4yLCmF5jaPFmC4ZNGca8IfOIjTl9Fmts2r+J37J+454O93B1s+LKcv2+9dzy7S28+fub/LPzP0tMrkgqiN+BDBFpBGwDBgLXeaTZAvQAfhaRWkBTYAOw30ZeRyhUhYyYOYKrml1F57TOkbjESbyZmELt6VtdXT0VRDCRXHcc2sHri173ORH47bpvqV+lPs9d8pzPMi4562JmV81k1V+N8WbCCMaDadKqSdSrXC/iz8IXuw7v4pUFrzDg7AG0rdOWyuUrc9/0+/hu3Xfc1+m+Epdn0bZF/Lr1V17t+erJBq9FjRZ8PeBrLvv4Mq6aeBUzbphBQpxR71PWTuHu7++md5PevHb5a4gI5ePKU7FSAVl794cshxnF9W9/M1ydcwtyGdZ2GBc3uthr+osbXcyzvzzLY7MfY+mOpUy6dhIta7U8ef7l317m9+2/81m/z6hRqUbA68fFxAUcFdROrM2rPV/lxq9v5PVFrzO80/Cg7jGafLn6SwCGdxpO42rFlfjFjS5m0upJ/PeX/3JLu1uokhDhUM0mSqmIfYArgHXAeuDf7mO3A7e7/68LzAD+BFYCg/3lDfQ599xzVbDkHM1R6a+lq5TnUtTavWuDzh8MTz+tFCh1/LjxvVUrpfr0Ca2sXbuMskaPLnr8zjuVSk62V8asDbNUrRdqqdgnYlWV/1bx+qn3Uj01e+Nsv+UszFqoaPKNSkvf5/X8+PGGrH/95V+enKM5qvxT5VWN52uo3Yd327sJh7ln2j0q9olYtW7vupPHmr7eVF364aVRkWfQF4NU5f9WVgePHyx2bsKfExQjUf0n9VcFhQVqYdZCVeHpCuq8d85Th08cLpI2rfl2Rfp3atvBbSHJsWWL8QzHjFHqlim3qKRnktSJ/BMB883eOFvVfrG2qvB0BTV22VillFJr9qxR5Z8qr66eeLUqLCwMSR5fFBYWql6f9FIVnq6gXNkuR8uOJJ3e66Tavd3Ob5rF2xYrRqIe++kxR68NLFa+2nBfJ07HTygKQimlXNkuVf356qrxq43VrsO7QirDDmPGGDW+fbvxPTVVqaFDQyursFCpypWVuuuuoscHDlQqI8N/3oLCAvXknCdVzBMxqtkbzdTKXStDE8JNfkG+ir/gdRUbf0IVFBQ//+ijSsXEKHUiQHvyv0X/U4xExTwRowZ+MTAsmUJh476NqtyT5dQtU24pcvyB6Q+o+Kfi1aETh0pUnq0Htqq4J+PUA9Mf8JnmuV+eU4xEDf16qKrxfA3VaHQjtfPQzmLp2nc5qKg/T324/MOQZPnpJ+Pd/fHHQlXvpXrqms+usZ13x6Ed6uJxF5+U8/z3zlfJzyWrHYd2hCRLILIOZKnK/62sLhp7kSoo9PJCljK27N+iGIl6Zt4zAdP2+7yfqjSqkqPtlD8FoVdSY3gHTB00lR2HdnDlhCs5knskItfxDLcRjonJl6troDhMe47s4YpPruDxOY8z8JyB/H7L75xd8+zQhHATGxNL86ZxFOTGe3V1zcy05+I6bsU4WtVqxX+6/oeJKyfy9V9fhyVXsIycM5IYieHxro8XOd6rSS9yC3KZuWFmicrz5qI3KVSF3NPxHp9pHuz8IHe2v5Oxy8dSoAr4/vrvqZVYq1i62smJxOZXYdbGWSHJYr5neVVXs+3QNnpl9LKdt3ZibX684Uceu+gxxi0fx29Zv/Fqz1epnVg7JFkCUa9yPV6+7GXmbZ7HW4vfcrTsnYd38sL8F2j3djsajG7g8zPm9zG2y/xyjWFe6teiX8C0T138FMfyj/HMz8+EfA/BoBWEm46pHZlwzQQWb1/MdV9dR0FhgePXsCqIY8eMTzjeRunpxVdT+1M687fMp+3bbZmzaQ5v936bj6/+mMT4xNAFsHBhmzoAzF1aXEPYieK6es9qFm1bxNA2Q3n4godpXas1t0+9nZxjXrbhiwCr96zmoz8+4u4Od5Nauaiv7gX1LyApPonv1vn22nGaI7lHeHvJ21zd7GqfnkpgTM6+dvlrPNP9GWYMnuHTTl85SahQWIOZG2aaJtygMF1cFx/+GoDLMy4PKr/ptjrjhhm8dNlLXN8ysi5tw9oO47KzLuNfP/6LTfs3hVVWbkEuk9dMps+EPqS+nMq/Zv6L8nHl6d6ou9dPpXKVGDl3JMfzj9sqf9LqSbSu1ZqMlMBeHM2qN2Nom6GMWTyGzfv97DvsFL6GFqfjJ1QTk5U3Fr6hGIm6c+qdjttHly41humTJyu1bZvx/1tvhV6eN9NNo0ZKDR5cNF1hYaF6Yf4LKvaJWHXWq2eppduXhn5RH8xc6lKg1PWPzPW4tndTmCcPznhQxT0Zd3LovHT7UhX3ZJy6cfKNYct2NPdowCH51ROvVknPJKk9R/Z4PX/NZ9eoui/Vdfyd8MWY38coRqJ+3vyzI+XdeqtSlVOOKEai/toTYDLIC1ddpVSLFkqd/975qv077R2RKdJs3r9ZJT6TqC758JKQntuq3avU8O+Hq+rPV1eMRNV5sY566MeH1Jo9a/zmm7l+pmIk6oOlHwS8RtaBLMVI1NNzn7Yt15b9W1T5p8