src/2.Experiments/0.TrainTestSplit.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train&Validation - Test Split\n",
    "****************\n",
    "\n",
    "**<div itemscope itemtype=\"https://schema.org/Person\"><a itemprop=\"sameAs\" content=\"https://orcid.org/0000-0001-5567-801X\" href=\"https://orcid.org/0000-0001-5567-801X\" target=\"orcid.widget\" rel=\"me noopener noreferrer\" style=\"vertical-align:top;\"><img src=\"https://orcid.org/sites/default/files/images/orcid_16x16.png\" style=\"width:1em;margin-left:0em;\"  alt=\"ORCID iD icon\">Adrián Arnaiz-Rodríguez</a></div>**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc": true
   },
   "source": [
    "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n",
    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#Approach\" data-toc-modified-id=\"Approach-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>Approach</a></span></li><li><span><a href=\"#Get-the-valid-number-of-slices-of-each-volume\" data-toc-modified-id=\"Get-the-valid-number-of-slices-of-each-volume-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>Get the valid number of slices of each volume</a></span><ul class=\"toc-item\"><li><span><a href=\"#How-many-RELEVANT-Slides-has-each-volume?\" data-toc-modified-id=\"How-many-RELEVANT-Slides-has-each-volume?-2.1\"><span class=\"toc-item-num\">2.1&nbsp;&nbsp;</span>How many RELEVANT Slides has each volume?</a></span></li><li><span><a href=\"#How-many-volumes-should-be-in-each-set?\" data-toc-modified-id=\"How-many-volumes-should-be-in-each-set?-2.2\"><span class=\"toc-item-num\">2.2&nbsp;&nbsp;</span>How many volumes should be in each set?</a></span></li><li><span><a href=\"#How-many-slides-should-be-on-each-set?\" data-toc-modified-id=\"How-many-slides-should-be-on-each-set?-2.3\"><span class=\"toc-item-num\">2.3&nbsp;&nbsp;</span>How many slides should be on each set?</a></span></li></ul></li><li><span><a href=\"#Stratified-vol-splitting\" data-toc-modified-id=\"Stratified-vol-splitting-3\"><span class=\"toc-item-num\">3&nbsp;&nbsp;</span>Stratified vol splitting</a></span><ul class=\"toc-item\"><li><span><a href=\"#Select-metadata-from-our-MRI-volumes\" data-toc-modified-id=\"Select-metadata-from-our-MRI-volumes-3.1\"><span class=\"toc-item-num\">3.1&nbsp;&nbsp;</span>Select metadata from our MRI volumes</a></span></li><li><span><a href=\"#Identify-non-age-volumes-and-remove-duplicated\" data-toc-modified-id=\"Identify-non-age-volumes-and-remove-duplicated-3.2\"><span class=\"toc-item-num\">3.2&nbsp;&nbsp;</span>Identify non-age volumes and remove duplicated</a></span></li><li><span><a href=\"#Identify-attribute-distribution\" data-toc-modified-id=\"Identify-attribute-distribution-3.3\"><span class=\"toc-item-num\">3.3&nbsp;&nbsp;</span>Identify attribute distribution</a></span><ul class=\"toc-item\"><li><span><a href=\"#Age\" data-toc-modified-id=\"Age-3.3.1\"><span class=\"toc-item-num\">3.3.1&nbsp;&nbsp;</span>Age</a></span></li><li><span><a href=\"#Sex\" data-toc-modified-id=\"Sex-3.3.2\"><span class=\"toc-item-num\">3.3.2&nbsp;&nbsp;</span>Sex</a></span></li><li><span><a href=\"#Ethnic\" data-toc-modified-id=\"Ethnic-3.3.3\"><span class=\"toc-item-num\">3.3.3&nbsp;&nbsp;</span>Ethnic</a></span></li></ul></li><li><span><a href=\"#Stratify-by-age-and-ethnic\" data-toc-modified-id=\"Stratify-by-age-and-ethnic-3.4\"><span class=\"toc-item-num\">3.4&nbsp;&nbsp;</span>Stratify by age and ethnic</a></span><ul class=\"toc-item\"><li><span><a href=\"#How-are-going-to-be-stratified?\" data-toc-modified-id=\"How-are-going-to-be-stratified?-3.4.1\"><span class=\"toc-item-num\">3.4.1&nbsp;&nbsp;</span>How are going to be stratified?</a></span><ul class=\"toc-item\"><li><span><a href=\"#Test\" data-toc-modified-id=\"Test-3.4.1.1\"><span class=\"toc-item-num\">3.4.1.1&nbsp;&nbsp;</span>Test</a></span></li><li><span><a href=\"#Train-and-validation\" data-toc-modified-id=\"Train-and-validation-3.4.1.2\"><span class=\"toc-item-num\">3.4.1.2&nbsp;&nbsp;</span>Train and validation</a></span></li></ul></li><li><span><a href=\"#Check-Number-of-slices-in-each-set\" data-toc-modified-id=\"Check-Number-of-slices-in-each-set-3.4.2\"><span class=\"toc-item-num\">3.4.2&nbsp;&nbsp;</span>Check Number of slices in each set</a></span><ul class=\"toc-item\"><li><span><a href=\"#Train&amp;Val\" data-toc-modified-id=\"Train&amp;Val-3.4.2.1\"><span class=\"toc-item-num\">3.4.2.1&nbsp;&nbsp;</span>Train&amp;Val</a></span></li><li><span><a href=\"#Test\" data-toc-modified-id=\"Test-3.4.2.2\"><span class=\"toc-item-num\">3.4.2.2&nbsp;&nbsp;</span>Test</a></span></li></ul></li></ul></li></ul></li><li><span><a href=\"#Conclussion\" data-toc-modified-id=\"Conclussion-4\"><span class=\"toc-item-num\">4&nbsp;&nbsp;</span>Conclussion</a></span><ul class=\"toc-item\"><li><span><a href=\"#Size-of-each-set\" data-toc-modified-id=\"Size-of-each-set-4.1\"><span class=\"toc-item-num\">4.1&nbsp;&nbsp;</span>Size of each set</a></span></li><li><span><a href=\"#Check-attribute-distributions-in-each-set\" data-toc-modified-id=\"Check-attribute-distributions-in-each-set-4.2\"><span class=\"toc-item-num\">4.2&nbsp;&nbsp;</span>Check attribute distributions in each set</a></span></li></ul></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.069069Z",
     "start_time": "2020-11-12T10:11:02.902312Z"
    }
   },
   "outputs": [],
   "source": [
    "import pickle as pkl\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Approach\n",
    "We will dived in a stratified way the train and test data. Train images will be splitted in train and validation through ImageDataGenerator of keras.\n",
    "\n",
    "The data will be **stratified by sex and age**\n",
    "\n",
    "**Approach**: Get aprox 90% for training and validation and 8% for final testing. Then get 85% of this 92% as training set and the 15% as validation set. Thus, we had 76.5% for training, 13.5% for validation and 10% for testing.\n",
    "\n",
    "* Get metadata from volumes\n",
    "* Stratify volumes from this metadata distributions\n",
    "* Set separate folder for images extracted from train&validation volumes and for test volumes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Get the valid number of slices of each volume\n",
    "\n",
    "**This will be used later. Onces volumes will be stratified, we check if the number of relevant slices are similar to he required one**\n",
    "\n",
    "**WE DO NOT STRATIFY STRIGHT FROM IMAGES DUE TO THE BIG CORRELATION OF THE IMAGES OF A SAME VOLUME. IF WE SPLIT IN A STRATIFY WAY FROM THE IMAGES, TRAIN AND TEST WILL BE ALMOST THE SAME. MAYBE, IT COULD HAPPENS THAT 2 SLICES, WHICH ARE ALMOST IDENTICAL BECAUSE THEY ARE NEAR IN THE SAME VOLUME, BELONG TO DIFFERENT SETS. IT IS BETTER TO STRATIFY  FROM VOLUMES TO AVOID THIS BIG CORRELATION**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.130131Z",
     "start_time": "2020-11-12T10:11:04.072924Z"
    }
   },
   "outputs": [],
   "source": [
    "with open('../1.DataPreprocessing/deepbrain_image_data.pickle', 'rb') as handle:\n",
    "    df_deepbrain_data = pkl.load(handle) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.366511Z",
     "start_time": "2020-11-12T10:11:04.143742Z"
    }
   },
   "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>ID</th>\n",
       "      <th>BRAIN_QUANTITY</th>\n",
       "      <th>VOL_ID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9468</th>\n",
       "      <td>IXI077-Guys-0752-T1_82</td>\n",
       "      <td>15997.0</td>\n",
       "      <td>77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5777</th>\n",
       "      <td>IXI052-HH-1343-T1_141</td>\n",
       "      <td>0.0</td>\n",
       "      <td>52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73767</th>\n",
       "      <td>IXI551-Guys-1065-T1_133</td>\n",
       "      <td>0.0</td>\n",
       "      <td>551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34724</th>\n",
       "      <td>IXI263-HH-1684-T1_12</td>\n",
       "      <td>0.0</td>\n",
       "      <td>263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28357</th>\n",
       "      <td>IXI216-HH-1635-T1_71</td>\n",
       "      <td>17939.0</td>\n",
       "      <td>216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8906</th>\n",
       "      <td>IXI073-Guys-0755-T1_120</td>\n",
       "      <td>5592.0</td>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49523</th>\n",
       "      <td>IXI377-Guys-0937-T1_57</td>\n",
       "      <td>14312.0</td>\n",
       "      <td>377</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            ID  BRAIN_QUANTITY  VOL_ID\n",
       "9468    IXI077-Guys-0752-T1_82         15997.0      77\n",
       "5777     IXI052-HH-1343-T1_141             0.0      52\n",
       "73767  IXI551-Guys-1065-T1_133             0.0     551\n",
       "34724     IXI263-HH-1684-T1_12             0.0     263\n",
       "28357     IXI216-HH-1635-T1_71         17939.0     216\n",
       "8906   IXI073-Guys-0755-T1_120          5592.0      73\n",
       "49523   IXI377-Guys-0937-T1_57         14312.0     377"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_deepbrain_data['VOL_ID'] = df_deepbrain_data['ID'].apply(lambda x: x.split('-')[0][3:])\n",
    "df_deepbrain_data = df_deepbrain_data.astype({'VOL_ID': 'int64'})\n",
    "df_deepbrain_data.sample(7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.429267Z",
     "start_time": "2020-11-12T10:11:04.371555Z"
    }
   },
   "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>BRAIN_QUANTITY</th>\n",
       "      <th>VOL_ID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>86794.000000</td>\n",
       "      <td>86794.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>9051.895419</td>\n",
       "      <td>327.90451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6956.157762</td>\n",
       "      <td>185.38492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>167.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>10785.500000</td>\n",
       "      <td>328.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>15382.000000</td>\n",
       "      <td>485.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>22830.000000</td>\n",
       "      <td>662.00000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       BRAIN_QUANTITY       VOL_ID\n",
       "count    86794.000000  86794.00000\n",
       "mean      9051.895419    327.90451\n",
       "std       6956.157762    185.38492\n",
       "min          0.000000      2.00000\n",
       "25%          0.000000    167.00000\n",
       "50%      10785.500000    328.00000\n",
       "75%      15382.000000    485.00000\n",
       "max      22830.000000    662.00000"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_deepbrain_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.506646Z",
     "start_time": "2020-11-12T10:11:04.433776Z"
    }
   },
   "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>BRAIN_QUANTITY</th>\n",
       "      <th>VOL_ID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>59278.000000</td>\n",
       "      <td>59278.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>13122.107257</td>\n",
       "      <td>328.464186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4285.977381</td>\n",
       "      <td>185.880795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>3001.000000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>10387.250000</td>\n",
       "      <td>166.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>14275.000000</td>\n",
       "      <td>328.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>16264.750000</td>\n",
       "      <td>486.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>22830.000000</td>\n",
       "      <td>662.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       BRAIN_QUANTITY        VOL_ID\n",
       "count    59278.000000  59278.000000\n",
       "mean     13122.107257    328.464186\n",
