lesson03/mnist/mnist2-kaggle/mnist02-download-and-convert.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "875bcbdb-b248-454d-8f91-ffc366278ef0",
   "metadata": {},
   "source": [
    "# Downloading and Converting the Kaggle MNIST-dataset\n",
    "\n",
    "Since the Kaggle dataset contains the images of the digits in text files, I convert the data into images in this notebook, so that the conversion process does not clutter up other notebooks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "b8befe84-bd03-43c8-a841-365fce78d716",
   "metadata": {},
   "outputs": [],
   "source": [
    "from fastai.vision.all import *"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0948d5bb-2eb3-467b-b762-24ccf91a5244",
   "metadata": {},
   "source": [
    "## Downloading the dataset from Kaggle\n",
    "\n",
    "Let's download the dataset from Kaggle. Uncomment the next line of you do not have it on your machine yet:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e9c3f2a5-4e00-42d5-8389-366371a602b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#!kaggle competitions download -c digit-recognizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "72c97bb0-a488-4ea1-8246-3ffa6e7c8e7a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#!unzip digit-recognizer.zip"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4cdb241-e4ce-44bd-8877-0cf541c23dc2",
   "metadata": {},
   "source": [
    "As presented in the book, let's do all the file manipulation directly from the notebook:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d6b254f3-8ab3-4a27-9c9a-2fd6ed563d86",
   "metadata": {},
   "outputs": [],
   "source": [
    "path = Path('.')\n",
    "Path.BASE_PATH = path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c5bbfbbc-f480-4690-b5cb-c98f912dc6e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(#14) [Path('.ipynb_checkpoints'),Path('digit-recognizer.zip'),Path('mnist02-download-and-convert.ipynb'),Path('MNIST02-dummy.ipynb'),Path('MNIST02-temp.ipynb'),Path('MNIST02.ipynb'),Path('MNIST02a.ipynb'),Path('sample_submission.csv'),Path('sub1-099442.PNG'),Path('submission.csv')...]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path.ls()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8bded2e3-a8c3-4784-900e-6632dea9322d",
   "metadata": {},
   "source": [
    "## Inspecting the data\n",
    "\n",
    "Let's have a look at the data so that we can convert it to actual files. For a start, let's load the csv into a dataframe:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "083af398-ae1b-4f9c-a15d-829f94cd8808",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label</th>\n",
       "      <th>pixel0</th>\n",
       "      <th>pixel1</th>\n",
       "      <th>pixel2</th>\n",
       "      <th>pixel3</th>\n",
       "      <th>pixel4</th>\n",
       "      <th>pixel5</th>\n",
       "      <th>pixel6</th>\n",
       "      <th>pixel7</th>\n",
       "      <th>pixel8</th>\n",
       "      <th>...</th>\n",
       "      <th>pixel774</th>\n",
       "      <th>pixel775</th>\n",
       "      <th>pixel776</th>\n",
       "      <th>pixel777</th>\n",
       "      <th>pixel778</th>\n",
       "      <th>pixel779</th>\n",
       "      <th>pixel780</th>\n",
       "      <th>pixel781</th>\n",
       "      <th>pixel782</th>\n",
       "      <th>pixel783</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 785 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   label  pixel0  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  \\\n",
       "0      1       0       0       0       0       0       0       0       0   \n",
       "1      0       0       0       0       0       0       0       0       0   \n",
       "2      1       0       0       0       0       0       0       0       0   \n",
       "3      4       0       0       0       0       0       0       0       0   \n",
       "4      0       0       0       0       0       0       0       0       0   \n",
       "\n",
       "   pixel8  ...  pixel774  pixel775  pixel776  pixel777  pixel778  pixel779  \\\n",
       "0       0  ...         0         0         0         0         0         0   \n",
       "1       0  ...         0         0         0         0         0         0   \n",
       "2       0  ...         0         0         0         0         0         0   \n",
       "3       0  ...         0         0         0         0         0         0   \n",
       "4       0  ...         0         0         0         0         0         0   \n",
       "\n",
       "   pixel780  pixel781  pixel782  pixel783  \n",
       "0         0         0         0         0  \n",
       "1         0         0         0         0  \n",
       "2         0         0         0         0  \n",