qroV+HaJ/2AG1iss9dHe7iwc4P8r/F/+OFX19wtGzrCMKJ9QoZGYZ3ycaNp455MzFNXDmRB398kKuaXcWSW5fQtk7b0C/qg26tzoK44yxeub/I8b174eBB/x5M+YX5fPTHR/TK6EXNSjUBaFunLSO6jODDFR+G3XMf+s1Qar9Ym8s+uowJf07gWF7RvTcXbVvE5L8m88/O/6R6xepey+iV0Yvth7azfOfysGSxQ6EqZPSC0bSv254uaV0CZ7BBYiLkHzc8nUIxlblc0KDxCRZkLQjKvBRN6lepzwuXvsDMDTMZ9fOowBksfO/6nlZjWvHm72/StUFXvrvuO7bcv4VnL3mWZtWb+c3bvVF3WtZsyeiFowOO1oIxL5mkVUnjzvPuZPyK8azZs8Z2vlAIqCBEpLeIF8fjM5hnL3mWgecM5KGZDzHhzwmOlWtVEE6seDYbXdPMZEZy9Szz3aXvcla1s5h07aSIucfFxgpVamezYX0shepUUCY7UVx/yPzBCJXQZmiR449e9Chn1zib26bexoHjB0KSa+O+jUxaPYkL6l/Auux1XPfVddR5qQ53TL2DRdsWoZTikVmPUKNiDe7vdL/PckyTir/FYU4xPXM6a7PXMrzjcJ+LzIIlMRGOHomhQeVGQc9DmFFcSV6PQp02CgLgtnNvY3CrwTw2+zE+WvGRrTxLdyzl2knX0rJWS7IeyOKL/l9wRcYVxMXYWxUgIgzvNJw/dv3B7E2z/ab9YvUXnFPznKAX9j18wcNULFeRx2Y/FlS+YLHT8A8EXCLyvIg0j6g0pYQYiWFc33F0bdCVId8McSwWT2Kisep59eadzFy1FAhvQZtn2O99+yhW5qb9m5i9aTZD2gxxrLHxReP0QvJ2NygSgsGOghi3Yhw1KtbgiowrihwvH1eesX3HsuPwDv5vxv+FJNMbi94gRmL49JpP2XDfBmbdOIsrm17JuBXj6PheRzJez2DWxlk8cuEjxRZzWamdWJv2dduXiIIYvWA0dZPqcu3Z1zpWprlpUNe6l/PTxp+CmmPbts3Y/XBPwq/UqlTLb4iP0oaI8H6f9+neqDvDpgxj1gb/ynHT/k30+rQXKRVTmHbdtJMj2mC5ruV11KhYg9ELRvtMs+PQDn7Z8gvXtgj+OdeoVIP/O///+HLNlyze7iXmjkMEVBBKqcFAW4z1CGNF5DcRuVVEfP+azgDKx5Vn8oDJ1Emsw+NzHg+cwQYiULlqHh8vnMaoHwzvinBGENWrGwvizEbY26hk/PLxCMKNrW8M/UI26dCyGuw7ixmZp0wYmZlGaIaGDb3nyT6azZS1UxjcajDlYssVO39evfP45/n/5P1l7/Pj+h+DkufQiUO8t+w9rm1xLamVU4mRGLo36s5HV3/Ezv/bydu936ZGpRq0qd2G29vfHrC8Xhm9WJi1kD1H9gQlRzCs3L2SHzf8yF3n3UV8rHORDc2Q3+fXupQDJw6wZMcS23nN92tN4RQuz7jc60rm0kx8bDxf9f+KZtWb8ffP/+5zr4l9x/ZxxSdXcDz/ON9f/z11kuqEfM2EuATuaH8HU9dNxZXtfbOUr9Z8hUIFZV6y8sD5D5BSIYVHZj0SspyBsPWklVIHgS8xoqrWAa4GloqIb/+7M4BqFapxU+ubmLd5HrsO7wq7vK0HtnIgZiOxJ2qSWGAEFkysEnooaZGinkye8xqFqpDxK8bTvVF36leJfCDDNi0SIb8C05auOHnM5TKUgy8X1wkrJ5BbkMuQNkN8lvvExU/QNKUpN397c1C7jY1dPpaDJw56XVFbJaEKt557K7/94zeW3bbs5Epkf/TK6IVCMT1zum0ZguXVBa+SEJfAbefe5mi5poJoU+1CgIA9aSumgjiSuPy0Mi9ZqZJQhWnXTSMpPonLP7mcrINZRc6fyD/BVZ9dxfp96/l6wNe0qBH+7gJ3nHcH5WLL8drC17ye/2LNF7So0SLka1UuX5lHLnyEHzf8yOyN/k1ZoWJnDuJKEZkM/ASUAzoopS4HWgMlFxQkSvRr0Y9CVRi2T/7+4/u54tMrUBX20jKpG5fWHQCxJ3ht6X/DKte6FsIzDtPPm39m4/6NxWz7kcI0Iy1YkX1yD4VALq5jl4+lXZ12tKrVymeahLgEPuj7AVsPbOWhmQ/ZkqWgsIBXF75K57TOdKjXwfY9+OPcuudSq1KtsMxMWQez6PheR8753zleP+NXjOfGVjeSUtHZeO2mgihfmEKrWq2YudH+RHVmJsTF5xFbdSeXNr7UUblKkrQqaUy7fhoHTxzkik+uODmvVagKuelroyM4/qrxdG3Y1ZHr1U6szaBzBjF2+Vj2H99f5Nyuw7uYt3ke/ZqHNnowufO8O0mtnMrDsx4OyX05EHZGENcCryilWimlXlBK7QZQSh0FhjkuUSnjnJrn0DSlKZNWTwq5jNyCXP7+2d9Zu3ctHdIzyD2cSLLKICHpKKN+edpv2ORAZGTA5s2Qm1vcxDR2+Vgql6/sM1Km05iK4MTuVBZkLUAp/2G+/9j1B0t3LGVI6yEBy+6c1pnhnYYzZvEYW72lqeumsmHfBr8Tz8ESIzFcnnE5P6z/IeR9B95Z8g6/b/udptWb0qx6s2Kffi368ciFzpsMrLvK9WjUg/lb5hfz5vKFywWxKZu4qOEFJRcDKEK0qtWKL/t/yZq9a7jm82vILchlxMwRfLbqM5675DkGnjPQ0esN7zScI3lHeG/pe0WOT/5rMoWqMGTzkklCXAJPXfwUbWu3tb3uIih8+b+aH6ARkGD5XgFoGChfND5OrIPwxr9n/VvFPhEbUnygwsJCNfirwYqRqA+Xf6iGDFEqLU2pv/9dqabN81TNF2qqc98+V+UV5IUkmzXO0csvG//n5Ch16MQhVWlUpWJhIyJJQYFS5csXKjq/oB776TGf8aJM7p9+vyr3ZDmfaw88OZJ7RJ316lmq8auNi8Ua8qTr2K6qwSsNQq5XX0xaNUkxEjV309zAiT0oKCxQ9V+pr/720d8clckOCxYYz2LaNKW+W/edYiTqx/U/2sqb0eyEoulk9eL8FyMsZckxfvl4xUhU27faRmzdk0m3cd1U/VfqF3kXu4/vrpq+3rTE1tX4gzDXQUwCrJtJFriPlRn6tehHgSoIycz06E+P8vEfHzOq+yhuaH0DKSnGXEFODtSsHsebV7zJkh1LeGF+aGsurJ5MOTnGhHCVKkZU1CN5R/za9p0mJgbOOkuodvQ8Zm2c5deDKa8gj4//+Jg+Tfv4XHvgScVyFXm/z/ts2LeBf//0b5/plu1YxtzNc7mnwz22XRPtcmnjS4mLiQtpbcbsjbPZcmBLiZn8rFhHEBc1uIi4mDhb8xCFhbBxQwykuOjV5PScf/DGja1v5KmLn2LZzmX0bdr3ZOTbSDC843C2HNjC5DWTASPczZxNc+jXol/EPQvDxY6CiFPGtp8AuP8/TTaOdIbWtVqTnpzOF2u+CCrfO0ve4ZlfnuHWdrfy8AUPA4b55+hRY3vO5GRD+fRr0Y+Rc0f63U3KF54Kolo1o6Eet2IcTVKacH7q+UGXGQ4ZGRC3vxkLsxby55pjRWS08n3m9+w5uidoBda1YVfuOu8uXlv4GvO3zPeaZvTC0STGJ3Jzu5uDFT8gVRKqcGH9C0Oahxi3YhxVylehb7OS3z3XVBCHDkFifCKdUjvZmofIyoL83Diqp+2nacrptwmPP/594b+ZN2QeE/tNjOjeEb2b9OasamfxyoJXgFPmpVDcW0saOwpij4j0Mb+ISF9gb+REKn2ICP2a92PWhllkH80OnAGY5prGHd/dQa+MXrzZ682TPQVzfmDjxlP/v3nFmyTFJzHsm2E+/dOVUvy69VcmrSo6eEtJMUYMmZmn4jCtz1nPvM3zGNI68msfPElPhwM7qlNQUMjUBWuJjVVeXVzHLh9LrUq16JneM+hrPHvJszSo2oBhU4YVs6PvPLyTCX9OYGiboRGzl/fK6MWqPauCioVz8MRBvlz9JYPOGWTLY8pprCMIgEsaXcKS7UvYd2yf33wr15wA4ILWtUt9bzdYRIQLG1wY8ecRGxPLvR3v5bes31iYtZAvVn9BenK6X8eM0oIdBXE78IiIbBGRrcBDgLM+eKcB1559LQWqgG/WfhMw7dG8o9w85WZa1mzJxH4Ti5g5TA+jvLxT/9esVJPXL3+dhdsWFltYs/3Qdp795Vmav9mcLh90of8X/Xnlt1dOnrdGdTV3qBu/YjwxEsMNrW8I+76DJSMDck/EUpeOfLdgLTHVtvDywufYfuhUEL89R/Ywdd1Ubmh1Q0gmoMT4RN678j3WZa/j8dlF16j87/f/kV+Yz70d7w37XnxhmlqCGUV8vupzjuUfY2jbkjcvwamFcqaC6NG4BwoVcKXv94vWAXDNBaW/MSvNDG0zlMrlK/OfOf/hp40/0a956Tcvgb2FcuuVUp2AFkALpVRnpVRmoHxnGm1rt6VR1UZ8sTqwmemNRW+w4/AOXr/89WLRUq2L2Kz/DzxnIH2b9uXR2Y+ycvdKvlj9Bb0+7UXaK2k8POthalSqwft93qdfi37834z/KyKHVUFUS1aMXzGeSxtfWiwqaUlgmpPePn8W9Qt6UKn2DkbMGkHaK2n0+rQXX6z+grHLx5JfmB/W/EiPxj24td2tvLzgZRZmLQTgWN4xxiwew5VNr4zo9qBNU5rSuFrjoBTEuOXjaF69OefVPS9icvkjPt5YxW8qiI71OpIYnxhwHuLnZbsg7hh/79ixBKQ8c0kqn8TNbW/mh/U/UKAKHF0lH0lsLZQTkV7AncD9IvK4iDiztPg0QkTo16IfMzfM9Dss3398P8/+Ymx4f2GDC4ud96UgzD11E+ISaDmmJddOupYVO1cwossI1t29jp+H/sywtsP46OqP6JzWmcFfDeaXLb8AhllnyxbYsQPy4ney5cCWEp2ctmK6tG7bXJF926ozuFsn1t69loe6PMSKnSu4dtK1PDTzIc6re17Y+1A8f+nz1E2qy7ApwziRf4JP//yUvUf3Oura6g0RoVdGL37a+FMx/3ZvrMtex/yt8xnaZmhUe42JicYcBEC52HJc1OAiv/MQSinWuRRJtXdTMb7kzWJnGvd0vIcYiaFR