       "std       4285.977381    185.880795\n",
       "min       3001.000000      2.000000\n",
       "25%      10387.250000    166.000000\n",
       "50%      14275.000000    328.000000\n",
       "75%      16264.750000    486.000000\n",
       "max      22830.000000    662.000000"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "slices_selected = df_deepbrain_data[df_deepbrain_data['BRAIN_QUANTITY']>3000].copy()\n",
    "slices_selected.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.522650Z",
     "start_time": "2020-11-12T10:11:04.509566Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(581,)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "slices_selected['VOL_ID'].unique().shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### How many RELEVANT Slides has each volume?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.569405Z",
     "start_time": "2020-11-12T10:11:04.527244Z"
    }
   },
   "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>Count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>581.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>102.027539</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.573077</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>87.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>99.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>102.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>105.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>114.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Count\n",
       "count  581.000000\n",
       "mean   102.027539\n",
       "std      4.573077\n",
       "min     87.000000\n",
       "25%     99.000000\n",
       "50%    102.000000\n",
       "75%    105.000000\n",
       "max    114.000000"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "slices_per_volume = slices_selected.groupby('VOL_ID').count().drop(columns=['BRAIN_QUANTITY'], \n",
    "                                                                   axis=1).rename(columns={'ID':'Count'})\n",
    "slices_per_volume.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.585392Z",
     "start_time": "2020-11-12T10:11:04.572131Z"
    }
   },
   "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>Count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VOL_ID</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>431</th>\n",
       "      <td>93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>160</th>\n",
       "      <td>97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>342</th>\n",
       "      <td>106</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Count\n",
       "VOL_ID       \n",
       "431        93\n",
       "160        97\n",
       "342       106"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "slices_per_volume.sample(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.775900Z",
     "start_time": "2020-11-12T10:11:04.587358Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "slices_per_volume.plot()\n",
    "plt.ylim(20);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### How many volumes should be in each set?\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.791388Z",
     "start_time": "2020-11-12T10:11:04.777848Z"
    }
   },
   "outputs": [],
   "source": [
    "train_and_val_size = 0.92\n",
    "train_size = 0.85 #out of 0.9\n",
    "validation_size = 0.15\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.807468Z",
     "start_time": "2020-11-12T10:11:04.793383Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(534, 0.92)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_vols = len(slices_per_volume)\n",
    "train_and_val_vol = int(total_vols*train_and_val_size)\n",
    "train_and_val_vol, train_and_val_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.822336Z",
     "start_time": "2020-11-12T10:11:04.809341Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(47, 0.07999999999999996)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_vols = total_vols - train_and_val_vol\n",
    "test_vols, 1-train_and_val_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.838217Z",
     "start_time": "2020-11-12T10:11:04.825859Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(453, 0.782)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_vols = int(train_and_val_vol*train_size)\n",
    "train_vols, train_and_val_size*train_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.853344Z",
     "start_time": "2020-11-12T10:11:04.841496Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(81, 0.138)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validation_vols = train_and_val_vol - train_vols\n",
    "validation_vols, train_and_val_size*validation_size"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### How many slides should be on each set?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.869211Z",
     "start_time": "2020-11-12T10:11:04.856217Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(53350, 0.9)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_slices = len(slices_selected)\n",
    "train_and_val_slices = int(total_slices*0.9)\n",
    "train_and_val_slices, 0.9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.884227Z",
     "start_time": "2020-11-12T10:11:04.872176Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5928, 0.1)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_slices = total_slices - train_and_val_slices\n",
    "test_slices, 0.1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.899132Z",
     "start_time": "2020-11-12T10:11:04.886143Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(45347, 0.765)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_slices = int(train_and_val_slices*0.85)\n",
    "train_slices, 0.9*0.85"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.914578Z",
     "start_time": "2020-11-12T10:11:04.902092Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8003, 0.135)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validation_slices = train_and_val_slices - train_slices\n",
    "validation_slices, 0.15*0.9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Stratified vol splitting\n",
    "\n",
    "* Get only volumes from whic we have MRI data\n",
    "    * If any Volume dont have metadata: it will go to train test\n",
    "* Identify volumes without metadata\n",
    "* Remove duplicated\n",
    "* Identify the distribution of each attribute\n",
    "* Stratify volumes by Age and Sex\n",
    "\n",
    "### Select metadata from our MRI volumes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:04.930051Z",
     "start_time": "2020-11-12T10:11:04.917055Z"
    }
   },
   "outputs": [],
   "source": [
    "IXI_metadata_xls = '../IXI-T1/IXI.xls'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.069679Z",
     "start_time": "2020-11-12T10:11:04.933010Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 619 entries, 0 to 618\n",
      "Data columns (total 4 columns):\n",
      " #   Column     Non-Null Count  Dtype   \n",
      "---  ------     --------------  -----   \n",
      " 0   IXI_ID     619 non-null    int64   \n",
      " 1   SEX        619 non-null    category\n",
      " 2   ETHNIC_ID  619 non-null    category\n",
      " 3   AGE        590 non-null    float64 \n",
      "dtypes: category(2), float64(1), int64(1)\n",
      "memory usage: 11.5 KB\n"
     ]
    }
   ],
   "source": [
    "IXI_metadata = pd.read_excel(IXI_metadata_xls)[['IXI_ID',\n",
    "                                                'SEX_ID (1=m, 2=f)',\n",
    "                                                'ETHNIC_ID',\n",
    "                                                'AGE']].rename(columns={'ID':'Count', 'SEX_ID (1=m, 2=f)': 'SEX'})\n",
    "IXI_metadata['SEX'] = IXI_metadata['SEX'].astype('category')\n",
    "IXI_metadata['ETHNIC_ID'] = IXI_metadata['ETHNIC_ID'].astype('category')\n",
    "IXI_metadata.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.099785Z",
     "start_time": "2020-11-12T10:11:05.072667Z"
    }
   },
   "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>IXI_ID</th>\n",
       "      <th>SEX</th>\n",
       "      <th>ETHNIC_ID</th>\n",
       "      <th>AGE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>551</th>\n",
       "      <td>577</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>64.186174</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>562</th>\n",
       "      <td>591</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>59.893224</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     IXI_ID SEX ETHNIC_ID        AGE\n",
       "551     577   2         1  64.186174\n",
       "562     591   2         1  59.893224"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "IXI_metadata.sample(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T18:51:51.060949Z",
     "start_time": "2020-11-11T18:51:51.018423Z"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**We select only metadata of OUR MRI VOLUMES**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.130517Z",
     "start_time": "2020-11-12T10:11:05.104550Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 591 entries, 1 to 618\n",
      "Data columns (total 4 columns):\n",
      " #   Column     Non-Null Count  Dtype   \n",
      "---  ------     --------------  -----   \n",
      " 0   IXI_ID     591 non-null    int64   \n",
      " 1   SEX        591 non-null    category\n",
      " 2   ETHNIC_ID  591 non-null    category\n",
      " 3   AGE        588 non-null    float64 \n",
      "dtypes: category(2), float64(1), int64(1)\n",
      "memory usage: 15.5 KB\n"
     ]
    }
   ],
   "source": [
    "IXI_metadata = IXI_metadata[IXI_metadata['IXI_ID'].isin(list(slices_selected['VOL_ID'].unique()))]\n",
    "IXI_metadata.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**We discover 15 volumes with no data which will be included in train test**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.146780Z",
     "start_time": "2020-11-12T10:11:05.133503Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "15"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "MRI_with_no_data = set(slices_selected['VOL_ID'].unique()) - set(IXI_metadata['IXI_ID'].unique())\n",
    "len(MRI_with_no_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**We join the information about the number of relevant slices**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.272132Z",
     "start_time": "2020-11-12T10:11:05.148434Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 591 entries, 1 to 618\n",
      "Data columns (total 5 columns):\n",
      " #   Column             Non-Null Count  Dtype   \n",
      "---  ------             --------------  -----   \n",
      " 0   IXI_ID             591 non-null    int64   \n",
      " 1   SEX                591 non-null    category\n",
      " 2   ETHNIC_ID          591 non-null    category\n",
      " 3   AGE                588 non-null    float64 \n",
      " 4   N_Relevant_Slices  591 non-null    int64   \n",
      "dtypes: category(2), float64(1), int64(2)\n",
      "memory usage: 20.1 KB\n"
     ]
    }
   ],
   "source": [
    "IXI_metadata['N_Relevant_Slices'] = IXI_metadata.apply(lambda x: slices_per_volume.loc[x['IXI_ID']], axis=1)\n",
    "IXI_metadata.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.288199Z",
     "start_time": "2020-11-12T10:11:05.274243Z"
    }
   },
   "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>IXI_ID</th>\n",
       "      <th>SEX</th>\n",
       "      <th>ETHNIC_ID</th>\n",
       "      <th>AGE</th>\n",
       "      <th>N_Relevant_Slices</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>324</th>\n",
       "      <td>360</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>54.193018</td>\n",
       "      <td>99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553</th>\n",
       "      <td>579</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>41.215606</td>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     IXI_ID SEX ETHNIC_ID        AGE  N_Relevant_Slices\n",
       "324     360   2         1  54.193018                 99\n",
       "553     579   1         1  41.215606                 98"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "IXI_metadata.sample(n=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Identify non-age volumes and remove duplicated"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**We discover other 3 volumes with no data which will be included in train test**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.320262Z",
     "start_time": "2020-11-12T10:11:05.291328Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Duplicados: False\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(3, 5)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "non_data = IXI_metadata[IXI_metadata['AGE'].isna()]\n",
    "\n",
    "print('Duplicados:',np.any(non_data.duplicated('IXI_ID')))\n",
    "if np.any(non_data.duplicated('IXI_ID')):\n",
    "    non_data = non_data.drop_duplicates(subset = 'IXI_ID')\n",
    "non_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.335551Z",
     "start_time": "2020-11-12T10:11:05.323251Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "18"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "MRI_with_no_data.update(list(non_data['IXI_ID']))\n",
    "len(MRI_with_no_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.351184Z",
     "start_time": "2020-11-12T10:11:05.338448Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(563, 5)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "IXI_metadata = IXI_metadata[IXI_metadata['AGE'].isna()==False]\n",
    "IXI_metadata.drop_duplicates(subset = 'IXI_ID', inplace=True)\n",
    "IXI_metadata.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.381611Z",
     "start_time": "2020-11-12T10:11:05.354136Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 563 entries, 1 to 618\n",
      "Data columns (total 5 columns):\n",
      " #   Column             Non-Null Count  Dtype   \n",
      "---  ------             --------------  -----   \n",
      " 0   IXI_ID             563 non-null    int64   \n",
      " 1   SEX                563 non-null    category\n",
      " 2   ETHNIC_ID          563 non-null    category\n",
      " 3   AGE                563 non-null    float64 \n",
      " 4   N_Relevant_Slices  563 non-null    int64   \n",
      "dtypes: category(2), float64(1), int64(2)\n",
      "memory usage: 19.2 KB\n"
     ]
    }
   ],
   "source": [
    "IXI_metadata.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Identify attribute distribution\n",
    "#### Age"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.397200Z",
     "start_time": "2020-11-12T10:11:05.383135Z"
    }
   },
   "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>AGE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>563.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>48.651764</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>16.472913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>19.980835</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>33.724846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>48.678987</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>62.169747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>86.318960</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              AGE\n",
       "count  563.000000\n",
       "mean    48.651764\n",
       "std     16.472913\n",
       "min     19.980835\n",
       "25%     33.724846\n",
       "50%     48.678987\n",
       "75%     62.169747\n",
       "max     86.318960"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(IXI_metadata['AGE'].describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.412012Z",
     "start_time": "2020-11-12T10:11:05.399043Z"
    }
   },
   "outputs": [],
   "source": [
    "IXI_metadata.loc[IXI_metadata['AGE'] >= 65, 'AGE_GROUP'] = 'elderly'\n",
    "IXI_metadata.loc[IXI_metadata['AGE'] < 25, 'AGE_GROUP'] = 'young'\n",
    "IXI_metadata.loc[(IXI_metadata['AGE'] < 65) & (IXI_metadata['AGE'] > 25), 'AGE_GROUP'] = 'adult'\n",
    "IXI_metadata['AGE_GROUP'] = IXI_metadata['AGE_GROUP'].astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.427968Z",
     "start_time": "2020-11-12T10:11:05.413977Z"
    }
   },
   "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>AGE_GROUP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>adult</th>\n",
       "      <td>0.733570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>elderly</th>\n",
       "      <td>0.191829</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>young</th>\n",
       "      <td>0.074600</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         AGE_GROUP\n",
       "adult     0.733570\n",
       "elderly   0.191829\n",
       "young     0.074600"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(IXI_metadata['AGE_GROUP'].value_counts(normalize=True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.647539Z",
     "start_time": "2020-11-12T10:11:05.429933Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.DataFrame(IXI_metadata['AGE_GROUP'].value_counts(normalize=True)).plot(kind='barh')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Sex"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.662125Z",
     "start_time": "2020-11-12T10:11:05.649471Z"
    }
   },
   "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>SEX</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.55595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.44405</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       SEX\n",
       "2  0.55595\n",
       "1  0.44405"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(IXI_metadata['SEX'].value_counts(normalize=True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.802253Z",
     "start_time": "2020-11-12T10:11:05.664084Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.DataFrame(IXI_metadata['SEX'].value_counts(normalize=True)).plot(kind='barh')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Ethnic\n",
    "Due to the big difference about 1 (white) and all others, we will categorize in 1 and others."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.817703Z",
     "start_time": "2020-11-12T10:11:05.803744Z"
    }
   },
   "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>ETHNIC_ID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ETHNIC_ID\n",
       "1        451\n",
       "3         50\n",
       "2         15\n",
       "5         14\n",
       "4         14\n",
       "6         10\n",
       "0          9"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "IXI_metadata['ETHNIC_ID'] = IXI_metadata['ETHNIC_ID'].astype('int32')\n",
    "pd.DataFrame(IXI_metadata['ETHNIC_ID'].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.832850Z",
     "start_time": "2020-11-12T10:11:05.819670Z"
    }
   },
   "outputs": [],
   "source": [
    "IXI_metadata.loc[IXI_metadata['ETHNIC_ID'] != 1, 'ETHNIC_ID'] = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:05.863640Z",
     "start_time": "2020-11-12T10:11:05.834818Z"
    }
   },
   "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>ETHNIC_ID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.801066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.198934</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ETHNIC_ID\n",
       "1   0.801066\n",
       "2   0.198934"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "IXI_metadata['ETHNIC_ID'] = IXI_metadata['ETHNIC_ID'].astype('category')\n",
    "pd.DataFrame(IXI_metadata['ETHNIC_ID'].value_counts(normalize=True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.036192Z",
     "start_time": "2020-11-12T10:11:05.865547Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.DataFrame(IXI_metadata['ETHNIC_ID'].value_counts(normalize=True)).plot(kind='barh')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Stratify by age and ethnic\n",
    "\n",
    "**We know that we have to stratify volumes in 90%-10% (train&val - test)**\n",
    "\n",
    "**We have already 18 non_metatada volumes in train&val set, so we have to stratify the remaining 563 volumes in 522 and 59.**\n",
    "\n",
    "**We defined a module which is able to stratify a DataFrame given the categories to stratify**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.051779Z",
     "start_time": "2020-11-12T10:11:06.038085Z"
    }
   },
   "outputs": [],
   "source": [
    "import stratifier_complex"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.067738Z",
     "start_time": "2020-11-12T10:11:06.053743Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(534, 47)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_and_val_vol, test_vols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.083017Z",
     "start_time": "2020-11-12T10:11:06.070698Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "563"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "IXI_metadata.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### How are going to be stratified?\n",
    "\n",
    "##### Test\n",
    "\n",
    "**Poblation and test sample distribution**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.114613Z",
     "start_time": "2020-11-12T10:11:06.084667Z"
    }
   },
   "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>ETHNIC_ID</th>\n",
       "      <th>AGE_GROUP</th>\n",
       "      <th>SEX</th>\n",
       "      <th>size</th>\n",
       "      <th>samp_size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>adult</td>\n",
       "      <td>1</td>\n",
       "      <td>143</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>adult</td>\n",
       "      <td>2</td>\n",
       "      <td>185</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>elderly</td>\n",
       "      <td>1</td>\n",
       "      <td>36</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>elderly</td>\n",
       "      <td>2</td>\n",
       "      <td>59</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>young</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>young</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2</td>\n",
       "      <td>adult</td>\n",
       "      <td>1</td>\n",
       "      <td>44</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2</td>\n",
       "      <td>adult</td>\n",
       "      <td>2</td>\n",
       "      <td>41</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2</td>\n",
       "      <td>elderly</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2</td>\n",
       "      <td>elderly</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2</td>\n",
       "      <td>young</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2</td>\n",
       "      <td>young</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ETHNIC_ID AGE_GROUP SEX  size  samp_size\n",
       "0          1     adult   1   143         12\n",
       "1          1     adult   2   185         15\n",
       "2          1   elderly   1    36          3\n",
       "3          1   elderly   2    59          5\n",
       "4          1     young   1    11          1\n",
       "5          1     young   2    17          1\n",
       "6          2     adult   1    44          4\n",
       "7          2     adult   2    41          3\n",
       "8          2   elderly   1     7          1\n",
       "9          2   elderly   2     6          1\n",
       "10         2     young   1     9          1\n",
       "11         2     young   2     5          0"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stratifier_complex.stratified_sample_report(IXI_metadata, ['ETHNIC_ID','AGE_GROUP','SEX'],test_vols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.304169Z",
     "start_time": "2020-11-12T10:11:06.116698Z"
    }
   },
   "outputs": [],
   "source": [
    "final_test_volumes = stratifier_complex.stratified_sample(IXI_metadata, ['ETHNIC_ID','AGE_GROUP','SEX'],test_vols)\n",