1Ua0re18wMxIYGeh3FvAAOAeQDDWRTSIsFylkmtbXEteYR5T1k7xmebFX19k3/F9jOruPXqkNU6SZxymOkl1+OjqjxjSZgjTr5/O5uGbGdVjVJE48QlxCXwz8BsaVm1I34l9+WvvXyejum7fDlvzVlClfBWuanZVOLcaMqmpkJAA8+cbjVFGBjRJacIzPZ5h8/DNfH/99wxtM5RneoS/4UmVhCq8e+W7rN6zmifmPsErC16hda3WdG3gzEInfwxpM4Tcglz+9eO/AqYdv3w8sRLL4FaDIy6XPxITT40gwJiHWJe9jq0HtnpNv3rPao7tqstZ6YVez2uCo2HVhjx/yfM8e8mzp4V5CbC1DuIPj7+JwIxA+aLxidQ6CJPCwkLV4JUGqvenvb2e33lop6o0qpIaMGmAzzIOHTL80UGpWbNCl2VDzgZV84WaquHohmrqrL0ny4zr/qS6/dvbQy/YAc4+W6kaNU753UeaIV8PUYxEMRI1btm4yF/QzYMzHgy4niC/IF/Ve6me6vVJrxKTyxfNmyvVr9+p7yt2rlCMRL2+8HW189DOYp/HZ41UxB5Tt99bfD9szZkDYa6DMJfnHRWRukAexuK5ModpZpqxfobX8NOjfh7F8fzjPHXxUz7LqFTJsAVDeIH6GlVrxHfXfcfuI7v59/L+J4/nl98VNfOSSXo67Nlz6v9I8/JlL1MnsQ61KtVyfCWsP57o9gRNUppw8xTfMaJmbZzFtkPbov5MwJioto4gzql5DrUq1eKe7++h9ku1i32e/PYDKEig7dlndFxOjR/suJB8KyJVgReApYAC3o2kUKWZfi368dJvL/Htum+LmAw27d/EW4vfYljbYX63DhQxTEs7d4YX6hugfd32fNbvM/pM6EtcxcPkH02kbq0Ex2IPhYo5UR0b6zuKq5NUq1CNX//xK8fyjlE+rnzkL+imQrkKfNDnAy4ceyEjZo7gzV5vFkszdvlYkiskc2WTK0tMLl94mphiJIZvBn7D0h1Lvab/a1E9XiPwdrGaMxe/CsK9UdAspdR+4EsRmYoRdiO03VvOADrU60Bq5VQmrZ5UREH42vDeG8nJhoIIZwRh0rtJb8b0+h+3v7MajnagZ8sOUbdvmg1Ko0anRkuRxt++zZGkS/0u3NvxXl5d+Cr9z+5fJNDb/uP7mbxmMre0u6VEFZcvEhONuF1WOqZ2pGOqdw+lt9wRwbWCKLv4NTEppQqBlyzfT5Rl5QBGr6tf8378kPkDB08cBPxveO+N5GTD7bBiRWdkuq39bZzdzPAy6deuuzOFhoFpVioJ81JpYFT3UTSu1ph/TPkHR/OOnjz+2crPOFFwolSYl6D4CCIQLpfhcFC3buRk0pRu7MxBzBCRayTa3dJSRL8W/ThRcIKp66YCRrylSuUqMeKCEbbyp6QYHydr1FzI1LS+vbhGkcTscZaVnmel+Eq83+d91u9bz79nnYoRNXb5WFrWbEm7Ou2iKN0pPOcgAmFG4o05vfYH0jiInTmIB4BKQL6IHMdwdVVKqcoRlawUc37a+dRNqntyyfzkvybzRLcnbAed69cPzjnHWZmuugpWroS0NGfLDYV69eC66+Caa6ItScnRrWE37mh/B68ufJV+LfqRXCGZhdsW8tJlL0Xd5GcS7AgiKwvqR36fKU0pJqCCUEppFwYPYiSGa5pfw7tL32XP0T0BN7z3ZHAE3OHbtoUvv3S+3FCIiYFPPom2FCXPc5c8xzTXNIZNGUbPs3oSK7Fc3/L6aIt1ksREOHLEWDNjZ1SQkwNnh7eWUXOaE1BBiMhF3o4rpeY5L87pQ78W/Xh90ev8suUXXvnbK343vNeUDZLKJ/Hule9y2ceXsS57HX2a9qFWYq1oi3USM2DfkSOnYjP5wwz+qCm72DExPWj5PwHoACwBoj8bGkW6pHWhdmJtysWUs7XhvaZscOlZl3Jz25t5b9l7tnbKK0msEV0DKYi8PGMlfLiu2JrTGzsmpiIO3CKSBjwfMYlOE2JjYpk8YDIV4ipEJXyzpvQyuudoLmpwUVT2ffCHZ0RXf+xzhxvTI4iyTSjbbWUBDk+xnp50Su0UbRE0pZBK8ZWiEmo9EJ57Qvgj273tiVYQZRs7cxCvY6yeBsMttg2wIoIyaTSaCGDdVS4QOTnGX21iKtvYGUEstvyfD0xQSnnf61Gj0ZRaghlBmApCjyDKNnYUxBfAcaVUAYCIxIpIRaXU0QD5NBpNKSKYOQhtYtKAvZXUs4AKlu8VgMC7nQMi0lNE1opIpogUW2YsIg+KyHL3Z6WIFIhIsvvcJhH5031ucfHSNRpNMIQygtAmprKNnRFEglLq5CullDosIgGjCIlILPAmcCnGxPbvIjJFKbXaUtYLGFFiEZErgfuVUjmWYi5WSu21dysajcYfwc5BxMZC5TIbL0ED9kYQR0TkZDAZETkXOGYjXwcgUym1QSmVC0wE/Pn9DQIm2ChXo9GEQLBeTNWqORsvTHP6YWcEMRyYJCLb3d/rYGxBGoh6gHUvwyzAa1xh94ikJ3C35bDCCBSogLeVUu/4yHsrcCtAfR04RqPxSXy8EX7drolJm5c0dhbK/S4izYCmGIH6/lJK5dko21vfQ3k5BnAlMN/DvNRFKbVdRGoCP4rIX97Ce7gVxzsA7du391W+RqPBfkTXnBw9Qa2xYWISkbuASkqplUqpP4FEEbnTRtlZgDW2aCqw3UfagXiYl5RS291/dwOTMUxWGo0mDBIT7c1B6DhMGrA3B3GLe0c5AJRS+4BbbOT7HcgQkUYiEo+hBKZ4JhKRKkBX4BvLsUoikmT+D1wGrLRxTY1G4we7Ib+1iUkD9uYgYkRElFIKTnonxQfKpJTKF5G7gR+AWOADpdQqEbndff4td9KrgRlKqSOW7LWAye44+nHAp0qp6XZvSqPReCcYBaFHEBo7CuIH4HMReQtjDuF24Hs7hSulpgHTPI695fF9HDDO49gGoLWda2g0GvvYmYPIzTXMUFpBaOwoiIcwvITuwJh4XobhyaTRaE4zEhNhb4CVRWYkV21i0gScg1BKFQILgA1Ae6AHsCbCcmk0mghgZ5Jax2HSmPgcQYhIE4yJ5UFANvAZgFLq4pIRTaPROI2dOQgdh0lj4s/E9BfwM3ClUioTQETsb7ys0WhKHXbmIHQcJo2JPxPTNcBOYLaIvCsiPfC++E2j0ZwmJCbC0aNQUOA7jTYxaUx8Kgil1GSl1ACgGTAHuB+oJSJjROSyEpJPo9E4iBmP6cgR32m0iUljYmeS+ohS6hOlVG+M1dDLgWKhuzUaTenHTsA+HclVY2JnJfVJlFI5Sqm3lVLdIyWQRqOJHHYVRHKyjuSqCVJBaDSa0xs7u8rpOEwaE60gNJoyhN0RhPZg0oBWEBpNmcLOrnI6DpPGRCsIjaYMYWcEoU1MGhOtIDSaMoSdOQhtYtKYaAWh0ZQhAo0gcnONc3oEoQGtIDSaMkWlSsZfX3MQehW1xopWEBpNGSI+3vj4GkHoOEwaK1pBaDRlDH8RXfUIQmNFKwiNpozhL6KrjsOksaIVhEZTxrAzgtAmJg1oBaHRlDn87SqnTUwaK1pBaDRlDH8jiOxsiIs7tV5CU7bRCkKjKWP4m4PQkVw1VrSC0GjKGIHmILR5SWOiFYRGU8bwNweh4zBprGgFodGUMQKNILQHk8ZEKwiNpoyRmAjHjkFBQfFz2sSksaIVhEZTxjA9lI4cKX5Om5g0VrSC0GjKGL42DTpxwlAa2sSkMdEKQqMpY/gK+b1vn/FXjyA0JlpBaDRlDF8KQsdh0niiFYRGU8bwtaucjsOk8UQrCI2mjOFrBKHjMGk80QpCoylj+Jqk1iYmjScRVRAi0lNE1opIpoiM8HL+QRFZ7v6sFJECEUm2k1ej0YRGoBGENjFpTCKmIEQkFngTuBxoAQwSkRbWNEqpF5RSbZRSbYCHgblKqRw7eTUaTWj4UxBxcafOazSRHEF0ADKVUhuUUrnARKCvn/SDgAkh5tVoNDbx58WUkqIjuWpOEUkFUQ/Yavme5T5WDBGpCPQEvgwh760islhEFu/ZsydsoTWaM51y5aB8+eJzEDrMhsaTSCoIb/0Q5SPtlcB8pVROsHmVUu8opdorpdrXqFEjBDE1mrKHt4B9WkFoPImkgsgC0izfU4HtPtIO5JR5Kdi8Go0mSLwpCNPEpNGYRFJB/A5kiEgjEYnHUAJTPBOJSBWgK/BNsHk1Gk1oeNtVTo8gNJ7ERapgpVS+iNwN/ADEAh8opVaJyO3u82+5k14NzFBKHQmUN1KyajRlDW1i0tghYgoCQCk1DZjmcewtj+/jgHF28mo0Gmfw3FVOR3LVeEOvpNZoyiCeIwgdZkPjDa0gNJoyiOcchFYQGm9oBaHRlEE8RxBmHCZtYtJY0QpCoymDeM5B6BGExhtaQWg0ZZDERDh+HPLzje9aQWi8oRWERlMGMeMxHXE7l2sTk8YbWkFoNGUQz13lcnKMGE2VKkVPJk3pQysIjaYM4hnR1VwkpyO5aqxoBaHRlEE8d5XTcZg03tAKQqMpg/gaQWg0VrSC0GjKIN7mILSC0HgS0VhMpYG8vDyysrI4fvx4tEXRRJiEhARSU1MpV65ctEUp9XiOILKzoV276MmjKZ2c8QoiKyuLpKQkGjZsiOgZuDMWpRTZ2dlkZWXRqFGjaItT6vGcg9AjCI03zngT0/Hjx0lJSdHK4QxHREhJSdEjRZtYRxDHj8PRo1pBaIpzxisIQCuHMoJ+zvaxKghzFbX2YtJ4UiYUhEajKUpcHCQkFFUQegSh8UQriAiSnZ1NmzZtaNOmDbVr16ZevXonv+fm5vrNu3jxYu69996A1+jcubNT4gJw3333Ua9ePQoLCx0tV1P6MAP2mWE2tILQeHLGT1JHk5SUFJYvXw7AyJEjSUxM5J///OfJ8/n5+cTFeX8E7du3p3379gGv8euvvzoiK0BhYSGTJ08mLS2NefPm0a1bN8fKtlJQUEBsbGxEytbYxwz5rU1MGl+UKQUxfPpwlu9c7miZbWq3YXTP0bbTDxkyhOTkZJYtW0a7du0YMGAAw4cP59ixY1SoUIGxY8fStGlT5syZw4svvsjUqVMZOXIkW7ZsYcOGDWzZsoXhw4efHF0kJiZy+PBh5syZw8iRI6levTorV67k3HPP5eOPP0ZEmDZtGg888ADVq1enXbt2bNiwgalTpxaTbfbs2ZxzzjkMGDCACRMmnFQQu3bt4vbbb2fDhg0AjBkzhs6dO/Phhx/y4osvIiK0atWKjz76iCFDhtC7d2/69etXTL4nnniCOnXqsHz5clavXs1VV13F1q1bOX78OPfddx+33norANOnT+eRRx6hoKCA6tWr8+OPP9K0aVN+/fVXatSoQWFhIU2aNGHBggVUr149jKdXtvFUEHoEofGkTCmI0sK6deuYOXMmsbGxHDx4kHnz5hEXF8fMmTN55JFH+PLLL4vl+euvv5g9ezaHDh2iadOm3HHHHcX8/ZctW8aqVauoW7cuXbp0Yf78+bRv357bbruNefPm0ahRIwYNGuRTrgkTJjBo0CD69u3LI488Ql5eHuXKlePee++la9euTJ48mYKCAg4fPsyqVasYNWoU8+fPp3r16uSYrYwfFi1axMqVK0+6oX7wwQckJydz7NgxzjvvPK655hoKCwu55ZZbTsqbk5NDTEwMgwcP5pNPPmH48OHMnDmT1q1ba+UQJuauctrEpPFFmVIQwfT0I8m111570sRy4MABbrrpJlwuFyJCXl6e1zy9evWifPnylC9fnpo1a7Jr1y5SU1OLpOnQocPJY23atGHTpk0kJibSuHHjk43yoEGDeOedd4qVn5uby7Rp03jllVdISkqiY8eOzJgxg169evHTTz/x4YcfAhAbG0uVKlX48MMP6dev38lGOtlG69KhQ4ciaxRee+01Jk+eDMDWrVtxuVzs2bOHiy666GQ6s9xhw4bRt29fhg8fzgcffMDQoUMDXk/jH+sIIj5eR3LVFKdMKYjSQiXLL/Gxxx7j4osvZvLkyWzatMmn3b98+fIn/4+NjSXf3OklQBqllC2Zpk+fzoEDB2jZsiUAR48epWLFivTq1ctreqWUV7fSuLi4kxPcSqkik/HW+54zZw4zZ87kt99+o2LFinTr1o3jx4/7LDctLY1atWrx008/sXDhQj755BNb96XxTWIi7NihI7lqfKO9mKLMgQMHqFevHgDjxo1zvPxmzZqxYcMGNm3aBMBnn33mNd2ECRN477332LRpE5s2bWLjxo3MmDGDo0eP0qNHD8aMGQMYE8wHDx6kR48efP7552S77ROmialhw4YsWbIEgG+++cbniOjAgQNUq1aNihUr8tdff7FgwQIAzj//fObOncvGjRuLlAtw8803M3jwYPr3768nuR3AHEFkZ2vzksY7WkFEmX/96188/PDDdOnShYKCAsfLr1ChAv/73//o2bMnF1xwAbVq1aJKlSpF0hw9epQffvihyGihUqVKXHDBBXz77be8+uqrzJ49m5YtW3LuueeyatUqzj77bP7973/TtWtXWrduzQMPPADALbfcwty5c+nQoQMLFy4sMmqw0rNnT/Lz82nVqhWPPfYYnTp1AqBGjRq88847/P3vf6d169YMGDDgZJ4+ffpw+PBhbV5yCKuJSXswabwhdk0QpwPt27dXixcvLnJszZo1NG/ePEoSlQ4OHz5MYmIiSinuuusuMjIyuP/++6MtVtAsXryY+++/n59//tlnGv287fPQQ/Daa9CkCTRqBF9/HW2JNNFARJYopbz61OsRRBng3XffpU2bNpx99tkcOHCA2267LdoiBc2zzz7LNddcw3//+99oi3LGkJhoxGHatUubmDTe0ZPUZYD777//tBwxWBkxYgQjRoyIthhnFGY8pl27tIlJ4x09gtBoyiimggA9gtB4RysIjaaMYu4qB1pBaLyjFYRGU0axjiC0iUnjDa0gNJoyijYxaQKhFUSE6datGz/88EORY6NHj+bOO+/0m8fTXddkz549lCtXjrfffttROTVlD60gNIHQCiLCDBo0iIkTJxY5NnHiRL9B8/wxadIkOnXqxIQJE5wQzyfeQnloziy0iUkTiIi6uYpIT+BVIBZ4Tyn1rJc03YDRQDlgr1Kqq/v4JuAQUADk+1rIEQzDh4N7ewbHaNMGRo/2fb5fv348+uijnDhxgvLly7Np0ya2b9/OBRdcwB133MHvv//OsWPH6NevH0888UTA602YMIGXXnqJ6667jm3btp0M0+Et9La3MN1169ald+/erFy5EoAXX3yRw4cPM3LkSLp160bnzp2ZP38+ffr0oUmTJjz99NPk5uaSkpLCJ598Qq1atTh8+DD33HMPixcvRkT4z3/+w/79+1m5ciWvvPIKYKy9WLNmDS+//HJY9auJHHqSWhOIiCkIEYkF3gQuBbKA30VkilJqtSVNVeB/QE+l1BYRqelRzMVKqb2RkrEkSElJoUOHDkyfPp2+ffsyceJEBgwYgIgwatQokpOTKSgooEePHvzxxx+0atXKZ1lbt25l586ddOjQgf79+/PZZ5/xwAMP+Ay97S1M9759+/zKu3//fubOnQvAvn37WLBgASLCe++9x/PPP89LL73EU089RZUqVfjzzz9PpouPj6dVq1Y8//zzlCtXjrFjx2ozWCnHHEHEx0PFitGVRVM6ieQIogOQqZTaACAiE4G+wGpLmuuAr5RSWwCUUrsjKI/fnn4kMc1MpoL44IMPAPj888955513yM/PZ8eOHaxevdqvgpg4cSL9+/cHYODAgfzjH//ggQce4KeffvIaettbmO5ACsIa+ygrK4sBAwawY8cOcnNzT4bgnjlzZhGzWbVq1QDo3r07U6dOpXnz5uTl5Z2MDKspnZhhslJSdCRXjXciOQdRD9hq+Z7lPmalCVBNROaIyBIRudFyTgEz3Mdv9XUREblVRBaLyOI9e/Y4JryTXHXVVcyaNYulS5dy7Ngx2rVrx8aNG3nxxReZNWsWf/zxB7169eL48eN+y5kwYQLjxo2jYcOG9OnThxUrVuByuXyGyPaGNRw3UOya1uB699xzD3fffTd//vknb7/99sm0vq538803M27cOMaOHasD6p0GxMVBQoI2L2l8E0kF4a3F8owMGAecC/QC/gY8JiJN3Oe6KKXaAZcDd4nIRd4uopR6RynVXinVvkaNGg6J7iyJiYl069aNYcOGnZycPnjwIJUqVaJKlSrs2rWL77//3m8Za9eu5ciRI2zbtu1kSO6HH36YiRMn+gy97S1Md61atdi9ezfZ2dmcOHHC69ajJtZQ5OPHjz95/LLLLuONN944+d0clXTs2JGtW7fy6aefhjwJrylZkpK0gtD4JpIKIgtIs3xPBbZ7STNdKXXEPdcwD2gNoJTa7v67G5iMYbI6bRk0aBArVqxg4MCBALRu3Zq2bdty9tlnM2zYMLp06eI3/4QJE7j66quLHLvmmmuYMGGCz9Db3sJ0lytXjscff5yOHTvSu3dvmjVr5vOaI0eO5Nprr+XCCy8ssr3no48+yr59+zjnnHNo3bo1s2fPPnmuf//+dOnS5aTZSVO6SUzUCkLjm4iF+xaROGAd0APYBvwOXKeUWmVJ0xx4A2P0EA8sAgYCG4EYpdQhEakE/Ag8qZSa7u+aOtx39Onduzf3338/PXr0iMr19fMOjg8/hNRU6N492pJoooW/cN8Rm6RWSuWLyN3ADxhurh8opVaJyO3u828ppdaIyHTgD6AQwxV2pYg0Bia77dxxwKeBlIMmuuzfv58OHTrQunXrqCkHTfDceGPgNJqyS0TXQSilpgHTPI695fH9BeAFj2MbcJuaNKcHVatWZd26ddEWQ6PROEiZWEl9Ju2ap/GNfs4ajbOc8QoiISGB7Oxs3Xic4SilyM7OJiEhIdqiaDRnDGf8jnKpqalkZWVRWtdIaJwjISGB1NTUaIuh0ZwxnPEKoly5cidXAGs0Go3GPme8iUmj0Wg0oaEVhEaj0Wi8ohWERqPRaLwSsZXU0UBE9gCb/SSpDpzW4cMjhK4X3+i68Y2uG9+cTnXTQCnlNZDdGaUgAiEii53YeOhMQ9eLb3Td+EbXjW/OlLrRJiaNRqPReEUrCI1Go9F4pawpiHeiLUApRdeLb3Td+EbXjW/OiLopU3MQGo1Go7FPWRtBaDQajcYmWkFoNBqNxitlQkGISE8RWSsimSIyItryRBMR+UBEdovISsuxZBH5UURc7r9lcr9QEUkTkdkiskZEVonIfe7jZbp+RCRBRBaJyAp3vTzhPl6m68WKiMSKyDIRmer+fkbUzRmvIEQkFngTuBxoAQwSkRbRlSqqjAN6ehwbAcxSSmUAs9zfyyL5wP8ppZoDnYC73O9KWa+fE0B3pVRroA3QU0Q6oevFyn3AGsv3M6JuzngFAXQAMpVSG5RSucBEoG+UZYoaSql5QI7H4b7AePf/44GrSlKm0oJSaodSaqn7/0MYP/h6lPH6UQaH3V/LuT+KMl4vJiKSCvQC3rMcPiPqpiwoiHrAVsv3LPcxzSlqKaV2gNFIAjWjLE/UEZGGQFtgIbp+TBPKcmA38KNSStfLKUYD/wIKLcfOiLopCwpCvBzTvr0an4hIIvAlMFwpdTDa8pQGlFIFSqk2QCrQQUTOibJIpQIR6Q3sVkotibYskaAsKIgsIM3yPRXYHiVZSiu7RKQOgPvv7ijLEzVEpByGcvhEKfWV+7CuHzdKqf3AHIx5LF0v0AXoIyKbMMzX3UXkY86QuikLCuJ3IENEGolIPDAQmBJlmUobU4Cb3P/fBHwTRVmihogI8D6wRin1suVUma4fEakhIlXd/1cALgH+oozXC4BS6mGlVKpSqiFG2/KTUmowZ0jdlImV1CJyBYadMBb4QCk1KroSRQ8RmQB0wwhHvAv4D/A18DlQH9gCXKuU8pzIPuMRkQuAn4E/OWVPfgRjHqLM1o+ItMKYaI3F6FR+rpR6UkRSKMP14omIdAP+qZTqfabUTZlQEBqNRqMJnrJgYtJoNBpNCGgFodFoNBqvaAWh0Wg0Gq9oBaHRaDQar2gFodFoNBqvaAWh0QRARApEZLnl41jgNRFpaI2sq9GUJuKiLYBGcxpwzB1mQqMpU+gRhEYTIiKySUSec++VsEhE0t3HG4jILBH5w/23vvt4LRGZ7N5XYYWIdHYXFSsi77r3WpjhXq2MiNwrIqvd5UyM0m1qyjBaQWg0gangYWIaYDl3UCnVAXgDY7U+7v8/VEq1Aj4BXnMffw2Y695XoR2wyn08A3hTKXU2sB+4xn18BNDWXc7tkbk1jcY3eiW1RhMAETmslEr0cnwTxkY6G9xB/nYqpVJEZC9QRymV5z6+QylVXUT2AKlKqROWMhpihM/OcH9/CCinlHpaRKYDhzFCoXxt2ZNBoykR9AhCowkP5eN/X2m8ccLyfwGn5gZ7YeyGeC6wRET0nKGmRNEKQqMJjwGWv7+5//8VI7InwPXAL+7/ZwF3wMkNeCr7KlREYoA0pdRsjM1oqgLFRjEaTSTRPRKNJjAV3LupmUxXSpmuruVFZCFGZ2uQ+9i9wAci8iCwBxjqPn4f8I6I/ANjpHAHsMPHNWOBj0WkCsamV6+492LQaEoMPQeh0YSIew6ivVJqb7Rl0WgigTYxaTQajcYregSh0Wg0Gq/oEYRGo9FovKIVhEaj0Wi8ohWERqPRaLyiFYRGo9FovKIVhEaj0Wi88v9w5ueiFshk1QAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "loss_train = history.history['accuracy']\n",
    "loss_val = history.history['val_accuracy']\n",
    "epochs = range(1,45)\n",
    "plt.plot(epochs,loss_train,'g',label = \"Training Accuracy\")\n",
    "plt.plot(epochs,loss_val,'b',label = \"Val Accuracy\")\n",
    "plt.title('Training and Val accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss: 0.467608779668808\n",
      "Accuracy: 0.8823529481887817\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(X_test,y_test,verbose=0)\n",
    "print('Loss:',score[0])\n",
    "print('Accuracy:',score[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('new.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "ripeness_detector = keras.models.load_model('new.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss: 0.20008975267410278\n",
      "Accuracy: 0.9411764740943909\n"
     ]
    }
   ],
   "source": [
    "score = ripeness_detector.evaluate(X_test,y_test,verbose=0)\n",
    "print('Loss:',score[0])\n",
    "print('Accuracy:',score[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "array = np.array([[0,1,2],[255,10,23],[203,210,33],[100,100,100],[0,255,30]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = ripeness_detector(array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "ripeness_class_name = ['Early Ripe','Partially Ripe','Ripe','Decay']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Input[1] Decay\n",
      "Input[2] Ripe\n",
      "Input[3] Ripe\n",
      "Input[4] Early Ripe\n",
      "Input[5] Partially Ripe\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(array)):\n",
    "    print(\"Input[\"+str(i+1)+\"]\",ripeness_class_name[np.argmax(predictions[i])])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "97ae724bfa85b9b34df7982b8bb8c7216f435b92902d749e4263f71162bea840"
  },
  "kernelspec": {
   "display_name": "Python 3.9.7 64-bit ('base': conda)",
   "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": "3.9.7"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}