    "final_test_volumes.drop(columns=['index'], axis=1, inplace = True)\n",
    "final_test_volumes.shape\n",
    "\n",
    "# dump information to that file\n",
    "pkl.dump(final_test_volumes, open('data_test_volumes_df.pkl', 'wb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Train and validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.336050Z",
     "start_time": "2020-11-12T10:11:06.306073Z"
    }
   },
   "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>ETHNIC_ID</th>\n",
       "      <th>AGE_GROUP</th>\n",
       "      <th>SEX</th>\n",
       "      <th>size</th>\n",
       "      <th>samp_size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>adult</td>\n",
       "      <td>1</td>\n",
       "      <td>143</td>\n",
       "      <td>135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>adult</td>\n",
       "      <td>2</td>\n",
       "      <td>185</td>\n",
       "      <td>175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>elderly</td>\n",
       "      <td>1</td>\n",
       "      <td>36</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>elderly</td>\n",
       "      <td>2</td>\n",
       "      <td>59</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>young</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>young</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2</td>\n",
       "      <td>adult</td>\n",
       "      <td>1</td>\n",
       "      <td>44</td>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2</td>\n",
       "      <td>adult</td>\n",
       "      <td>2</td>\n",
       "      <td>41</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2</td>\n",
       "      <td>elderly</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2</td>\n",
       "      <td>elderly</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2</td>\n",
       "      <td>young</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2</td>\n",
       "      <td>young</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ETHNIC_ID AGE_GROUP SEX  size  samp_size\n",
       "0          1     adult   1   143        135\n",
       "1          1     adult   2   185        175\n",
       "2          1   elderly   1    36         34\n",
       "3          1   elderly   2    59         56\n",
       "4          1     young   1    11         10\n",
       "5          1     young   2    17         16\n",
       "6          2     adult   1    44         42\n",
       "7          2     adult   2    41         39\n",
       "8          2   elderly   1     7          7\n",
       "9          2   elderly   2     6          6\n",
       "10         2     young   1     9          9\n",
       "11         2     young   2     5          5"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stratifier_complex.stratified_sample_report(IXI_metadata, ['ETHNIC_ID','AGE_GROUP','SEX'],train_and_val_vol-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.351241Z",
     "start_time": "2020-11-12T10:11:06.339043Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(516, 6)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_train_val_volumes = IXI_metadata[~IXI_metadata['IXI_ID'].isin(final_test_volumes['IXI_ID'])]\n",
    "final_train_val_volumes.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* **47 test volumes: 8%**\n",
    "* **516+18=534 train and validation volumes: 92%**\n",
    "   * **454 Train volumes 78.2% of global**, 85% of train-val.\n",
    "   * **80 Validation volumes 13.8% of global**, 15% of train-val."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.444020Z",
     "start_time": "2020-11-12T10:11:06.353006Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(534, 6)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for i in MRI_with_no_data:\n",
    "    final_train_val_volumes = final_train_val_volumes.append({'IXI_ID':i,\n",
    "                                                             'SEX': None,\n",
    "                                                              'ETHNIC_ID':None,\n",
    "                                                              'AGE': None,\n",
    "                                                              'N_Relevant_Slices':None,\n",
    "                                                              'AGE_GROUP': None},ignore_index=True)\n",
    "final_train_val_volumes['IXI_ID'] = final_train_val_volumes['IXI_ID'].astype('int64')\n",
    "final_train_val_volumes.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.458769Z",
     "start_time": "2020-11-12T10:11:06.446788Z"
    }
   },
   "outputs": [],
   "source": [
    "pkl.dump(final_train_val_volumes, open('data_train_val_volumes_df.pkl', 'wb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Check Number of slices in each set\n",
    "##### Train&Val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.474681Z",
     "start_time": "2020-11-12T10:11:06.461718Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "59278"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_slices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.490727Z",
     "start_time": "2020-11-12T10:11:06.477972Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(54454, 91.86207361921792)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_slices-final_test_volumes['N_Relevant_Slices'].sum(), (total_slices-final_test_volumes['N_Relevant_Slices'].sum()) / total_slices *100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.506617Z",
     "start_time": "2020-11-12T10:11:06.492632Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "76.26168224299066"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "53184*0.85/total_slices *100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.522582Z",
     "start_time": "2020-11-12T10:11:06.509621Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13.457943925233645"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "53184*0.15/total_slices *100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T19:40:16.448606Z",
     "start_time": "2020-11-11T19:40:16.437624Z"
    }
   },
   "source": [
    "##### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.538609Z",
     "start_time": "2020-11-12T10:11:06.525544Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4824, 8.137926380782078)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_test_volumes['N_Relevant_Slices'].sum(), final_test_volumes['N_Relevant_Slices'].sum() / total_slices *100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclussion\n",
    "\n",
    "### Size of each set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:06.833093Z",
     "start_time": "2020-11-12T10:11:06.541517Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "IMGS: [46285.9, 8168.099999999999, 4824]\n",
      "VOLS: [453.9, 80.1, 47]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1,2, figsize=(15,8))\n",
    "print('IMGS:',[(total_slices-final_test_volumes['N_Relevant_Slices'].sum())*train_size,\n",
    "            (total_slices-final_test_volumes['N_Relevant_Slices'].sum())*validation_size,\n",
    "            final_test_volumes['N_Relevant_Slices'].sum()])\n",
    "\n",
    "print('VOLS:',[final_train_val_volumes.shape[0]*train_size,\n",
    "            final_train_val_volumes.shape[0]*validation_size,\n",
    "            final_test_volumes.shape[0]])\n",
    "\n",
    "axs[0].set_title('IMAGE split')\n",
    "axs[0].pie([(total_slices-final_test_volumes['N_Relevant_Slices'].sum())*train_size,\n",
    "            (total_slices-final_test_volumes['N_Relevant_Slices'].sum())*validation_size,\n",
    "            final_test_volumes['N_Relevant_Slices'].sum()],\n",
    "          autopct='%1.1f%%',\n",
    "          labels = ['Train', 'validation', 'test'])\n",
    "\n",
    "axs[1].set_title('Volume split')\n",
    "axs[1].pie([final_train_val_volumes.shape[0]*train_size,\n",
    "            final_train_val_volumes.shape[0]*validation_size,\n",
    "            final_test_volumes.shape[0]],\n",
    "          autopct='%1.1f%%',\n",
    "          labels = ['Train', 'validation', 'test']);\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T20:22:25.463461Z",
     "start_time": "2020-11-11T20:22:25.401590Z"
    }
   },
   "source": [
    "### Check attribute distributions in each set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:07.252567Z",
     "start_time": "2020-11-12T10:11:06.835715Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJoAAAFTCAYAAABxtmZdAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABFbUlEQVR4nO3de5xVVfn48c8jIIggKXghUMG74gUTSw0N09AyL6nhhQTNa5mZd8sbavT1lqaplZdCy2uaVlreUswriooZoqGIij8UQUVRFNH1+2MdmMMwwMywZ+acmc/79dqvOXvvtdd+9oE5a/Zz1lo7UkpIkiRJkiRJS2uZlg5AkiRJkiRJrYOJJkmSJEmSJBXCRJMkSZIkSZIKYaJJkiRJkiRJhTDRJEmSJEmSpEKYaJIkSZIkSVIhTDRJkiRJkiSpECaaJElqoyJidES8GxEda20fFRE/X8QxKSLWqUfdIyLi04iYFRHvRcSjEbF12f7eEXFdRMyIiA8j4omI+HYd5/qwVMcbEXFhRLRbVBylc/6p9HpQRExZQoyDSvWcWFpfo3SueUv5+WdFxLa135uI6BgR/xcRr0XE7IiYGBEnRETUep8/jojVy7btGBGTl/Q+SpIkVRsTTZIktUER0QfYFkjAbk10mptSSl2AlYGHgb9EtlJpfQ7QD+gBXARcHxF716pjs1IdOwD7A4cWGN9w4J3ST1JKr6WUusxbys9fWh6qo44/l2L7FtAVOAA4DLi4VrkPgdMKjF2SJKkimWiSJKltGgY8DoyilGhpKimlT4FrgNWA7sAxwCzg4JTSmyml2SmlG4CRwC/LewOV1fEC8BCwcRExRURnYG/gSGDdiBjQiDp2AAYDe6WU/ptSmptSehz4HnBkrR5XlwD71ac3mCRJUjUz0SRJUts0DLiutOwUEas21YlKQ/MOBKaklKYD3wBuTSl9XqvozcAawHp11LERuQfWMwWFtRc52fVn4G7y+9FQ3wDGpJReL9+YUhoDTCH3dJrnDeBKYERjgpUkSaoWJpokSWpjImIgsCZwc0rpKeBl8rC0og2JiPeA14EtgD1K23sAU+soP7Vs/zxPR8S7wN+Bq4A/FBTbcPLQvs+A68m9jTo0sI5FXQel7T1qbfs/YNeI6NfA80iSJFUNE02SJLU9w4F7Sr2LICdammL43M0ppS+klFZJKX29lNQCmA70rKN8z7L983wppbRiSmntlNKpZb2gPgNqJ4Y6AJ8uKajSpNzbk3tzAfwV6ATssuRLWsCiroPS9vLrIKX0NnApcFYDzyNJklQ1TDRJktSGRMRywBDgaxHxZkS8SZ4zabOI2KyZwrgP2Csiav8dMoTc++l/9ajjNaBPrW19gVfrcewB5L+B/l66/knkRFNDh8/dB3yl/GlyABHxZWB14P46jjmfnOTaooHnkiRJqgommiRJalv2IPcG2gjoX1o2JE+0XZ5oaRcRncqWZcv2LVtrX7sGxnARsAJwdUSsVqpjP+AU4ISUUqpHHTcBp0ZE74hYJiJ2BHYFbikvVCvOTqWJxocBZ5Zdf3/ynE27RET3+l5ESuk+4F/ArRHRLyLaRcRW5J5Sv0kpTazjmPeAXwIn1vc8kiRJ1cREkyRJbctw4A8ppddKT3x7M6X0JnlI19CIaF8qdzIwu2wp750zvta+gxoSQEppBjCQ3IvoeWAGcCxwQErppnpWcxbwKPAw8C5wHjA0pfTfsjK9asU5G/gquSfUZeXXn1L6G/ASsF9DroWcoHoAuIs8ufifgKuBoxZzzMXkZJ8kSVKrE/X70lCSJEmSJElaPHs0SZIkSZIkqRAmmiRJkiRJklQIE02SJEmSJEkqhIkmSZIkSZIkFcJEkyRJkiRJkgphokmSJEmSJEmFMNEkSZIkSZKkQphokiRJkiRJUiFMNEmSJEmSJKkQJpokSZIkSZJUCBNNkiRJkqQ2JyIGRsSjETEzIt6JiEciYsuIODAiPouIWbWWL0ZEl4iYHBH7l9XTNSJei4i9W/J6pEphoknNovRhPLvWB/XnZa8/rvVhPr50XIqIdWrVNSIi/lS2niLiuYhYpmzbzyNiVOl1n1KZ9mX7vxwR/4iI90qNyhMRcdASrmFQREwpWx9divuDiHg/Ip6KiJMjouNSv2GS1AaUPkffrf25GRH7RsSYiPgwIqaVXv8wIqK0f1REzKnVpjxbj/MtGxGnR8SLpbrfiIh/RsTgsjLl7dWbpXN1qVXPNhFxf+nzf2ZE/D0iNirbf2BEPFzH+SdHxI51XMM7EXFvRGzQ8HdRktQYEbECcAfwa2AloBdwJvBJqchjKaUutZb/l1KaBRwGXBwRK5fKngeMTSnd0syXIVUkE01qTrvW+qBeZt5r4AgW/DDv18C6vwjsW5+CEbE1cD/wILAO0B34AfDNBp4T4Ecppa5AT+C4Ugz/mHczJEmqW0T0AbYFErBb2fbjgIuB84HVgFXJbcRXgWXLqjivVpuyWT1OewuwOzAMWBHoWzrXLrXK7Vpqm/oDmwM/LYtva+Ae4K/ktqcv8CzwSESsVZ9rr30NQG9gGjCqgcdLkhpvPYCU0g0ppc9SSrNTSveklP6zpANTSvcAdwKXRMQgYAhwZFMGK1UTE01qLc4DzizvtbQY5wPXpJTOTSlNT9lTKaUhjT15SunDlNJo8s3S1ix80yJJWtAw4HFycmU4QER0A84CfphSuiWl9EHpM/qZlNLQlNIni65u8Uo9ib4B7J5SGpNSmlNa7kopHV3XMSmlN4G7yQmnec4Drk0pXVyK752U0qmlaxnRmNhSSh8B1wMbN+Z4SVKj/A/4LCKuiYhvRsSKDTz+GGAQ+UuM41NKU4sOUKpWJprUWvwFeB84cHGFIqIzORHUJN1aU0qvAWPJ39JLkhZtGHBdadkpIlYlfz53JPcWKtqOwJiU0pQlliyJiN7k3q4vldY7A9sAf66j+M3kRFaDlYbmDQWeaczxkqSGSym9Dwwk96y9Eng7Iv5Wao8AtipNszFvebnW8e8C44HO5HsRSSUmmtScbq/1YX1oPY97uvw44OQ6yiTgNOD0JcyRtCL5/31TfuPw/8jjvCVJdYiIgcCawM0ppaeAl4H9gR7A9JTS3LKyj5Y+/2dHxHZl1Rxfq025Zgmn7QG8WVbvSqXjZkbEx7XK3h4RHwCvk4e0nVHavhKLbkOmls7REMeX2rWXgC4s4csSSVKxUkoTUkoHppR6k3uVfhH4VWn34ymlL5Qta5cfGxHfA/oA9wHnNmPYUsUz0aTmtEetD+sr63ncl8qPA86pq1BK6R/Aa+TJ+RblXeBz8pxKTaUX8E4T1i9J1W44cE9KaXpp/frSthlAj/Jh0CmlbUqf/TNY8O+WC2q1KcOXcM4ZlH32l4a8fQHYgtyLqtwepfn3BgEbUJNAWlwb0hOYdz1zgQ51lOkAfFrHNayWUtotpfRyHcdIkppBSukF8nDuJQ5jjohVgIuAQ4HDgSG1vgyR2jQTTWptTgVOIXdhXUhpHozHgL2a4uQRsTr5puWhpqhfkqpdRCxHnjT1a6Wnur1JnudiM+Aj8tN+dm+CU/8L2LI0HK5eUkoPkm86Liitf0huQ75bR/EhpXNA/tJjjfIHQ5SG3a0CvNqY4CVJxYqIDSLiuHntQunv+P3Ic+4tyaXA7SmlB0pzM50IXOnTp6XMRJNaldKE3M9Rmlh2EU4EDoyIEyKiO0BEbBYRNzb2vBHROSK+Rp5X5AngH42tS5JauT2Az4CNyJNs9wc2JCfodyM/WvryiNg7IrpExDIR0R9YfmlOWnpC0APkYXFfiYhlI6IDsNUSDv0V8I1SDJCHbw+PiB9HRNeIWDEifk6eX+rMUpkxwMfAyRHRKSKWJ/fGHYuJJkmqFB8AXwHGRMSH5ATTf8lPkgbYOiJm1Vq2jIg9yHM7nTCvopTSVcAU4PRmvQKpQploUnP6e60P6tua6Dynspg5klJKjwJfLy2TIuId4Aoalxy6tDSPx1vkm5FbgZ1TSp83oi5JaguGA39IKb2WUnpz3kL+dngocCFwLPlLgWnkz9ffAScBj5bVc2KtNmU6S7YncAfwJ+A94JXSOXde1AEppbeBa8nzAJJSehjYqVTXVHLiaHNgYEppYqnMJ+Snjw4i33hMIs/7MSSllOoRpySpiaWU3kgpDUkp9UopLV/6eXhK6f2U0qiUUruUUpday5MppdtTSl9MKb1Xq74dUkqntNDlSBUl/HtHkiRJkiRJRbBHkyRJkiRJkgphokkqExE/q2Ms9qyI+GdLxyZJWryIGLqIz/DxLR2bJElSW+HQOUmSJEmSJBXCHk2SJEmSJEkqRPumqLRHjx6pT58+TVG1JFW1p556anpKaeWWjqMl2UZI0qLZTthOSNLiVEM70SSJpj59+jB27NimqFqSqlpEvNrSMbQ02whJWjTbCdsJSVqcamgnHDonSZIkSZKkQphokiRJkiRJUiFMNEmSJEmSJKkQTTJHk6TK9+mnnzJlyhQ+/vjjlg6lVerUqRO9e/emQ4cOLR2KJC3ENqDl2U5I0tJrze1ZNbcTJpqkNmrKlCl07dqVPn36EBEtHU6rklJixowZTJkyhb59+7Z0OJK0ENuAlmU7IUnFaK3tWbW3Ew6dk9qojz/+mO7du7eqD+RKERF07969VX6zIql1sA1oWbYTklSM1tqeVXs7YaJJasNa2wdyJfG9lVTp/JxqWb7/klSM1vp5Ws3XZaJJkiRJkiSpkUaOHEm/fv3YdNNN6d+/P2PGjGHQoEGsv/769O/fn/79+7P33nsD8OMf/5izzz57gWOPPPLIlgq9SThHkyQA+px8Z6H1TT5nlyWWadeuHZtsssn89X333ZcxY8bwyiuvMGvWLN5+++35Y5Ivv/xyfvazn3HBBRcwYMCAfI7Jk/n2t7/Nf//7X0aPHs3222/P3/72N3bddVcAvv3tb3P88cczaNAgBg0aNP/YWbNmcdxxx3HffffRqVMnunfvzvnnn89XvvKVOuPs0qULs2bNYvLkyWy44YZssMEGfPzxx3Tt2pUjjzyS4cOHL+3bJUktqiXaAIDbbruNPffckwkTJrDBBhsA8MQTT3DiiSfyxhtv0LVrV3r27Mk555zDJptswogRI7jyyitZeeWV59cxevRovvCFL9RZf33rmjNnDqeddhr77bcfkOfGGDlyJNdccw0RQa9evbj00kvp168fUNMuzDNq1CjGjh3LpZdeukC9c+fO5Re/+AW77bZbY95GSVIDtUR79thjj3HHHXfw9NNP07FjR6ZPn86cOXMAuO666+bfu8zz85//nP79+zN06FAigquuuopnnnmm0LhbmokmSS1mueWWY9y4cXXuGz16NBdccAF33HFHvevr3bs3I0eOnJ9oWpRDDjmEvn37MnHiRJZZZhkmTZrEhAkT6nWOtddee35DMGnSJPbcc08+//xzDjrooHrHKUnKbrjhBgYOHMiNN97IiBEjeOuttxgyZAjXX38922yzDQAPP/wwL7/88vwvJo455hiOP/74JdbdkLomTpzIFltswd57702HDh247LLLePTRR3n22Wfp3Lkz99xzD7vtthvjx4+nU6dOSzz3vHonTJjAtttuy7Rp01hmGQcSSFJrNHXqVHr06EHHjh0B6NGjx2LLr7DCCowcOZIf/ehHAJx11lmL/MKkWtniSWo1NttsM7p168a99967yDIvv/wyY8aM4ec///n8P/rXWmstdtmlft++l1trrbW48MILueSSSxodsyS1VbNmzeKRRx7h6quv5sYbbwTg0ksvZfjw4fMTQwADBw5kjz32aHD9Dalr3XXXpXPnzrz77rsAnHvuufz617+mc+fOAAwePJhtttmG6667rkExbLjhhrRv357p06c3OH5JUnUYPHgwr7/+Ouuttx4//OEPefDBB+fvGzp06PyhcyeccML87fvttx/vvvsu77//PgcccEBLhN2k7NEkqcXMnj2b/v37z1//6U9/yj777LPYY4YOHcpyyy0HwJw5cxb6hvjUU0/l1FNP5Rvf+Eadx48fP57+/fvTrl27pQu+5Etf+hIvvPBCIXVJUlty++23s/POO7Peeuux0kor8fTTTzN+/PglDke+6KKL+NOf/gTAiiuuyAMPPFBnufrUNc/TTz/NuuuuyyqrrML777/Phx9+yNprr71AmQEDBjB+/Ph61TfPmDFjWGaZZRYY6idJal26dOnCU089xUMPPcQDDzzAPvvswznnnAPUPXQOYMqUKbz55ptEBLNmzaJLly7NHXaTMtEkqcUsbujcopR/WM+bo6nctttuC8BDDz1USIxLklJqlvNIUmtzww038JOf/ATIc/TdcMMNC5X5yle+wvvvv8/gwYO5+OKLgfoPnatPXRdddBFXXnklkyZN4q677lrs8SmlxT4BqHzfvGRY165duemmm6r6yUGSpCVr167d/HlhN9lkE6655prFlj/66KMZMWIEEyZM4Mwzz+T8889vpkibh0PnJLU6p5xyCiNHjqxzX79+/Xj22Wf5/PPPCznXM888w4YbblhIXZLUVsyYMYP777+fQw45hD59+nD++edz00030a9fP55++un55caMGcPZZ5/NzJkzG3yO+tR1zDHH8OKLL3LTTTcxbNgwPv74Y1ZYYQWWX355Jk2atEB9Tz/9NBtttBGQvyiZN9ErwDvvvLPAnBzHHHMM48aN46GHHpr/BYgkqXV68cUXmThx4vz1cePGseaaay6y/D//+U+mTZvGsGHDOO2007jtttt4/vnnmyPUZmOiSVKrM3jwYN59912effbZhfatvfbaDBgwgDPOOGN+b6SJEyfy17/+tcHnmTx5MscffzxHHXXUUscsSW3JLbfcwrBhw3j11VeZPHkyr7/+On379mXw4MGMGjWKRx99dH7Zjz76qFHnOPLII+td15577smAAQPmfwN9wgkn8OMf/5jZs2cDcN999/Hwww+z//77A/C1r31t/vC92bNnc/PNN7P99ts3Kk5JUnWbNWsWw4cPZ6ONNmLTTTfl+eefZ8SIEcCCczTtuOOOfPzxx/zkJz/h8ssvJyJYfvnlOe+88+ZPDN5aOHROElD/R1EXqfYcTTvvvPP88cxL65RTTmH33Xevc99VV13FcccdxzrrrEPnzp3p3r17vburvvzyy2y++eZ8/PHHdO3alaOOOsonzkmqes3dBtxwww2cfPLJC2zba6+9uP7667nppps46aSTeOONN1hllVXo0aMHp59++vxy5XM0QZ7rqU+fPgudY7XVVltiXeVOP/109t9/fw499FCOOuoo3n33XTbZZBPatWvHaqutxl//+tf5cwRefPHFHH744VxyySWklBg2bBjbbbddAe+MJGlptMQ9zRZbbLHAlxrzjB49us7yL7744gLre+65J3vuuWdThNZioinmFxkwYEAaO3Zs4fVKKs6ECRMc8tXE6nqPI+KplNLCMwK2IbYRUsuzDagMthN1s52QVF+tvT2r1nbCoXOSJEmSJEkqhEPnJIk8Me0OO+yw0PZ//etfdO/evQUikiTVx913381JJ520wLa+ffty2223tVBEkiS1bSaaJAno3r0748aNa+kwJEkNtNNOO7HTTju1dBiSpBaSUiIiWjqMwjXFNEfNxaFzUhtWzR9elc73VlKl83OqZfn+S9LS69SpEzNmzGh1n6kpJWbMmEGnTp1aOpRGsUeT1EbN+1Du3r17q/wGoCVVe8PQ1J57YyZ9Tr6zxc7fEk8jkSqNbUDLsp2QpGL07t2bKVOm8Pbbb7d0KIXr1KkTvXv3bukwGsVEk9RGteYP5UpQzQ2DpNbPNqDl2U5I0tLr0KEDffv2bekwVIuJJqmN8kNZktou2wBJktRUnKNJkiRJkiRJhTDRJEmSJEmSpEKYaJIkSZIkSVIhTDRJkiRJkiSpECaaJEmSJEmSVAgTTZIkSZIkSSqEiSZJkiRJkiQVwkSTJEmSJEmSCtG+KSp97o2Z9Dn5zqaoutWafM4uLR2CJEmS1OK8l5BaP+9/Wzd7NEmSJEmSJKkQJpokSZIkSZJUCBNNkiRJkiRJKoSJJkmSJEmSJBXCRJMkSZIkSZIKYaJJkiRJkiRJhTDRJEmSJEmSpEKYaJIkSZIkSVIhTDRJUisWEWdHxNFl6yMj4uiIOD8i/hsRz0XEPqV9gyLijrKyl0bEgaXXkyPizIh4unTMBqXtK0fEvaXtv4uIVyOiRzNfpiRJkqQKYaJJklq3q4HhABGxDLAvMAXoD2wG7AicHxE961HX9JTSl4DfAMeXtp0B3F/afhuwRqHRS5IkSaoqJpokqRVLKU0GZkTE5sBg4BlgIHBDSumzlNJbwIPAlvWo7i+ln08BfUqvBwI3ls51F/BuXQdGxGERMTYixn720cxGXo0kSZKkSmeiSZJav6uAA4GDgN8DsYhyc1mwXehUa/8npZ+fAe1LrxdV1wJSSleklAaklAa069ytPodIkiRJqkImmiSp9bsN2Jnca+lu4N/APhHRLiJWBrYDngBeBTaKiI4R0Q3YoR51PwwMAYiIwcCKTRC/JEmSpCrRfslFJEnVLKU0JyIeAN5LKX0WEbcBWwPPAgk4MaX0JkBE3Az8B5hIHma3JGcCN5QmFH8QmAp80ASXIUmSJKkK2KNJklq50iTgW5EnBidlJ6SUNk4pbZJSumle2ZTSiSml9VNK304p7ZlSGlXa3ielNL30emxKaVDpkJnATqXJwG8GpqWUPqEa9ekDEQsvu+yy5GMnToSuXaFLl4X3zZkDp58OfftCx46wxhpwySU1+++9F9ZbD1ZYAQ44IJefZ9YsWHddGD9+qS9PkiRJag4mmiSpFYuIjYCXgH+llCY2wSnWAJ6MiGeBS4BDm+AczePJJ2Hq1Jrl6adzomnIkMUfN2cO7LsvbLdd3fv32w/uuguuuAJefBH+/GfYdNO87/PPYehQOOIIeOwxGDs2l5vn1FNz3f36FXONkiRJUhNz6JwktWIppeeBtZqw/onA5k1Vf7NaeeUF16++Ovcy+u53F3/cSSflxNHXvgYPPrjgvnvugfvug5dfhh498rY+fWr2T58Ob78NP/whdOoEu+0GEybkfU88kY9/pj4jGCVJkqTKYI8mSZJqSyknmr73PejcedHl7rwT7rhjwaFw5W6/HbbcEi68EHr3zsPgfvzjPCQOcnKrZ8+cUJo9Gx56KCet5s6Fww+H3/wmD7eTJEmSqoSJJkmSarv3XnjlFTjkkEWXmToVDj0U/vjHPD9TXSZNgocfhmefhVtvhUsvzcPoDjww74+Am2+Gs8+GjTaCzTeH738fzj8/J6hWXTUPyVt3XRgxouirlCRJkgrn0DlJkmq78sqc6Onff9Flvvc9+MEPYKutFl3m889zMun666Fbt7zt0kthp53grbdyImngwDw/1DwvvZTnaXrmGdhxx3yOIUNyPFtuWb/JySVJkqQWYo8mSZLKTZsGf/1r7q20OPffD2eeCe3b5+Xgg+HDD/PreRN69+wJvXrVJJkANtww/3zttbrrPfxwOO88WGYZeOqpPBl4166w6675nJIkSVIFs0eTJEnl/vCHPC/Svvsuvtxzzy24/te/wsiReRLvXr3ytq9+NT9lbtYs6NIlb/vf//LPNdes+9zLL58nIH/vvbzt00/zzzlzcu8oSZIkqYLZo0mSpHlSgquuqulFVO6nP4UddqhZ33jjBZdevXIvpI03hhVXzGX23x+6d4eDDoLx4+GRR+Doo2HvvWGVVRasf9q03EPqssvy+he+AP36wS9/mYfR3XJLHmYnSZIkVTATTZIkzTN6dJ4jqa5hc1OnwssvN6y+Ll3gvvtg5sw8v9KQIfC1r8Hvf79w2aOPhuOOg9VXr9l2zTX5yXXbbw977ZUXSZIkqYI5dE6SpHm23z73aqrLqFGLP/bAA2ueJldu/fXhnnuWfO4bblh42xZbLDxET5IkSapg9miSJEmSJElSIUw0SZIkSZIkqRAmmiRJkiRJklQIE02SJEmSJEkqxBITTRGxekQ8EBETImJ8RBzdHIFJkiRJqn4R8fuImBYR/23pWCRJTa8+PZrmAsellDYEtgKOjIiNmjYsSZIkSa3EKGDnlg5CktQ8lphoSilNTSk9XXr9ATAB6NXUgUmSJEmqfimlfwPvtHQckqTm0aA5miKiD7A5MKZJopEkSZIkSVLVal/fghHRBbgV+ElK6f069h8GHAbQboWVCwtQktS6bNKrG2PP2aWlw5AkVRDvJSSp9ahXj6aI6EBOMl2XUvpLXWVSSleklAaklAa069ytyBglSZIktWLeS0hS61Gfp84FcDUwIaV0YdOHJEmSJEmSpGpUnx5NXwUOAL4eEeNKy7eaOC5JkiRJrUBE3AA8BqwfEVMi4uCWjkmS1HSWOEdTSulhIJohFkmSJEmtTEppv5aOQZLUfBr01DlJkiRJkiRpUUw0SZIkSZIkqRAmmiRJkiRJklQIE02SJEmSJEkqhIkmSZIkSZIkFcJEkyRJkiRJkgphoqnS/fvfsNtu0KsXRMCoUQvuf+stOPBA+OIXoXNn2HlnmDhxyfVefz3075+PWW01+N734M03a/Z/+imcdRasvTZ06gSbbQZ33bVgHdddB6uvDiutBMceu+C+N96APn1yfJIkSZIkNdQvfpHvg3/0o7r3H3ZY3n/BBfWv8+GHoX172HjjBbffey+stx6ssAIccADMmVOzb9YsWHddGD++4dfQBploqnSzZuVfgIsvhuWWW3BfSrDHHjmxdPvt8MwzsOaasOOO8OGHi67zkUfyL87w4fkX5fbb4fnnYejQmjKnngq//S1cckned8QR8J3v5HMATJ8OhxySf6Hvvhv+9Ce4446a4488Ek47DVZdtaA3QpIkSZLUZjz+OFx5JWy6ad37b7kFnnwyd7qor3ffhWHDYIcdFtz++ef5fviII+Cxx2DsWLjiipr9p54K++4L/fo1/DraIBNNle5b38pZ3L33hmVq/XNNnJh/+S6/HL78ZVh/ffjNb2D2bLjhhkXX+dhj0Ls3HHMM9O0LW20FRx0FY8bUlPnjH+Gkk2CXXWCtteAHP8ix/PKXef+kSdCtG+yzD2y5JWy/PUyYkPfdeivMnAnf/36x74UkSZIkqfWbOTMnfq6+GlZcceH9r74KRx+dR+p06FD/eg8+OHe42HrrBbdPnw5vvw0//GFOJu22W8397RNPwD335GST6sVEUzX75JP8s1Onmm3LLAMdO+bugIvy1a/C1Knw97/nXlHTp8ONN+ZEUnnd5fVC7lE1r95114WPPso9nN55J2eSN900fyCccAL87ne5C6MkSZIkSQ1x2GG5s8XXv77wvrlzYb/9cuJnww3rX+fll+fpYupKGK28MvTsmRNKs2fDQw/l+9u5c+Hww3OHjo4dG389bYyJpmq2wQZ5qNzPfpaTPXPmwLnnwpQpOZG0KFtvnXs8DR0Kyy6bf6lSgmuuqSmz007wq1/Biy/mboT33gt/+UtNvSuumMsPG5Z7Uw0blo858cQ8pG76dBgwIP/i//a3Tfo2SJIkSZJaiSuvhJdegrPPrnv/GWdA9+551E19PfccnHlmnme4XbuF90fAzTfnc260EWy+eR6hc/75eQTPqqvCdtvlDhcjRjTqstqS9i0dgJZChw55mNrBB+dftHbt8vxM3/zm4o97/nn48Y/zHEo77ZSTRyeckDO1116by1x8MRx6aP4li8iTgh90EPzhDzX1fOc7eZnn4YfzUL5f/jIP47vmmtztcNNNcy+qTTYp/j2QJEmSJLUOL76YO1I89FDuFFHbgw/mB2SNG1f/Oj/5JM+vdMEFeeqYRRk4MI/Umeell/I8Tc88k++zf/ADGDIkJ5623DJPM6M62aOp2m2xRf4le++9nDC66y6YMWPxv0D/93+5F9IJJ+Qk0E475W6Ef/wjvP56LrPyynmS8A8/zONfX3gBunRZdL1z5uSJ0373uzx/05w5+ZexZ08YNAhGjy70siVJkiRJrcxjj+XRMRtvnJ8M1759Ti5dfnl+fddd+b63Z8+a/a++mucX7t277jqnTs2dLQ46qOaYs87KD8Zq3z4Pl6vL4YfDeefl6Wmeeionq7p2hV13hfvvb7r3oBWwR1Nr0a1b/jlxYp4hf1HdDCHPrVS7u+C89ZQW3N6pE/TqBZ9+mntPDRlSd50jR+bxs1ttlRNfc+fW7JszBz77rEGXI0mSJElqY/bYI0/BUu6gg/KQtZ/9LHeIKH9aOuSOE/vtl0fk1KVXrzx0rtzll+fpYW67Dfr0WfiYP/wBll8evvvd3KkD8j0x5Ptb5yNeLBNNlW7WrNxlD/JcSa+9lhM5K60Ea6wBf/4z9OiR52p67rk88/4ee8DgwTV1DBuWf84bFrfrrvmX8De/qRk695OfwJe+lOuE/AS6N96A/v3zzxEj8vlPPHHhGJ9/Po91feaZvL7++jkz/Nvf5qFz//pXHqYnSZIkSdKifOELeSm3/PL5/nfjjfP6qqsuuL9DB1httXwfOk/5PXCHDjXHzrPKKnly79rbAaZNy/M5PfRQTUz9+uUpYvbcE265JU81o0Uy0VTpxo6F7bevWT/jjLwMH57Hpk6dCsceC2+9lbsPDhu2cFLntdcWXD/wQPjgA7j0UjjuuNwbavvtc7fAeT7+OM/GP2lSHjL3rW/loXW1f+lTyk8EuOii3I0Q8tPp/vhHOPLI/BS6U05ZOCstSZIkSVJTqH0P3BBHH53vk1dfvWbbNdfk++hf/zrfc++111KH2JpFqj1UqgAde66beg7/VeH1tmaTz3EiMaktiIinUkptOvM6YMCANHbs2JYOQ5Iqku2E9xJSW+D9b+NVQzvhZOCSJEmSJEkqhIkmSZIkSZIkFcJEkyRJkiRJkgphokmSJEmSJEmFMNEkSZIkSZKkQphokiRJkiRJUiFMNEmSJEmSJKkQJpokSZIkSZJUCBNNkiRJkiRJKoSJJkmSJEmSJBXCRJMkSZIkSZIKYaJJkiRJkiRJhWjfFJVu0qsbY8/ZpSmqliRJktSKeS8hSdXNHk2SJEmSJEkqhIkmSZIkSZIkFcJEkyRJkiRJkgphokmSJEmSJEmFMNEkSZIkSZKkQphokiRJkiRJUiFMNEmSJEmSJKkQJpokSZIkSZJUCBNNkiRJkiRJKoSJJkmSJEmSJBWifUsHIElqW557YyZ9Tr6zpcOoepPP2aWlQ5AkSZIWYo8mSZIkSZIkFcJEkyRJkiRJkgphokmSJEmSJEmFMNEkSZIkSZKkQphokiRJkiRJUiFMNEmSJEmSJKkQJpokSZIkSZJUCBNNkiRJkiRJKoSJJkmSJEmSJBXCRJMkSZIkSZIKYaJJkiRJkiRJhTDRJEmSJEmSpEKYaJIkSZIkSVIhTDRJkiRJkiSpECaaJEmSJEmSVAgTTZIkSZIkSSqEiSZJkiRJkiQVwkSTJEmSJEmSCmGiSZLaiIiYHBE96tg+IiKOb2BdB0bEpcVFJ0mSJKk1MNEkSWqQiGjf0jFIkiRJqkwmmiSpFYqI70XEExExLiJ+FxHtau0/JSJejIj7gPXLtq8dEXdFxFMR8VBEbFDaPioiLoyIB4Bzy8p3jYhXIqJDaX2FUs+pDs1zpZIkSZIqiYkmSWplImJDYB/gqyml/sBnwNCy/VsA+wKbA3sCW5YdfgVwVEppC+B44PKyfesBO6aUjpu3IaX0ATAa2KW0aV/g1pTSp8VelSRJkqRq4PAHSWp9dgC2AJ6MCIDlgGll+7cFbkspfQQQEX8r/ewCbAP8uXQcQMey4/6cUvqsjvNdBZwI3A4cBBxau0BEHAYcBtBuhZUbeVmSJEmSKp2JJklqfQK4JqX00wU2RhxYtprqOG4Z4L1SL6i6fFjXxpTSIxHRJyK+BrRLKf23jjJXkHtL0bHnunWdW5IkSVIr4NA5SWp9/gXsHRGrAETEShGxZtn+fwPfiYjlIqIrsCtASul94JWI+G7puIiIzep5zmuBG4A/FHURkiRJkqqPiSZJamVSSs8DpwL3RMR/gHuBnmX7nwZuAsYBtwIPlR0+FDg4Ip4FxgO71/O01wErkpNNkiRJktooE02S1AqllG5KKfVPKW2aUtoipfR4SqlPSml6af/IlNL6KaXBKaXvp5QuKG1/JaW0c0pps5TSRimls0rbD0wp3VJW/6iU0o/KTjkQuCWl9F4zXqbq8u9/w267Qa9eEAGjRi24/6234MAD4YtfhM6dYeedYeLExdc5dSrsvz9ssAG0a5ePr8vFF+cyyy0HvXvDkUfCrFk1+6+7DlZfHVZaCY49dsFj33gD+vTJ8UmSJKlqmWiSJC2ViPg1cA5wdkvHInJiZ+ONc9JnueUW3JcS7LFHTizdfjs88wysuSbsuCN8WOcUXNknn0CPHnDyyfCVr9Rd5vrr4cQT4ZRTYMIEuPZa+Mc/4Oij8/7p0+GQQ+CCC+Duu+FPf4I77qg5/sgj4bTTYNVVl+bqJUmS1MKcDFyStFRSSke1dAwq861v5QUW7nk0cSI8/jiMGweblabf+s1vYLXV4IYbciKoLn36wCWX5Ne33FJ3mUcfha22ggMOqDlm2DC49da8PmkSdOsG++yT17ffPiekvv3tXGbmTPj+9xt+vZIkSaoo9miSJKmt+OST/LNTp5ptyywDHTvCww8vXd0DB+YE1uOP5/XXXoO//a0m6bXuuvDRR7kX1TvvwJNPwqab5gTTCSfA736Xh/pJkiSpqplokiSprdhggzxU7mc/y8meOXPg3HNhypQ8D9PS2Hdf+MUvYLvtoEOHfJ5NNsn1A6y4IlxzTe7l9OUv55877ZSH2x1ySB5aN2AAbLgh/Pa3S3+tkiRJahEOnZMkqa3o0CEPUzv4YOjePU/sveOO8M1vLn3dDz4IZ58Nl1+e53F66aU8P9MZZ8BZZ+Uy3/lOXuZ5+OHcA+qXv4T118+JqH79ck+nr341J6okSZJUVUw0SZLUlmyxRR7iNnNm7tG08so5MTRgwNLVe+qpsN9+NfM8bbJJnmD8kEPg9NOhfa0/OebMgSOOgKuuyvM3zZmTk14AgwbB6NEmmiRJkqqQQ+ckSWqLunXLSaaJE2HsWNh996Wr76OPcg+pcu3a5Sfd1WXkSPj61/ME4p9/DnPn1uybMwc++2zp4pEkSVKLsEeTJEmtyaxZedga5ATOa6/lHkwrrQRrrAF//jP06JHnUHruuTy8bY89YPDgmjqGDcs/r722Ztu4cfnn++/nCcTHjYNll4WNNsrbd90VLrww94yaN3TutNPyU+Vq92Z6/nm47ro8MTjkYXPt2+e5mfr1g3/9Kx8rSZKkqmOiSZKk1mTsWNh++5r1M87Iy/DhMGpUnvT72GPhrbegZ8+cVKqd1HnttYXr3XzzBdf//vecrJo8Oa+femp+atxpp+XJxXv0yMmnkSMXPC4lOOwwuOgi6No1b1tuOfjjH+HII/OQvlNOWfqhfJIkSWoRkRbVpX0pDBgwII0dO7bweiWp2kXEUymlNn0H3bHnuqnn8F+1dBhVb/I5u7R0CJKagO2E9xKStDjV0E44R5MkSZIkSZIKYaJJkiRJkiRJhTDRJEmSJEmSpEKYaJIkSZIkSVIhTDRJkiRJkiSpECaaJEmSJEmSVAgTTZIkSZIkSSqEiSZJkiRJkiQVon1LByBJals26dWNsefs0tJhSJIkSWoC9miSJEmSJElSIUw0SZIkSZIkqRAmmiRJkiRJklQIE02SJEmSJEkqhIkmSZIkSZIkFcJEkyRJkiRJkgphokmSJEmSJEmFMNEkSZIkSZKkQphokiRJkiRJUiFMNEmSJEmSJKkQ7Zui0ufemEmfk+9siqolqUVNPmeXlg5BkqRWzXsJSa1ZW7ifsEeTJEmSJEmSCmGiSZIkSZIkSYUw0SRJkiRJkqRCmGiSJEmSJElSIUw0SZIkSZIkqRAmmiRJkiRJklQIE02SJEmSJEkqhIkmSZIkSZIkFcJEkyRJkiRJkgphokmSJEmSJEmFMNEkSZIkSZKkQphokiRJkiRJUiFMNEmSJEmSJKkQJpokSZIkSZJUCBNNkiRJkiRJKoSJJkmSJEmSJBXCRJMkSZIkSZIKYaJJkiRJkiRJhTDRJEmSJEmSpEKYaJIkSZIkSVIhTDRJkiRJkiSpECaaJEmSJEmSVAgTTZIkSZIkSSqEiSZJkiRJkiQVwkSTJEmSJEmSCmGiSZIkSZIkSYUw0SRJkiRJkqRCmGiSJEmSJElSIUw0SZIkSZIkqRBLTDRFxO8jYlpE/Lc5ApIkSZLUekTE6hHxQERMiIjxEXF0S8ckSWo69enRNArYuYnjkCRJktQ6zQWOSyltCGwFHBkRG7VwTJKkJrLERFNK6d/AO80QiyRJkqRWJqU0NaX0dOn1B8AEoFfLRiVJairO0SRJkiSpWUREH2BzYEwLhyJJaiLti6ooIg4DDgNot8LKRVUrSWplnntjJn1OvrOlw5CkJjH5nF1aOoSKFRFdgFuBn6SU3q+1z3sJSWolCuvRlFK6IqU0IKU0oF3nbkVVK0mSJKnKRUQHcpLpupTSX2rv915CkloPh85JkiRJajIREcDVwISU0oUtHY8kqWktMdEUETcAjwHrR8SUiDi46cOSJEmS1Ep8FTgA+HpEjCst32rpoCRJTWOJczSllPZrjkAkSZIktT4ppYeBaOk4JEnNw6FzkiRJkiRJKoSJJkmSJEmSJBXCRJMkSZIkSZIKYaJJkiRJkiRJhTDRJEmSJEmSpEKYaJKkpfXZZ3DaadC3L3TqlH+eeirMnVtTJiUYMYK3YFMiZhMxmoh+i603oicR1xPxAhGfETGqjjL9iLiFiElEJCJG1FFmKBGvE/EOERfW2teLiMlErNqYS5ckSZJUkBEjIGLBZbXVavYfeCAJtij93T9veXyJ9UYsS8RZRLxCxCdEvEbEj8v2f4OI/xHxPhF/JGLZsn1diJi4xHuXMiaaJGlpnXsuXHYZXHIJvPACXHxxXv+//6spc9558MtfciK8BmwJTAPuJaLrYmruCEwHzgHGLKJMZ2AycCrwykJ7I3oAVwHHAzsB3yPi22UlLgPOJqW36nWtkiRJkprO+uvD1Kk1y3PPLbD7UXgf6Fm2fKsetd4A7AwcBqwPfBf4DwARywDXAb8FtgYGlMrN83PgRlIaX99LaF/fgpKkRXj0Udh117wA9OkDu+0GY0q5oZTgV7+Ck0/mmlNPfW9USv8lYjg52bQ/8Ls6601pMpC/aYjYexFlngSeLJX5WR0l1gJmktJNpTIPABsCdxCxF9AN+H2DrleSJElS02jffsFeTLV8AomU3qx3fRGDgR2BtUlpemnr5LISPYCVgctJ6WMi/ka+X4CILwODgc0bcAX2aJKkpTZwIDzwQO7NBPD883D//fCt0pcLr7wCb74JgwfXHJPSbODfwDZNHN1EoDMRmxOxErk31X+I6AacDxxOSqmJY5AkSZJUH5MmQa9eeTqOfffN62UGQBcippWGul1JxCpLqHEP8hfTxxIxpTQM7hIiupT2vw1MBQYTsRywLfl+oT35C/EfkNInDbkEE02StLROOgkOOAA22gg6dIB+/WD4cPjhD/P+N0tfOKy60DRIbwGL/rqiCCm9CwwHrgWeAK4lpbuB88hD6noQMZaICUQc0aSxSJIkSVq0r3wFRo2Cf/4Trrwy30dssw3MmJH377wzR+TpMnYAjgO+DNxPRMfF1LoWMBDYDNgL+BF5GN0ogNKXzkOA04DngWfIIx5OICeo3iLi36UE1Yj6XIZD5yRpad10E1x7LVx/fU4yjRsHRx+dv4U4+OCachG1jwyg6XsTpXQbcFtZHAOBrciN04vkRNR48jcXj5DSc3VVI0mSJKkJffObC65vtRWstRZccw0ceyzsuy/X77ffzOvy3+vPEfEU8CqwC/CXRdS6DPmeY39SmglAxI+Au4lYlZTeIqWHySMfKO1fhzxP0+bAfcBvgJuBJ4l4kpTuXNxlmGiSpKV1wglw/PG5ayvAJpvAq6/mycAPPrhmjPWbCw2lXoXcq6n55CdI/BY4hPztxrKkdF9p32hgEGCiSZIkSWppXbrkL7InTqx7f0r/j4gpwLqLqWUq8Mb8JFM2ofRzDeq+H/kdcCLwObAFeTLwD4n4O/B1YLGJJofOSdLS+ugjaNduwW3t2sHnn+fXffvmZNO999bsj+hEHv/8aHOFWXIKcD8pPU5uA8q/cFgWaFfnUZIkSZKa18cf53lge/ase39+wnQvcjJpUR4Bvlg2JxPAeqWfr9ZR50HAh6T0Z2pyRh1KP+t1v2CiSZKW1q67wjnnwJ13wuTJcNttcOGF8J3v5P0R8JOfwDnnMAy+QMTG5DHRs4Dr59cTcS0R1y5Qd0R/IvoDKwArldY3Ktu/bFmZTsBqpfV1FoozHzeUnGyCPGxuLhFHELEteaz3w0v7dkiSJElqhOOPhwcfzA8TGjMG9t4bPvwwz/86axYcfzw7wPJE9CFiEPB38pOsy6fJqH1PcT0wA/gDEf2I+CpwMXALKU1b4Px5YvEzgCMBSOk98hQbxxGxObA39bhfcOicJC2tX/8aTjstT/49bVr+xuHQQ+H002vKnHgizJ7NBWeeuQYwFhgDDCalD8pqWqOO2p+ptb4r+ZuHPqX1L9YqszZwOPAgeRhcFhHAFcAx88+Z0mwiDgAuA7oBI0lpbKn4gcCAlNKPFnXZEdEHuCOltHHkRNcXU0r/WFR5SZIkSYsxZQrstx9Mnw4rr5znaHr8cVhzTZg9G557jpthHeB/5F5MDwBDFntPkdIsInYEfk2e3Ptd4Hbg5DoiuBj4JSm9XrZtOPlL8qPIDxi6dUmXYaJJkpZW167wq1/lZVEiYMQIVjnzzP+klAbUWSalQXVsW2gG8Vr7J5MnFV+8/DSJgXVsv4ucnFpa/YEBgIkmSZIkqTFuvHHR+5ZbDu6+m+4Rzy7yfgIWdU/xIjB4iedPab86tj0FbLLEY8s4dE6S2qCIuD0inoqI8RFxWGnbQRHxv4h4EPhqWdlREbF32fqsWnUtC5wF7BMR4yJin2a6DEmSJEkVxh5NktQ2fT+l9E5ELAc8GRF3AmeSnyoxk9wNt/awvTqllOZExOksYaidJEmSpNbPHk2S1Db9OCKeBR4HVgcOAEanlN5OKc0BbiryZBFxWESMjYixn300c8kHSJIkSapKJpokqY2J/ISKHYGtU0qbkXsuvQCkRRwyl1J7EXlS8WUbes6U0hUppQEppQHtOndrTNiSJEmSqoCJJklqe7oB76aUPoqIDYCtgOWAQRHRPSI6AN8tKz+ZPKQOYHegQx11fgB0bbqQJUmSJFUDE02S1PbcBbSPiP8AZ5OHz00FRgCPAfcBT5eVvxL4WkQ8AXwF+LCOOh8ANnIycEmqQ58++emjtZdddsn7TzsNNtgAll+emdCfiH8Rsc1i64z4GhGPEjGDiNlEvEDE8XWUW4GIS4j4f0R8QsRLRAwp2z+UiNeJeIeIC2sd24uIyUSsutTvgSSpzXAycElqY1JKnwDfrGPXaOAPdZR/i9zraZ6flrZPBjYuvX4H2LLgUCWpdXjySfjss5r1qVNhiy1gSCnfs/76cNll0LcvX1977RfGwivAXUSsS/4Mrsss4BLgOeAj8tNCf0fER6R0OQC5h+o9wLvAEGAK0Bv4pLS/B3AVcCAwCbiTiPtJ6Y7SOS4Dzl5MDJIkLcREkyRJktSUVl55wfWrr4YVVoDvlkYpf+9783c9BR8DxwIHA/2Bu+usM6WncvH5XiFiT2Bb4PLStoOAVYDtyA96gDwcep61gJmklB8AEfEAsCFwBxF7kYda/77e1ylJEg6dkyRJkppPSjnR9L3vQefOC+3uCAEcBrwPjKt3vRGbA9sAD5Zt3QN4BPg1EW8S8TwRI0o9nQAmAp2J2JyIlcg9U/9DRDfgfOBwUlrUgyIkSaqTiSZJkiSpudx7L7zyChxyyILb77gDunThI/gScAzwjXoNWYuYQsQnwFjgclL6bdnetcgPd+gA7AKcBhwB/B8AKb0LDAeuBZ4AriWlu4HzyEPqehAxlogJRBzR+IuWJLUlDp2TJEmSmsuVV8KWW0L//gtu3357GDeOb6y77gv/gkeBm4nYmpSmLqHGbYEu5Ln0ziXiFVL6Y2nfMsA04FBS+gx4iojuwEVEnEBKiZRuA26bX1vEwFJdxwEvkhNR48k9nR4hpeeW6volSa2ePZokSZKk5jBtGvz1r3DooQvvW355WGcd7ocPSelg4FPgkIUL1pLSK6T0HCldCVxIfoLoPFOB/5WSTPNMADoDPRaqK2JZ4LfA4eTeUMuS0n2lZNdoYNCSL1KS1NaZaJIkSZKawx/+AB07wr771qf0MkDHBp6h9jGPAOsQUf43/3rkp9RNr+P4U4D7SenxUl3lox+WBdo1MB5JUhvUJEPnNunVjbHn7NIUVUuSJEnVJyW46qqcZOratWb7++/DeefBrrtCz54MzJNz/x7oDdw8v1zEtaV6hpXWjwJeIQ9vA9gOOJ6aJ84B/Ab4EXAxEZcCfYAzyXM5LTjJd8RGwFBg89KWF4G5pbmZxgM7AGcv1XtQT95LSFJ1c44mSZIkqamNHg0vvQTXXbfg9vbtYfx4+P3vYcYMboV1yE+D246U/lNWco1aNbYDziUnj+YCLwMnk4e+ZSm9TsRg8pC6ccCbwO+Bny9QU0QAVwDHkNIHpWNnE3EAcBnQDRhJSmMbd/GSpLbERJMkSZLU1LbfPvdqqq1zZ7itZi7uVSP+k1LafaFyKQ2qtf4r4FdLPG8eBrfNEsokYGAd2+8C1l7iOSRJKuMcTZIkSZIkSSqEiSZJkiRJkiQVwkSTJEmSJEmSCmGiSZIkSZIkSYUw0SRJkiRJkqRCmGiSJEmSJElSIUw0SZIkSZIkqRAmmiRJkiRJklSI9i0dgCSpbdmkVzfGnrNLS4chSZIkqQnYo0mSJEmSJEmFMNEkSZIkSZKkQphokiRJkiRJUiFMNEmSJEmSJKkQJpokSZIkSZJUCBNNkiRJkiRJKoSJJkmSJEmSJBXCRJMkSZIkSZIKYaJJkiRJkiRJhTDRJEmSJEmSpEKYaJIkSZIkSVIhTDRJkiRJkiSpECaaJEmSJEmSVAgTTZIkSZIkSSqEiSZJkiRJkiQVwkSTJEmSJEmSCmGiSZIkSZIkSYUw0SRJkiRJkqRCmGiSJEmSJElSIUw0SZIkSZIkqRAmmiRJkiRJklQIE02SJEmSJEkqRKSUiq804gPgxcIrLkYPYHpLB7EIxtY4xtY4xtY4SxvbmimllYsKphpVeBtRWyX/X6ytWmKtljjBWJtCtcQJLRer7UR1tRONVU2/C43h9VU3r6+yVXw70b6J6n0xpTSgiepeKhEx1tgaztgax9gax9havYptI2qrpn/vaom1WuIEY20K1RInVFesrVDVtBON1dr/f3l91c3r09Jy6JwkSZIkSZIKYaJJkiRJkiRJhWiqRNMVTVRvEYytcYytcYytcYytdaum99BYi1ctcYKxNoVqiROqK9bWpi289639Gr2+6ub1aak0yWTgkiRJkiRJanscOidJkiRJkqRCNDrRFBE7R8SLEfFSRJxcx/6IiEtK+/8TEV9aulALjW2DiHgsIj6JiOObK656xja09H79JyIejYjNKii23UtxjYuIsRExsFJiKyu3ZUR8FhF7V0psETEoImaW3rdxEXF6pcRWFt+4iBgfEQ9WSmwRcULZe/bf0r/rShUSW7eI+HtEPFt63w5qjriqTSW3E3XEUrHtRq04KrYNqSOWim1T6oilYtuYWuev2Pamjlgqtv2pFUPFtkVtQTW1E41RLW1LY1VTm9RY1dSWNUa1tH+NVU3tZquTUmrwArQDXgbWApYFngU2qlXmW8A/gQC2AsY05lxNFNsqwJbASOD45oirAbFtA6xYev3NCnvfulAz3HJT4IVKia2s3P3AP4C9KyU2YBBwR3P9P2tgbF8AngfWKK2vUimx1Sq/K3B/pcQG/Aw4t/R6ZeAdYNnm/jeu5KWS24lGxtoi7UYj4myRNqSRsbZIm9KYWMvKNWsb04j3tEXam0bG2iLtT2P+7cvKN1tb1BaWamonmvD6WrxtaeLrq4g2qYmvsSLasqa6vrJyLdb+NfG/X0W0m61xaWyPpi8DL6WUJqWU5gA3ArvXKrM7cG3KHge+EBE9G3m+QmNLKU1LKT0JfNoM8TQ0tkdTSu+WVh8HeldQbLNS6TcSWB5orgm+6vP/DeAo4FZgWjPF1ZDYWkJ9Ytsf+EtK6TXIvxsVFFu5/YAbmiWy+sWWgK4REeQ/MN4B5jZTfNWiktuJ2iq53ShXyW1IbZXcptRWyW1MuUpub2qr5PanXCW3RW1BNbUTjVEtbUtjVVOb1FjV1JY1RrW0f41VTe1mq9PYRFMv4PWy9SmlbQ0t0xRa6rz10dDYDiZ/i9Mc6hVbRHwnIl4A7gS+XymxRUQv4DvAb5sppnnq+2+6deRhVv+MiH7NE1q9YlsPWDEiRkfEUxExrIJiAyAiOgM7kxu45lCf2C4FNgT+H/AccHRK6fPmCa9qVHI7UVulxLEkldyG1FbJbUptldzGlKvk9qa2Sm5/ylVyW9QWVFM70RjVHHt9VFOb1FjV1JY1RrW0f41VTe1mq9O+kcdFHdtqZ2/rU6YptNR566PesUXE9uQP5OYa51uv2FJKtwG3RcR2wNnAjk0dGPWL7VfASSmlz3Ink2ZTn9ieBtZMKc2KiG8BtwPrNnVg1C+29sAWwA7AcsBjEfF4Sul/FRDbPLsCj6SU3mnCeMrVJ7adgHHA14G1gXsj4qGU0vtNHFs1qeR2orZKiWNJKrkNWSiEOrZVSptSWyW3MeUqub2prZLbn3KV3Ba1BdXUTjRGNcdeH9XUJjVWNbVljVEt7V9jVVO72eo0tkfTFGD1svXe5G/2G1qmKbTUeeujXrFFxKbAVcDuKaUZlRTbPCmlfwNrR0SPpg6M+sU2ALgxIiYDewOXR8QelRBbSun9lNKs0ut/AB0q6H2bAtyVUvowpTQd+DfQHBM1NuT/274071CF+sR2EHnIR0opvQS8AmzQTPFVi0puJ2qrlDiWpJLbkNoquU2prZLbmHKV3N7UVsntT+0YKrUtaguqqZ1ojGqOvT6qqU1qrGpqyxqjWtq/xqqmdrP1SY2bWKs9MAnoS83EWv1qldmFBSfve6Ix52qK2MrKjqB5JwOvz/u2BvASsE1zxdWA2NahZrK7LwFvzFtv6dhqlR9F800GXp/3bbWy9+3LwGuV8r6Rh3/9q1S2M/BfYONKiK1Urht5/qPlm+PfswHv22+AEaXXq5Z+F3o0V4zVsFRyO9GYWMvKNmu70Yj3tEXakEbG2iJtytL8+5fKN1sb04j3tEXam0bG2iLtT2P+7VuiLWoLSzW1E011fWVlW6xtaeJ/v4pok5r4GiuiLWuq66tVvkXavyb+96uIdrM1Lo0aOpdSmhsRPwLuJs/m/vuU0viIOKK0/7fkWem/Vfpw+YjcA6DJ1Se2iFgNGAusAHweET8hz0DfpMNe6vm+nQ50J2eLAeamlAY0ZVwNiG0vYFhEfArMBvZJpd/KCoitRdQztr2BH0TEXPL7tm+lvG8ppQkRcRfwH+Bz4KqU0n8rIbZS0e8A96SUPmzqmBoY29nAqIh4jvzH70kpfyOvkkpuJxoTa0u1Gw2NkxZqQxoZa4u0KY2MtcVVcnvTmFhbqv1paJylos3eFrUF1dRONEa1tC2NVU1tUmNVU1vWGNXS/jVWNbWbrVH4PkqSJEmSJKkIjZ2jSZIkSZIkSVqAiSZJkiRJkiQVwkSTJEmSJEmSCmGiSZIkSZIkSYUw0SRJkiRJkqRCmGiSJEmSJElSIUw0SZIkSZIkqRAmmiRJkiRJklSI/w+L4UuvtEQOqwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_ccas(df, ccas, title='-'):\n",
    "    fig, axs = plt.subplots(1,len(ccas), figsize=(20,5))\n",
    "    fig.suptitle(title)\n",
    "    for i, feat in enumerate(ccas):\n",
    "        pd.DataFrame(df[feat].value_counts(normalize=True)).plot(kind='barh', ax = axs[i])\n",
    "        axs[i].set_title(feat)\n",
    "        totals = []\n",
    "        for j in axs[i].patches:\n",
    "            totals.append(j.get_width())\n",
    "\n",
    "        # set individual bar lables using above list\n",
    "        total = sum(totals)\n",
    "\n",
    "        for j in axs[i].patches:\n",
    "        # get_width pulls left or right; get_y pushes up or down\n",
    "            axs[i].text(j.get_width()+.01, j.get_y()+.1, \\\n",
    "                    str(round((j.get_width()/total)*100, 2))+'%', fontsize=14, color='red')\n",
    "            \n",
    "plot_ccas(IXI_metadata, ['ETHNIC_ID','AGE_GROUP','SEX'], 'ALL POPULATION')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:07.596126Z",
     "start_time": "2020-11-12T10:11:07.254508Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_ccas(final_train_val_volumes, ['ETHNIC_ID','AGE_GROUP','SEX'], 'Train and validation')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:08.096153Z",
     "start_time": "2020-11-12T10:11:07.599086Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKMAAAFTCAYAAADlUM+vAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABDCklEQVR4nO3debxd0/n48c8jEUQihqT4UoIqktIgagqSUlQEVSUahNbQX1NCDdXGkJa0pmopqoY2tAjfqqGoqUTNRMQQqSJSjS/VGEJIGknW7491kjvkJrnjPuee+3m/Xvt1z9577bWffXLvWdnPWWvtSCkhSZIkSZIkFWG5cgcgSZIkSZKkjsNklCRJkiRJkgpjMkqSJEmSJEmFMRklSZIkSZKkwpiMkiRJkiRJUmFMRkmSJEmSJKkwJqMkSZIkSZJUGJNRkiSpIkTErFrLgoiYXWt9WDPqGx8RR7ZFrJIkSWq+zuUOQJIkCSCl1G3h64iYBhyZUrq/fBFJkiSpLdgzSpIkVbSIWC4iTo2I1yLi3Yi4KSJWL+1bMSL+UNr+QUQ8HRFrRsQYYCfgklLPqkvKexWSJElayGSUJEmqdMcB+wG7AP8DvA9cWto3HOgBfBZYA/gOMDulNAp4GPheSqlbSul7RQctSZKkhpmMkiRJle4YYFRKaXpK6b/AaOCAiOgMfEpOQn0upTQ/pfRMSunDMsYqSZKkZXDOKEmSVOnWB26JiAW1ts0H1gR+T+4VNS4iVgX+QE5cfVp4lJIkSWoUe0ZJkqRK9y/gqymlVWstK6aU3kwpfZpS+nFKqQ+wA7A3cFjpuFS2iCVJkrREJqMkSVKluxwYExHrA0REr4jYt/R6UERsHhGdgA/Jw/bml477N7BhOQKWJEnSkpmMkiRJle4i4Hbg3oj4CHgC2La0by3gj+RE1BTgIfJQvYXHHRAR70fExcWGLEmSpCWJlOzBLkmSJEmSpGLYM0qSJEmSJEmFMRklSZIkSZKkwpiMkiRJkiRJUmFMRkmSJEmSJKkwJqMkSZIkSZJUGJNRkiRJkiRJKozJKEmSJEmSJBXGZJQkSZIkSZIKYzJKkiRJkiRJhTEZJUmSJEmSpMKYjJIkSZIkdTgRMSAiHouImRHxXkQ8GhHbRMThETE/ImbVW/4nIrpFxLSI+GaterpHxBsRcUA5r0dqT0xGqRClD+zZ9T7MF9R6PafeB/7k0nEpIj5Xr67REfGHWuspIl6IiOVqbTs7IsaWXvculelca/+XIuKuiPig1PA8FRFHLOMaBkbE9Frr40txfxQRH0bEMxFxakSs0OI3TJI6gNLn6Pv1PzcjYmhEPBkRH0fEO6XX342IKO0fGxFz67UpzzXifF0i4oyIeLlU95sR8ZeI2L1Wmdrt1dulc3WrV88OEfFA6fN/ZkT8OSL61Np/eEQ80sD5p0XEbg1cw3sRcV9EbNr0d1GS1BwRsQpwB/ArYHVgHeDHwH9LRR5PKXWrt/xfSmkWcDRwUUT0KpU9D5iQUvpjwZchtVsmo1SkIfU+zJdb+Br4DnU/8Ps2se7/AYY2pmBEbA88ADwEfA5YA/h/wFebeE6A76WUugNrAyeWYrhr4Q2TJKlhEdEb2AlIwD61tp8IXAScD6wFrEluI3YEutSq4rx6bcoXG3HaPwL7AocBqwEblM41uF65IaW2qR+wJfDDWvFtD9wL3EZuezYAngMejYgNG3Pt9a8BWBd4BxjbxOMlSc33eYCU0g0ppfkppdkppXtTSs8v68CU0r3AncDFETEQOBAY0ZbBStXGZJSqxXnAj2v3flqK84FrUkrnppRmpOyZlNKBzT15SunjlNJ48g3V9ix+YyNJqusw4AlyAmY4QET0AH4CfDel9MeU0kelz+hnU0rDUkr/XXJ1S1fqkfQVYN+U0pMppbml5e6U0siGjkkpvQ3cQ05KLXQecG1K6aJSfO+llE4rXcvo5sSWUvoEuB74QnOOlyQ1yz+A+RFxTUR8NSJWa+LxJwADyV90nJRSequ1A5SqmckoVYs/AR8Chy+tUER0JSeL2qQLbUrpDWAC+dt+SdKSHQZcV1r2iIg1yZ/PK5B7HbW23YAnU0rTl1myJCLWJfeafbW03hXYAfjfBorfRE52NVlpGOAw4NnmHC9JarqU0ofAAHIP3SuB/0TE7aX2CGC70pQeC5fX6h3/PjAZ6Eq+F5HUBCajVKRb632gH9XI4ybWPg44tYEyCTgdOGMZczatRv69b8tvLv6PPO5cktSAiBgArA/clFJ6BngN+CbQE5iRUppXq+xjpc//2RGxc61qTqrXplyzjNP2BN6uVe/qpeNmRsScemVvjYiPgH+Rh8+dWdq+OktuQ94qnaMpTiq1a68C3VjGFyqSpNaVUpqSUjo8pbQuuXfq/wC/LO1+IqW0aq1lo9rHRsQhQG/gfuDcAsOWqoLJKBVpv3of6Fc28ritah8HnNNQoZTSXcAb5AkFl+R9YAF5jqe2sg7wXhvWL0nt3XDg3pTSjNL69aVt7wI9aw+5TintUPrsf5e6/2+5oF6bMnwZ53yXWp/9peF1qwJbk3tj1bZfaT7AgcCm1CSZltaGrA0svJ55wPINlFke+LSBa1grpbRPSum1Bo6RJBUgpfR38tDxZQ6ZjojPAL8AjgKOAQ6s94WJpGUwGaVqcxowitxddjGleTkeB77eFiePiM+Sb2webov6Jam9i4iVyBO97lJ6Wt3b5Hk3vgh8Qn6K0b5tcOq/AtuUht41SkrpIfKNyQWl9Y/Jbcg3Gih+YOkckL8YWa/2wyxKQ/w+A/yzOcFLklpXRGwaEScubBdK/48/mDwH4LJcAtyaUnqwNFfUKcCVPlVbajyTUaoqpUnEX6A0Ge4SnAIcHhEnR8QaABHxxYgY19zzRkTXiNiFPM/JU8Bdza1LkqrcfsB8oA95YvB+wGbkJP4+5MdqXxYRB0REt4hYLiL6ASu35KSlJx89SB6Ct21EdImI5YHtlnHoL4GvlGKAPFR8eEQcFxHdI2K1iDibPN/Vj0tlngTmAKdGxIoRsTK5V+8ETEZJUqX4CNgWeDIiPiYnoV4kPyEbYPuImFVv2SYi9iPPNXXywopSSlcB04EzCr0CqR0zGaUi/bneh/ktbXSe01jKnE0ppceAL5eWqRHxHnAFzUsgXVKaV+Tf5BuWm4E9U0oLmlGXJHUEw4HfpZTeSCm9vXAhf8s8DLgQ+D75i4N3yJ+vvwF+ADxWq55T6rUpM1i2/YE7gD8AHwCvl86555IOSCn9B7iWPC8hKaVHgD1Kdb1FTi5tCQxIKb1SKvNf8lNVB5JvTqaS5yE5MKWUGhGnJKmNpZTeTCkdmFJaJ6W0cunnMSmlD1NKY1NKnVJK3eotT6eUbk0p/U9K6YN69e2aUhpVpsuR2p3w/0SSJEmSJEkqij2jJEmSJEmSVBiTUVItEfGjBsaGz4qIv5Q7NknS0kXEsCV8hk8ud2ySJEmq4TA9SZIkSZIkFcaeUZIkSZIkSSpM57aotGfPnql3795tUbUktWvPPPPMjJRSr3LHUU62EZK0ZLYTthOStDTV0k60STKqd+/eTJgwoS2qlqR2LSL+We4Yys02QpKWzHbCdkKSlqZa2gmH6UmSJEmSJKkwJqMkSZIkSZJUGJNRkiRJkiRJKkybzBklqfJ9+umnTJ8+nTlz5pQ7lKq04oorsu6667L88suXOxRJWoxtQPnZTkhSy1Vze1bt7YTJKKmDmj59Ot27d6d3795ERLnDqSopJd59912mT5/OBhtsUO5wJGkxtgHlZTshSa2jWtuzjtBOOExP6qDmzJnDGmusUVUf2pUiIlhjjTWq8hsaSdXBNqC8bCckqXVUa3vWEdoJk1FSB1ZtH9qVxPdWUqXzc6q8fP8lqXVU6+dptV7XQiajJEmSJEmSmmnMmDH07duXLbbYgn79+vHkk08ycOBANtlkE/r160e/fv044IADADjuuOM466yz6hw7YsSIcoVeNs4ZJQmA3qfe2ar1TTtn8DLLdOrUic0333zR+tChQ3nyySd5/fXXmTVrFv/5z38WjZG+7LLL+NGPfsQFF1xA//798zmmTWPvvffmxRdfZPz48QwaNIjbb7+dIUOGALD33ntz0kknMXDgQAYOHLjo2FmzZnHiiSdy//33s+KKK7LGGmtw/vnns+222zYYZ7du3Zg1axbTpk1js802Y9NNN2XOnDl0796dESNGMHz48Ja+XZJUVuVoAwBuueUW9t9/f6ZMmcKmm24KwFNPPcUpp5zCm2++Sffu3Vl77bU555xz2HzzzRk9ejRXXnklvXr1WlTH+PHjWXXVVRusv7F1zZ07l9NPP52DDz4YyHN1jBkzhmuuuYaIYJ111uGSSy6hb9++QE27sNDYsWOZMGECl1xySZ16582bx09/+lP22Wef5ryNkqQmKkd79vjjj3PHHXcwceJEVlhhBWbMmMHcuXMBuO666xbduyx09tln069fP4YNG0ZEcNVVV/Hss8+2atztgckoSWWz0korMWnSpAb3jR8/ngsuuIA77rij0fWtu+66jBkzZlEyakmOPPJINthgA1555RWWW245pk6dypQpUxp1jo022mhRYzF16lT2339/FixYwBFHHNHoOCVJ2Q033MCAAQMYN24co0eP5t///jcHHngg119/PTvssAMAjzzyCK+99tqiLy9OOOEETjrppGXW3ZS6XnnlFbbeemsOOOAAll9+eS699FIee+wxnnvuObp27cq9997LPvvsw+TJk1lxxRWXee6F9U6ZMoWddtqJd955h+WWc0CCJFWjt956i549e7LCCisA0LNnz6WWX2WVVRgzZgzf+973APjJT36yxC9VqpmtoqSq8cUvfpEePXpw3333LbHMa6+9xpNPPsnZZ5+96MZgww03ZPDgxn2LX9uGG27IhRdeyMUXX9zsmCWpo5o1axaPPvooV199NePGjQPgkksuYfjw4YuSRwADBgxgv/32a3L9Talr4403pmvXrrz//vsAnHvuufzqV7+ia9euAOy+++7ssMMOXHfddU2KYbPNNqNz587MmDGjyfFLktqH3XffnX/96198/vOf57vf/S4PPfTQon3Dhg1bNEzv5JNPXrT94IMP5v333+fDDz/k0EMPLUfYZWfPKEllM3v2bPr167do/Yc//CEHHXTQUo8ZNmwYK620EgBz585d7Jvm0047jdNOO42vfOUrDR4/efJk+vXrR6dOnVoWfMlWW23F3//+91apS5I6kltvvZU999yTz3/+86y++upMnDiRyZMnL3Po8y9+8Qv+8Ic/ALDaaqvx4IMPNliuMXUtNHHiRDbeeGM+85nP8OGHH/Lxxx+z0UYb1SnTv39/Jk+e3Kj6FnryySdZbrnl6gwrlCRVl27duvHMM8/w8MMP8+CDD3LQQQdxzjnnAA0P0wOYPn06b7/9NhHBrFmz6NatW9Fhl53JKElls7RhektS+wN94ZxRte20004APPzww60S47KklAo5jyRVmxtuuIHjjz8eyHMG3nDDDYuV2Xbbbfnwww/Zfffdueiii4DGD9NrTF2/+MUvuPLKK5k6dSp33333Uo9PKS31yUa19y1MmHXv3p0bb7yx6p+IJEkdXadOnRbNU7v55ptzzTXXLLX8yJEjGT16NFOmTOHHP/4x559/fkGRVg6H6UmqOqNGjWLMmDEN7uvbty/PPfccCxYsaJVzPfvss2y22WatUpckdRTvvvsuDzzwAEceeSS9e/fm/PPP58Ybb6Rv375MnDhxUbknn3ySs846i5kzZzb5HI2p64QTTuDll1/mxhtv5LDDDmPOnDmsssoqrLzyykydOrVOfRMnTqRPnz5A/jJl4eS0AO+9916dOUJOOOEEJk2axMMPP7zoSxJJUnV6+eWXeeWVVxatT5o0ifXXX3+J5f/yl7/wzjvvcNhhh3H66adzyy238NJLLxURakUxGSWp6uy+++68//77PPfcc4vt22ijjejfvz9nnnnmol5Nr7zyCrfddluTzzNt2jROOukkjj322BbHLEkdyR//+EcOO+ww/vnPfzJt2jT+9a9/scEGG7D77rszduxYHnvssUVlP/nkk2adY8SIEY2ua//996d///6Lvsk++eSTOe6445g9ezYA999/P4888gjf/OY3Adhll10WDRWcPXs2N910E4MGDWpWnJKk9m3WrFkMHz6cPn36sMUWW/DSSy8xevRooO6cUbvtthtz5szh+OOP57LLLiMiWHnllTnvvPMWTWbekThMTxLQ+Mdwt6b6c0btueeei8ZXt9SoUaPYd999G9x31VVXceKJJ/K5z32Orl27ssYaazS6a+xrr73GlltuyZw5c+jevTvHHnusT9KT1O4V3QbccMMNnHrqqXW2ff3rX+f666/nxhtv5Ac/+AFvvvkmn/nMZ+jZsydnnHHGonK154yCPPdU7969FzvHWmuttcy6ajvjjDP45je/yVFHHcWxxx7L+++/z+abb06nTp1Ya621uO222xbNWXjRRRdxzDHHcPHFF5NS4rDDDmPnnXduhXdGktQS5bin2Xrrret88bHQ+PHjGyz/8ssv11nff//92X///dsitIoWbTHfSf/+/dOECRNavV5JrWfKlCkOL2tjDb3HEfFMSmnxWQw7ENsIqfxsAyqD7UTDbCckNVa1t2fV3E44TE+SJEmSJEmFcZieJJEn0911110X2/7Xv/6VNdZYowwRSZIa45577uEHP/hBnW0bbLABt9xyS5kikiRJy2IySpKANdZYg0mTJpU7DElSE+2xxx7sscce5Q5DklQmKSUiotxhtLq2mFKpkjhMT+rAqv0Drpx8byVVOj+nysv3X5JabsUVV+Tdd9+tus/UlBLvvvsuK664YrlDaTP2jJI6qIUf3GussUZVfpNQTh2h8WiJF96cSe9T7yzsfOV4qopU6WwDyst2QpJax7rrrsv06dP5z3/+U+5QWt2KK67IuuuuW+4w2ozJKKmDquYP7kpQ7Y2HpPbNNqD8bCckqeWWX355Nthgg3KHoWYwGSV1UH5wS1LHZRsgSZLKyTmjJEmSJEmSVBiTUZIkSZIkSSqMyShJkiRJkiQVxmSUJEmSJEmSCmMySpIkSZIkSYUxGSVJkiRJkqTCmIySJEmSJElSYUxGSZIkSZIkqTCd26LSF96cSe9T72yLqjukaecMLncIkiRJUiG8l5DUGryPrmz2jJIkSZIkSVJhTEZJkiRJkiSpMCajJEmSJEmSVBiTUZIkSZIkSSqMyShJkiRJkiQVxmSUJEmSJEmSCmMySpIkSZIkSYUxGSVJkiRJkqTCmIySpCoWEWdFxMha62MiYmREnB8RL0bECxFxUGnfwIi4o1bZSyLi8NLraRHx44iYWDpm09L2XhFxX2n7byLinxHRs+DLlCRJktSOmIySpOp2NTAcICKWA4YC04F+wBeB3YDzI2LtRtQ1I6W0FfBr4KTStjOBB0rbbwHWa9XoJUmSJFUdk1GSVMVSStOAdyNiS2B34FlgAHBDSml+SunfwEPANo2o7k+ln88AvUuvBwDjSue6G3i/oQMj4uiImBARE+Z/MrOZVyNJkiSpGpiMkqTqdxVwOHAE8FsgllBuHnXbhRXr7f9v6ed8oHPp9ZLqqiOldEVKqX9KqX+nrj0ac4gkSZKkKmUySpKq3y3AnuTeT/cAfwMOiohOEdEL2Bl4Cvgn0CciVoiIHsCujaj7EeBAgIjYHVitDeKXJEmSVEU6L7uIJKk9SynNjYgHgQ9SSvMj4hZge+A5IAGnpJTeBoiIm4DngVfIQ/qW5cfADaVJ0B8C3gI+aoPLkCRJklQl7BklSVWuNHH5duTJzEnZySmlL6SUNk8p3biwbErplJTSJimlvVNK+6eUxpa2904pzSi9npBSGlg6ZCawR2kC85uAd1JK/6U9eOstGD4cevWCFVeEPn3goYeWXP6ll2DQIFhzzVx+ww3hRz+CuXPrlrv+eujXD7p2hbXWgkMOgbffrtl/333w+c/DKqvAoYfWPX7WLNh4Y5g8uVUvVZIkSaokJqMkqYpFRB/gVeCvKaVX2uAU6wFPR8RzwMXAUW1wjtb3wQew446QEtx5J0yZAr/6FXzmM0s+pkuXnLy69154+WX45S/h6qvhtNNqyjz6aE4wDR+eE0q33pqTWMOG5f0LFuTX3/kOPP44TJgAV1xRc/xpp8HQodC3bxtctCRJklQZHKYnSVUspfQSsGEb1v8KsGVb1d9mzjsP1l4brr22ZtsGGyz9mM99Li8Lrb8+jB8PDz9cs+3xx2HddeGEE2rqPPbYvADMmAH/+Q9897u5d9U+++REGMBTT+VE17ONGR0pSZIktV/2jJIkdTy33grbbgsHHZR7Q/XrB5dckntKNdarr8Ldd8Muu9Rs23HHPPzvz3/Odc2YAePGwV575f29euUk2L33wuzZOZG1xRYwbx4ccwz8+tewwgqteaWSJElSxTEZJUnqeKZOhcsuy/M+3XMPjBwJp54Kl1667GN32CH3atp4YxgwAH7605p9228PN9yQh+J16ZKTTynBNdfk/RFw001w1ll5jqott4RvfQvOPx+22SbPR7Xzzrnu0aPb5NIlSZKkcjMZJUnqeBYsgK22gp/9LCeEjjgCjjuuccmoG2+EiRPzROV33QXnnluz76WXcj2nnw7PPJN7Tr39du71tNCAAfD00/D66/l8//xnnjfqvPPyZOfDh+f6x43L81lJkiRJVcY5oyRJHc/aa+eeSbVtthlcdNGyj/3sZ/PPPn1g/nw48kg4+WTo3Dknt770pbwOeQjeyivDTjvBmDE1x9Z2zDE5EbXccjmBNXRoPmbIEHjgARg8uGXXKkmSJFUYe0ZJkjqeHXfMT8Sr7R//yJOSN8WCBXm+p/nz8/onn0CnTnXLLFxvaD6q3/0uJ56+8Y1cF8Cnn+afc+fW1CtJkiRVEZNRkqSO54QT4Ikncm+lV1+F//1fuPhiGDGipswPfwi77lqz/vvf53J//3uec+qmm3KZAw6omXR8yBC47bY8EfnUqfDoo3nY3lZbwXrr1Y3hnXfgxz+uGRq46qrQty/8/Of5iXp//GMe0idJkiRVGYfpSZI6nm22yU/U+9GP8mTi662Xf373uzVl3noLXnutZn3hMLxXXsm9nNZfPyevTjihpszhh8NHH+Un8514IvToAYMG5WF49Y0cmcvUHrp3zTW5jl/9Cg47DL7+9Va+cEmSJKn8TEZJkjqmwYOXPh/T2LF11w8+OC/LcuyxeVmWG25YfNvWW8MLLyz7WEmSJKkdc5ieJEmSJEmSCmMySpIkSZIkSYUxGSVJkiRJkqTCmIySJEmSJElSYZaZjIqIz0bEgxExJSImR8TIIgKTJEmS1P5FxG8j4p2IeLHcsUiSKkNjekbNA05MKW0GbAeMiIg+bRuWJEmSpCoxFtiz3EFIkirHMpNRKaW3UkoTS68/AqYA67R1YJIkSZLav5TS34D3yh2HJKlyNGnOqIjoDWwJPNkm0UiSJEmSJKmqdW5swYjoBtwMHJ9S+rCB/UcDRwN0WqVXqwUoSaoum6/TgwnnDC53GJKkCuK9hCR1LI3qGRURy5MTUdellP7UUJmU0hUppf4ppf6duvZozRglSZIkVTHvJSSpY2nM0/QCuBqYklK6sO1DkiRJkiRJUrVqTM+oHYFDgS9HxKTSslcbxyVJkiSpCkTEDcDjwCYRMT0ivl3umCRJ5bXMOaNSSo8AUUAskiRJkqpMSungcscgSaosTXqaniRJkiRJktQSJqMkSZIkSZJUGJNRkiRJkiRJKozJKEmSJEmSJBXGZJQkSZIkSZIKYzJKkiRJkiRJhTEZ1V797GewzTawyirQqxcMGQIvvli3zJ/+BHvskfdHwPjxy673T3+C3XfPx3TvDttuC7ffvni5iy6CTTeFlVaCddeFESNg1qya/dddB5/9LKy+Onz/+3WPffNN6N0b/v3vpl61JEmSJEmt66c/zffM3/te3e3/+Afsvz+suip07QpbbQVTpiy9ruuvh379cvm11oJDDoG3367ZP3kyHHAAbLhhPufo0YvXsZT76d6wPBHTiFizGVdaMUxGtVfjx8N3vwuPPQYPPACdO8Nuu8F779WU+fhj2GEHuPDCxtf70EPw5S/DnXfCs8/CXnvB174GDz9cU+b66+GUU2DUqPyHeO21cNddMHJk3j9jBhx5JFxwAdxzD/zhD3DHHTXHjxgBp58Oa7brvx1JkiRJUnv3xBNw5ZWwxRZ1t7/+Ouy4I2ywQb7nfvFFOPts6NZtyXU9+igceigMH56TTrfeCi+9BMOG1ZT55JPcOePss3Pd9S3jfvpKWA84i5Tade+OzuUOQM10zz1113//e+jRI//yDxmStx16aP45Y0bj673oorrrZ56ZE1O33go77ZS3PfYYbLddTf29e8Nhh8HNN+f1qVNzLAcdlNcHDcpJq733zmVmzoRvfavxMUmSJEmS1NpmzsyJoquvhp/8pO6+UaPyqKGf/7xm24YbLr2+xx/PI4dOOCGvb7ABHHtsXhbaZpu8QO6RVd8y7qe7Qyfgt026zgpkz6hq8dFHsGABrLZa29Rdu94BA2DSpJxBBnjjjTyUb6+98vrGG+ds77PP5p5aTz+ds8wzZ8LJJ8NvfpO7I0qSJEmSVC5HH52HzH35y3W3L1gAf/4z9OkDe+6Zp7HZZhu48cal17fjjvDWW/nYlHLHkHHjau6VG2MZ99NHwT9JKTX9YiuLyahqMXJkHpe6/fatW++ll8L06TW9oACGDs0Z3J13huWXh/XXh803h3PPzftXWw2uuSb3lvrSl/LPPfbIQ/uOPDL/QfbvD5ttBpdf3rrxSpIkSZK0LFdeCa++Cmedtfi+d97JcyL/9Ke5d9R998HBB+deVLWnoKlv++3hhhtyuS5dchIrpXx/3FjLuJ9eEzoTMYGIKUR8p+kXXhkcplcNvv99eOSRvHTq1Hr13nxz7sk0blxOOC300EP5D/ayy/IE56++mpNhZ55Z07Xxa1/Ly0KPPJJ7Uv3857DJJvmPq2/fnOHdcceczJIkSZIkqa29/DL86Ed5buQuXRbfv2BB/rnvvjUTiPfrBxMm5A4be+/dcL0vvQTHHZfnSN5jj9xL6uST4Zhj8lzLjbWU++nrRo3aCNgLmAw8T8SjpPRC4yuvDCaj2rsTTsjJogcfXPb41aa4+ebcG+raa2GfferuO+20nBU+8si8vvnmebL0I4+EM87Ik6nXNncufOc7cNVVefzr3Ll5snWAgQPzZOwmoyRJkiRJRXj88Txi5wtfqNk2fz787W959M7HH+f72j596h632Wb5/ntJfvaz3Jvp5JPz+hZbwMor5/mXx4zJT8hrqnr308tDkNL9AESMBwYC7S4Z5TC99mzkyPxkuwcegE03bb16b7opP35y7Ng8fra+Tz5ZvAdWp065+2FDxozJY3C32y5nmOfNq9k3d27+o5ckSZIkqQj77QcvvJDnQl649O+fp6SZNCn3ltpmm9yDqrZ//KPuqKH6lnSvDEu+X16WevfTnaD2BMxdyBOatzv2jGqvRozIT9C79dY8pvTtt/P2bt1qHjX53nt5cvEPPsjrr74Kq64Ka62VF8jjT6Gmy+C4cblH1AUX5DmhFtbbpQusvnp+PWQIXHhh/mNdOEzv9NNzV8X6vaJeegmuuy5PvgZ5iF7nzjnb3Lcv/PWv+VhJkiRJkoqw6qp5qW3llfM978LeUqecAgcemHs1ffnLeTTSuHH5Hnyh+vfTQ4bAUUfBr39dM0zv+ONhq61gvfVymblz830ywJw5+Z570qR8H/+5z9WNqYH76fmQSnNFTQZ2BRqY9KrymYxqry67LP/cdde62888E0aPzq9vvx2OOKJm31FHLV7mjTfqHn/55bnn0vHH52WhXXbJw+kgD9OLyEmk6dOhZ8/8RzdmTN26UspPJ/jFL6B797xtpZVyEm3EiPw0gFGjclJLkiRJkqRKsd9+cMUVeRLzkSPzU+6uvRYGD64pU/9++vDD89PoL7kETjwRevSAQYPgvPNqyvzf/8GWW9asv/ZafuJ87XtuWOL99FHw+h/hZKAHMIaUJrTqdRck2uKJgCusvXFae/gvW73ejmraOYOXXUhSuxARz6SUOnQGtn///mnChHbZZkpSm7Od8F5CUuuo1vvoamknnDNKkiRJkiRJhTEZJUmSJEmSpMKYjJIkSZIkSVJhTEZJkiRJkiSpMCajJEmSJEmSVBiTUZIkSZIkSSqMyShJkiRJkiQVxmSUJEmSJEmSCmMySpIkSZIkSYUxGSVJkiRJkqTCmIySJEmSJElSYUxGSZIkSZIkqTCd26LSzdfpwYRzBrdF1ZIkSZKqmPcSklT97BklSZIkSZKkwpiMkiRJkiRJUmFMRkmSJEmSJKkwJqMkSZIkSZJUGJNRkiRJkiRJKozJKEmSJEmSJBXGZJQkSZIkSZIKYzJKkiRJkiRJhTEZJUmSJEmSpMKYjJIkSZIkSVJhOpc7AElSx/LCmzPpfeqd5Q6jw5h2zuByhyBJkiTVYc8oSZIkSZIkFcZklCRJkiRJkgpjMkqSJEmSJEmFMRklSZIkSZKkwpiMkiRJkiRJUmFMRkmSJEmSJKkwJqMkSZIkSZJUGJNRkiRJkiRJKozJKEmSJEmSJBXGZJQkSZIkSZIKYzJKkiRJkiRJhTEZJUmSJEmSpMKYjJIkSZIkSVJhTEZJkiRJkiSpMCajJEmSJEmSVBiTUZIkSZIkSSqMyShJkiRJkiQVxmSUJEmSJEmSCmMySpI6iIiYFhE9G9g+OiJOamJdh0fEJa0XnSRJkqSOwmSUJKlJIqJzuWOQJEmS1H6ZjJKkKhQRh0TEUxExKSJ+ExGd6u0fFREvR8T9wCa1tm8UEXdHxDMR8XBEbFraPjYiLoyIB4Fza5XvHhGvR8TypfVVSj2wli/mSiVJkiS1NyajJKnKRMRmwEHAjimlfsB8YFit/VsDQ4Etgf2BbWodfgVwbEppa+Ak4LJa+z4P7JZSOnHhhpTSR8B4YHBp01Dg5pTSp617VZIkSZKqhUMtJKn67ApsDTwdEQArAe/U2r8TcEtK6ROAiLi99LMbsAPwv6XjAFaoddz/ppTmN3C+q4BTgFuBI4Cj6heIiKOBowE6rdKrmZclSZIkqRqYjJKk6hPANSmlH9bZGHF4rdXUwHHLAR+UelM15OOGNqaUHo2I3hGxC9AppfRiA2WuIPe6YoW1N27o3JIkSZI6CIfpSVL1+StwQER8BiAiVo+I9Wvt/xvwtYhYKSK6A0MAUkofAq9HxDdKx0VEfLGR57wWuAH4XWtdhCRJkqTqZDJKkqpMSukl4DTg3oh4HrgPWLvW/onAjcAk4Gbg4VqHDwO+HRHPAZOBfRt52uuA1cgJKUmSJElaIpNRklSFUko3ppT6pZS2SCltnVJ6IqXUO6U0o7R/TEppk5TS7imlb6WULihtfz2ltGdK6YsppT4ppZ+Uth+eUvpjrfrHppS+V+uUA4A/ppQ+KPAy1RI/+xlssw2ssgr06gVDhsCL9UZY/ulPsMceeX8EjB+/7Hr/9CfYffd8TPfusO22cPvti5e76CLYdFNYaSVYd10YMQJmzarZf9118NnPwuqrw/e/X/fYN9+E3r3h3/9u6lVLkiSpApiMkiS1SET8CjgHOKvcsagJxo+H734XHnsMHngAOneG3XaD996rKfPxx7DDDnDhhY2v96GH4MtfhjvvhGefhb32gq99DR6u1QHv+uvhlFNg1CiYMgWuvRbuugtGjsz7Z8yAI4+ECy6Ae+6BP/wB7rij5vgRI+D002HNNVv0FkiSJKk8nMBcktQiKaVjyx2DmuGee+qu//730KMHPPpo7iUFcOih+eeMGY2v96KL6q6feWZOTN16K+y0U9722GOw3XY19ffuDYcdBjffnNenTs2xHHRQXh80KCet9t47l5k5E771rcbHJEmSpIpizyhJkgQffQQLFsBqq7VN3bXrHTAAJk2CJ57I62+8kYfy7bVXXt94Y/jkk9yz6r334OmnYYstchLq5JPhN7/JwwYlSZLULtkzSpIk5SFy/frB9tu3br2XXgrTp9f0ggIYOhTefRd23hlSgnnz8v5zz837V1sNrrkm95aaPTv/3GMPOOaYPHxvxgz45jfzMMKRI+E732ndmCVJktSmTEZJktTRff/78MgjeenUqfXqvfnm3JNp3DhYf/2a7Q89BGedBZddlic4f/XVnFQ680z4yU9yma99LS8LPfJI7kn185/DJpvkZFXfvrnH1I47wuabt17ckiRJalMmoyRJ6shOOCEnix58EDbcsPXqvfnm3Nvp2mthn33q7jvtNDj44NzLCXIi6eOP8/oZZ+TJ1GubOzf3frrqqjyf1Ny5ebJ1gIED82TsJqMkSZLaDeeMkiSpoxo5Mj/Z7oEHYNNNW6/em26CQw6BsWPhgAMW3//JJ4v3wOrUKQ/Za8iYMfkJfdttl+e1mjevZt/cuTB/fquFLkmSpLZnzyhJkjqiESPyE/RuvTXP0fT223l7t255gTx5+BtvwAcf5PVXX4VVV4W11soL5PmcIPeAgtzL6tBD4YIL8pxQC+vt0gVWXz2/HjIELrwQ+vevGaZ3+un5aXn1e0W99BJcd12ezBzyEL3OneHyy/Mwvb/+NR8rSZKkdsNklCRJHdFll+Wfu+5ad/uZZ8Lo0fn17bfDEUfU7DvqqMXLvPFG3eMvvzz3XDr++LwstMsueTgd5GF6ETmJNH069OyZE1RjxtStKyU4+mj4xS+ge/e8baWVchJtxIj8dL1Ro3JSS5IkSe1GpCV1iW+B/v37pwkTJrR6vZLU3kXEMymlDn3nvMLaG6e1h/+y3GF0GNPOGVzuECQ1ge2E9xKStDTV0k44Z5QkSZIkSZIKYzJKkiRJkiRJhTEZJUmSJEmSpMKYjJIkSZIkSVJhTEZJkiRJkiSpMCajJEmSJEmSVBiTUZIkSZIkSSqMyShJkiRJkiQVpnO5A5AkdSybr9ODCecMLncYkiRJksrEnlGSJEmSJEkqjMkoSZIkSZIkFcZklCRJkiRJkgpjMkqSJEmSJEmFMRklSZIkSZKkwpiMkiRJkiRJUmFMRkmSJEmSJKkwJqMkSZIkSZJUGJNRkiRJkiRJKozJKEmSJEmSJBWmc1tU+sKbM+l96p1tUbUkldW0cwaXOwRJkqqa9xKSqpn3E5k9oyRJkiRJklQYk1GSJEmSJEkqjMkoSZIkSZIkFcZklCRJkiRJkgpjMkqSJEmSJEmFMRklSZIkSZKkwpiMkiRJkiRJUmFMRkmSJEmSJKkwJqMkSZIkSZJUGJNRkiRJkiRJKozJKEmSJEmSJBXGZJQkSZIkSZIKYzJKkiRJkiRJhTEZJUmSJEmSpMKYjJIkSZIkSVJhTEZJkiRJkiSpMCajJEmSJEmSVBiTUZIkSZIkSSqMyShJkiRJkiQVxmSUJEmSJEmSCmMySpIkSZIkSYUxGSVJkiRJkqTCmIySJEmSJElSYUxGSZIkSZIkqTAmoyRJkiRJklQYk1GSJEmSJEkqjMkoSZIkSZIkFcZklCRJkiRJkgpjMkqSJEmSJEmFWWYyKiJ+GxHvRMSLRQQkSZIkqXpExGcj4sGImBIRkyNiZLljkiSVV2N6Ro0F9mzjOCRJkiRVp3nAiSmlzYDtgBER0afMMUmSymiZyaiU0t+A9wqIRZIkSVKVSSm9lVKaWHr9ETAFWKe8UUmSysk5oyRJkiQVIiJ6A1sCT5Y5FElSGXVurYoi4mjgaIBOq/RqrWolSVXmhTdn0vvUO8sdhiS1iWnnDC53CBUrIroBNwPHp5Q+rLfPewlJ6kBarWdUSumKlFL/lFL/Tl17tFa1kiRJktq5iFienIi6LqX0p/r7vZeQpI7FYXqSJEmS2kxEBHA1MCWldGG545Ekld8yk1ERcQPwOLBJREyPiG+3fViSJEmSqsSOwKHAlyNiUmnZq9xBSZLKZ5lzRqWUDi4iEEmSJEnVJ6X0CBDljkOSVDkcpidJkiRJkqTCmIySJEmSJElSYUxGSZIkSZIkqTAmoyRJkiRJklQYk1GSJEmSJEkqjMkoSWqp3r0hYvFl8OC8f9YsOPZYWHddZsNWRLxMxAlLrTNiLBGpgeXjWmWOIuJhIt4j4gMiHiRiQL16hhHxr1KZC+vtW4eIaUSs2SrvgyRJkqTmGz168XuKtdaq2X/66UyFvkR8TMT7RPyViB2WWmfE2kRcT8TfiZhPxNgGyixPxBlEvEbEHCKeI2LPemVa9b7CZJQktdTTT8Nbb9UsEyfmhuPAA/P+738f7rwTfv97vggvAmOAc4g4dCm1jgTWrrdMBW6qVWYgcCOwK7At8DJwDxEbAxDRE7gKOAnYAziEiL1rHX8pcBYp/btF1y9JkiSpdWyySd17ixdeqLPveHgD2BwYALwO3L2MJNAKwAzgHODJJZQ5G/gOcBzQB7gcuIWILYE2ua8wGSVJLdWrV/7GYuFy112wyirwjW/k/Y89BoceCoMG8Q+YS0rXAk+QE0gNS2kmKb29aIGNgA2BK2uVGUZKl5DSs6T0MvD/gI+Ahd9ibAjMJKUbSelp4EFgMwAivg70AH7biu+EJEmSpJbo3LnuvUWvXjX7DjmE2+EjUppKSpOB7wPdgX5LrC+laaR0HCmNBd5bQqlDgXNJ6c5S3b8G7gJOLO1v9fsKk1GS1JpSgquvhkMOga5d87YBA+DPf4Z//Suv5660/YC7m1DzUcBkUnpsKWW6ACsC75fWXwG6ErElEasD2wDPE9EDOB84hpRSE2KQJEmS1JamToV11oENNoChQ/N6QyK6AEcDHwKTWnjWFYA59bbNJve+gja4rzAZJUmt6b774PXX4cgja7ZdfDH06wfrrcensBXwEPADUrqjUXXmD/lvULtXVMPOBmYBtwOQ0vvAcOBa4CngWlK6BziP3M22JxETiJhCxHcafY2SJEmSWt+228LYsfCXv8CVV8Lbb8MOO8C77y4qMhR6EDGLnDw6AfhKK0y7cQ9wPBGbELEcEV8B9idPFdIm9xWdWxiwJKm2K6+EbbbJyaeFfvUrePRRuP12dtxnnylPwq+BC4iYRkqN6R11CNAJ+P0SS0SMBI4BdiOlDxdtT+kW4JZa5QYA25G73L5MblQmk7/ZeJSUag1KlyRJklSYr3617vp228GGG8I11+R5aIE787QcXwV6kkdP3ETE9qT0VgvOPJL8xfdLQAJeA34HHLGoRCvfV9gzSpJayzvvwG23wVFH1WybPRt++EM47zwYMoSnYDYpXQKMI08A2BhHATeTUsNjvHMi6mxgL1J6aom15K68l5OTVhsCXUjp/lLDNZ48IbokSZKkStCtG/TtC6+8smjTR7CAlF4lpSdI6dvAp8CRS6yjMVL6DyntB6wMrA9sSh5x8XqD5VvhvsJklCS1lt/9DlZYIY/tXujTT/PSqVP90vNpzGdwxLbAF1nSEL2I75OfzjeYlB5ZRm2jgAdI6YnSuWv3ju1C7n0lSZIkqRLMmQN//zusvfbSSi1HnvOp5VKaQ0pvku8Tvg7ctoSSLb6vcJieJLWGlOCqq3Iiqnv3mu2rrAK77AKnngrdurEJdCHicOAw4JRF5SKuLdVzWL2ajyJPGPjQYueMOJmciDoE+AcRa5X2zCalmfXK9gGGAVuWtrwMzCuN6Z4M7Aqc1eTrliRJktQ6TjoJhgyB9dbLoy7OOgs+/hiGD4cPP4TzzmMQrEzEekAvYASwLnDTojoauq+I6Fd6tQqwoLQ+l5ReKu3fFliHPBH6OsBocpLpvMVibKX7CpNRktQaxo+HV1+F665bfN+4cXmo3rBhTIIvAKcCpwOX1Cq13mLHRXQHhgI/WcLTKUYAywM31tt+DXB4rXoCuAI4gZQ+AiCl2UQcClxKfhTrGFKaUOuQw4H+KaXvLemSI6I3cEdK6QuRG7T/SSndtaTykiRJkpZi+nQ4+GCYMQN69cpzRj3xBKy/PnzyCUyezDjYiPxl9bvA08DOpPR8rVoWv6+AZ+utDwH+CfQura9InvZjQ/LwvLuAQ0npgzpHNeO+YklMRklSaxg0KPeOashaa+UhfMBKERNTSv0XK5PSwAa2fQR0W+I5U+rdqNhyImtAA9vvJjdmraEf0J/ccEmSJElqqnHjlryva1e45RbWjHi+wfuJhRq+r4ilnjelh4A+y4yvFe8rnDNKkjqgiLg1Ip6JiMkRcXRp2xER8Y+IeAjYsVbZsRFxQK31WfXq6gL8BDgoIiZFxEEFXYYkSZKkdsieUZLUMX0rpfReRKwEPB0RdwI/BrYGZgIPsnh33gallOZGxBksY1ifJEmSJIE9oySpozouIp4DngA+CxwKjE8p/SelNJfF56FqkYg4OiImRMSE+Z/MXPYBkiRJkqqWyShJ6mAiYiCwG7B9SumL5B5QfweWMOkV8yi1F5EnLezS1HOmlK5IKfVPKfXv1LVHc8KWJEmSVCVMRklSx9MDeD+l9ElEbApsB6wEDIyINSJieeAbtcpPIw/fA9iX/AS/+j4CurddyJIkSZKqhckoSep47gY6R8TzwFnkoXpvAaOBx4H7gYm1yl8J7BIRTwHbAh83UOeDQB8nMJekJejdGyIWXwYPhk8/hR/8ALbYgk9gSyLeIuJ6Ihp6PHeNiF2IeIyId4mYTcTfiThpKeUPJiIRcUe97cOI+BcR7xFxYb196xAxjYg1m33tkiTV4wTmktTBpJT+C3y1gV3jgd81UP7f5N5TC/2wtH0a8IXS6/eAbVo5VEmqHk8/DfPn16y/9RZsvTUceCB88glMnAijRrHd0KEvPQffBn4O3E3EFqQ0bwm1zgIuBl4APiE/CfU3RHxCSpfVKRmxIXA+8HC97T2Bq4DDganAnUQ8QEoLE1aXAmeR2wJJklqFyShJkiSprfXqVXf96qthlVXgG9+Arl3hvvsAeH7o0P+S0lNEHANMBjYjJ5sWl9IzwDO1trxOxP7ATkBNMioPv74BGAUMAnrWOmZDYCYp3Vgq+2DpnHcQ8XXy0O7fNuuaJUlaAofpSZIkSUVKKSejDjkkJ6Iatkrp5/uNrjdiS2AH4KF6e8YA00jpmgaOegXoSsSWRKxO7uX6PBE9yD2pjiGlJT3gQpKkZrFnlCRJklSk++6D11+HI49seH9EF/IwvT+T0vRl1hcxHehF/r/9j0np8lr7dgcOAvo1eGxK7xMxHLiW/DCLa0npHiJ+Qx6+15OI64GVgYvq1C1JUjOZjJIkSZKKdOWVsM020K/fYrtK/zn/A7AqsE8ja9wJ6Eae3+9cIl4npd+X5oMaC3yTlJbcwyqlW4BbFq1HDCjVdSLwMjCcPGTweSIeJaWGhw1KktRIJqMkSZKkorzzDtx2G1x66eL75s3j9jyHUzdgICm926g6U3q99OqF0lPvRgO/Jz9kYm3gfiIWls7TdETMA/qS0st16sq9si4HjiTH0oWU7i/tGw8MZElzWEmS1EgmoyRJkqSi/O53sMIKMHRo3e2ffgpDh7IZdAUGkdLbzTzDcsAKpddPA5vX2382sBowAnidxY0CHiClJ4joR937hS5Ap2bGJUnSIm2SjNp8nR5MOGdwW1QtSZIktU8pwVVX5URU9+412+fNy0/Ve/ppDoGpj0AiYq3S3pmkNBuAiGtL9RxWWj+WnFBa2LtpZ+AkFj5JL6WPgRfrxBDxAdCZlOpuz/v6AMOALUtbXgbmEfEd8jC9XYGzmnv5jeW9hCRVP3tGSZIkSUUYPx5efRWuu67u9unT89A94BHYDHir1t4jyPM+AaxXr8ZOwLlAb2Ae8BpwKnmYXdNEBHAFcAIpfQRASrOJOBS4FOgBjCGlCU2uW5KkekxGSZIkSUUYNCj3jqqvd+9F2yPimZRS/waPT2lgvfVfAr9sUgwpHb6E7QkY0MD2u4GNmnQOSZKWYblyByBJkiRJkqSOw2SUJEmSJEmSCmMySpIkSZIkSYUxGSVJkiRJkqTCmIySJEmSJElSYUxGSZIkSZIkqTAmoyRJkiRJklQYk1GSJEmSJEkqTOdyByBJ6lg2X6cHE84ZXO4wJEmSJJWJPaMkSZIkSZJUGJNRkiRJkiRJKozJKEmSJEmSJBXGZJQkSZIkSZIKYzJKkiRJkiRJhTEZJUmSJEmSpMKYjJIkSZIkSVJhTEZJkiRJkiSpMCajJEmSJEmSVBiTUZIkSZIkSSqMyShJkiRJkiQVxmSUJEmSJEmSCmMySpIkSZIkSYUxGSVJkiRJkqTCmIySJEmSJElSYUxGSZIkSZIkqTAmoyRJkiRJklQYk1GSJEmSJEkqjMkoSZIkSZIkFcZklCRJkiRJkgpjMkqSJEmSJEmFMRklSZIkSZKkwkRKqfUrjfgIeLnVK24dPYEZ5Q5iKYyv+So5NjC+lqrk+JoS2/oppV5tGUylq/A2oiGV/LtXn7G2jfYUK7SveI11cbYT7a+daIn29DfQUl5r9eko1wmVda1V0U50bqN6X04p9W+julskIiZUamxgfC1RybGB8bVUJcdXybFVqIptIxrSnv59jbVttKdYoX3Fa6xagnbVTrRER/q98lqrT0e5TuhY11oUh+lJkiRJkiSpMCajJEmSJEmSVJi2SkZd0Ub1toZKjg2MryUqOTYwvpaq5PgqObZK1N7er/YUr7G2jfYUK7SveI1VDelI77XXWp06yrV2lOuEjnWthWiTCcwlSZIkSZKkhjhMT5IkSZIkSYVpdjIqIvaMiJcj4tWIOLWB/RERF5f2Px8RW7Us1FaPb9OIeDwi/hsRJxUZWyPjG1Z6356PiMci4osVFNu+pbgmRcSEiBhQVGyNia9WuW0iYn5EHFBJ8UXEwIiYWXr/JkXEGZUSW634JkXE5Ih4qKjYGhNfRJxc6317sfTvu3oFxdcjIv4cEc+V3r8jioqtElV6O1EvlopuMxqIp2LbkAZiqeg2pV4sFd2+1IuhYtuaBmKp6LangVgqui2qJu2pnWip9tbOtER7aqNaoj21by3VntrHlmpP7Wu7l1Jq8gJ0Al4DNgS6AM8BfeqV2Qv4CxDAdsCTzTlXG8b3GWAbYAxwUlGxNSG+HYDVSq+/WtT718jYulEzxHML4O+V9N7VKvcAcBdwQCXFBwwE7ijyd64Jsa0KvASsV1r/TCXFV6/8EOCBSooP+BFwbul1L+A9oEvR/9aVsFR6O9GMWMvWZjQz3rK0Ic2MtWxtSlNjrVWu8PalGe9rWdqaZsZatranub8HtcoX2hZV09Ke2omCrrVi2pkCrrUi2qgCrrMi2rcirrVWubK2jwX9u1ZE+1oNS3N7Rn0JeDWlNDWlNBcYB+xbr8y+wLUpewJYNSLWbub5Wj2+lNI7KaWngU8Liqmp8T2WUnq/tPoEsG4FxTYrlf4SgZWBIicea8zvHsCxwM3AOwXGBo2PrxwaE9s3gT+llN6A/HdSYfHVdjBwQyGRZY2JLwHdIyLI/wF5D5hXYIyVpNLbidoqvc2or5LbkPoqvU2prdLbl9oqua2pr9LbnvoqvS2qJu2pnWip9tbOtER7aqNaoj21by3VntrHlmpP7Wu719xk1DrAv2qtTy9ta2qZtlLOczdGU+P7NvlboSI0KraI+FpE/B24E/hWQbFBI+KLiHWArwGXFxjXQo39t90+8lCuv0RE32JCa1RsnwdWi4jxEfFMRBxWUGzQhL+LiOgK7Elu8IrSmPguATYD/g94ARiZUlpQTHgVp9LbiUqMo7EquQ2pr9LblNoqvX2prZLbmvoqve2pr9LbomrSntqJlqqW62iM9tRGtUR7at9aqj21jy3VntrXdq9zM4+LBrbVz/Q2pkxbKee5G6PR8UXEIPKHdFFjjBsVW0rpFuCWiNgZOAvYra0DK2lMfL8EfpBSmp87qBSqMfFNBNZPKc2KiL2AW4GN2zowGhdbZ2BrYFdgJeDxiHgipfSPtg6Opv3dDgEeTSm914bx1NeY+PYAJgFfBjYC7ouIh1NKH7ZxbJWo0tuJ2ioljsaq5DZksRAa2FZJbUptld6+1FbJbU19ld721FfpbVE1aU/tREtVy3U0Rntqo1qiPbVvLdWe2seWak/ta7vX3J5R04HP1lpfl9wToKll2ko5z90YjYovIrYArgL2TSm9W0mxLZRS+huwUUT0bOvAShoTX39gXERMAw4ALouI/QqJrhHxpZQ+TCnNKr2+C1i+oPevsX+3d6eUPk4pzQD+BhQ1qWRTfveGUvywiMbEdwR5qElKKb0KvA5sWlB8labS24lKjKOxKrkNqa/S25TaKr19qa2S25r6Kr3tqa/S26Jq0p7aiZaqlutojPbURrVEe2rfWqo9tY8t1Z7a1/YvNW9ir87AVGADaib26luvzGDqTjj4VHPO1Vbx1So7muInMG/M+7ce8CqwQwXG9jlqJuPbCnhz4XolxFev/FiKncC8Me/fWrXevy8BbxTx/jUyts2Av5bKdgVeBL5QKe9dqVwP8lxMKxf179qE9+/XwOjS6zVLfxs9i4yzUpZKbyea87tXKlt4m9HM97YsbUgzYy1bm9Lc34NS+ULbl2a8r2Vpa5oZa9nanub+HpSrLaqmpT21E0Vca62yZW9nCvh3rYg2qoDrrIj2rYhrrVe+bO1jQf+uFdG+VsPSrGF6KaV5EfE94B7yjPO/TSlNjojvlPZfTp5Ff6/SB80n5B4DhWhMfBGxFjABWAVYEBHHk2fKb/PhNI18/84A1iBnlQHmpZT6V0hsXwcOi4hPgdnAQan011gh8ZVNI+M7APh/ETGP/P4NLeL9a0xsKaUpEXE38DywALgqpfRiW8fW2PhKRb8G3JtS+riIuJoY31nA2Ih4gfwf5x+k/C1/h1Pp7URTYy1nm9GceClTG9LMWMvWpjQj1opQyW1Nc2ItZ9vTnHhLRcvSFlWT9tROtFR7a2daoj21US3Rntq3lmpP7WNLtaf2tRqE75skSZIkSZKK0tw5oyRJkiRJkqQmMxklSZIkSZKkwpiMkiRJkiRJUmFMRkmSJEmSJKkwJqMkSZIkSZJUGJNRkiRJkiRJKozJKEmSJEmSJBXGZJQkSZIkSZIK8/8Bdrpcn9It4GoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1440x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_ccas(final_test_volumes, ['ETHNIC_ID','AGE_GROUP','SEX'], 'Test')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T21:12:01.625602Z",
     "start_time": "2020-11-11T21:12:01.467183Z"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-12T10:11:31.427948Z",
     "start_time": "2020-11-12T10:11:31.400084Z"
    }
   },
   "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>IXI_ID</th>\n",
       "      <th>SEX</th>\n",
       "      <th>ETHNIC_ID</th>\n",
       "      <th>AGE</th>\n",
       "      <th>N_Relevant_Slices</th>\n",
       "      <th>AGE_GROUP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>35.800137</td>\n",
       "      <td>106.0</td>\n",
       "      <td>adult</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.781656</td>\n",
       "      <td>109.0</td>\n",
       "      <td>adult</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>46.710472</td>\n",
       "      <td>101.0</td>\n",
       "      <td>adult</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>14</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>34.236824</td>\n",
       "      <td>97.0</td>\n",
       "      <td>adult</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>24.284736</td>\n",
       "      <td>102.0</td>\n",
       "      <td>young</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>529</th>\n",
       "      <td>88</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>530</th>\n",
       "      <td>345</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>531</th>\n",
       "      <td>347</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>532</th>\n",
       "      <td>637</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>533</th>\n",
       "      <td>638</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>534 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     IXI_ID  SEX ETHNIC_ID        AGE  N_Relevant_Slices AGE_GROUP\n",
       "0         2    2         1  35.800137              106.0     adult\n",
       "1        12    1         1  38.781656              109.0     adult\n",
       "2        13    1         1  46.710472              101.0     adult\n",
       "3        14    2         1  34.236824               97.0     adult\n",
       "4        15    1         2  24.284736              102.0     young\n",
       "..      ...  ...       ...        ...                ...       ...\n",
       "529      88  NaN       NaN        NaN                NaN       NaN\n",
       "530     345  NaN       NaN        NaN                NaN       NaN\n",
       "531     347  NaN       NaN        NaN                NaN       NaN\n",
       "532     637  NaN       NaN        NaN                NaN       NaN\n",
       "533     638  NaN       NaN        NaN                NaN       NaN\n",
       "\n",
       "[534 rows x 6 columns]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_train_val_volumes"
   ]
  },
  {
   "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": "3.7.9"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Table of Contents",
   "toc_cell": true,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  },
  "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
}