       "3         0         0         0         0  \n",
       "4         0         0         0         0  \n",
       "\n",
       "[5 rows x 785 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "train = pd.read_csv(\"train.csv\")\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ce8b6d6c-f00d-46d9-86f5-b2f71fd14768",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "42000"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "af6b802c-79ea-4ac3-878c-0800399d69a4",
   "metadata": {},
   "source": [
    "## Displaying one image\n",
    "\n",
    "To display one image, we can grab it from the csv-file. Essentiall, we pick a row and get all the pixel columns:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "195fe5bf-5376-4427-85a5-9543118c4917",
   "metadata": {},
   "outputs": [],
   "source": [
    "im = torch.tensor(train.iloc[1,1:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "abcdb743-04cc-44f5-ad8e-453615069eb9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "784"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(im)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4074b4b6-9e4a-481d-ada4-d9883f8a2bea",
   "metadata": {},
   "source": [
    "All images are 28x28 pixels:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "48bf2827-f1ac-402c-b23d-b1243607c245",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "28.0"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "math.sqrt(len(im))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "411ef402-7f1b-4159-998d-88c662b0a184",
   "metadata": {},
   "source": [
    "Just displaying the raw data does not look nice, but if we show a fraction, we can see the shape of a number emerging:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "64f77906-7b6b-4efe-a3a9-2cb6804ba446",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
       "        [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
       "        [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
       "        [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
       "        [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  18,  30, 137, 137],\n",
       "        [  0,   0,   0,   0,   0,   0,   0,   0,  13,  86, 250, 254, 254, 254],\n",
       "        [  0,   0,   0,   0,   0,   0,   0,  16, 179, 254, 254, 254, 254, 254],\n",
       "        [  0,   0,   0,   0,   0,   0,   0,  72, 254, 254, 254, 254, 254, 254],\n",
       "        [  0,   0,   0,   0,   0,   0,  61, 191, 254, 254, 254, 254, 254, 109],\n",
       "        [  0,   0,   0,   0,   0,   0, 172, 254, 254, 254, 202, 147, 147,  45],\n",
       "        [  0,   0,   0,   0,   0,   1, 174, 254, 254,  89,  67,   0,   0,   0],\n",
       "        [  0,   0,   0,   0,   0,  47, 254, 254, 254,  29,   0,   0,   0,   0],\n",
       "        [  0,   0,   0,   0,   0,  80, 254, 254, 240,  24,   0,   0,   0,   0],\n",
       "        [  0,   0,   0,   0,   0,  64, 254, 254, 186,   7,   0,   0,   0,   0]])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "im2 = torch.zeros((27,27), dtype=int)\n",
    "for i in range(27):\n",
    "    im2[i] = im[i*28:((i+1)*28)-1]\n",
    "im2[:14,:14]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0a3a04f5-2a36-40b7-90a3-1b0fea57896a",
   "metadata": {},
   "source": [
    "Using the elegant visualization from [chapter 4](https://github.com/fastai/fastbook/blob/master/04_mnist_basics.ipynb), here is a zero:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "06273eb1-c73d-4c6a-8768-3e353f5db164",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_951f5_row0_col0, #T_951f5_row0_col1, #T_951f5_row0_col2, #T_951f5_row0_col3, #T_951f5_row0_col4, #T_951f5_row0_col5, #T_951f5_row0_col6, #T_951f5_row0_col7, #T_951f5_row0_col8, #T_951f5_row0_col9, #T_951f5_row0_col10, #T_951f5_row0_col11, #T_951f5_row0_col12, #T_951f5_row0_col13, #T_951f5_row0_col14, #T_951f5_row0_col15, #T_951f5_row0_col16, #T_951f5_row0_col17, #T_951f5_row0_col18, #T_951f5_row0_col19, #T_951f5_row1_col0, #T_951f5_row1_col1, #T_951f5_row1_col2, #T_951f5_row1_col3, #T_951f5_row1_col4, #T_951f5_row1_col5, #T_951f5_row1_col6, #T_951f5_row1_col7, #T_951f5_row1_col8, #T_951f5_row1_col9, #T_951f5_row1_col10, #T_951f5_row1_col11, #T_951f5_row1_col12, #T_951f5_row1_col13, #T_951f5_row1_col14, #T_951f5_row1_col15, #T_951f5_row1_col16, #T_951f5_row1_col17, #T_951f5_row1_col18, #T_951f5_row1_col19, #T_951f5_row2_col0, #T_951f5_row2_col1, #T_951f5_row2_col2, #T_951f5_row2_col3, #T_951f5_row2_col4, #T_951f5_row2_col5, #T_951f5_row2_col6, #T_951f5_row2_col7, #T_951f5_row2_col15, #T_951f5_row2_col16, #T_951f5_row2_col17, #T_951f5_row2_col18, #T_951f5_row2_col19, #T_951f5_row3_col0, #T_951f5_row3_col1, #T_951f5_row3_col2, #T_951f5_row3_col3, #T_951f5_row3_col4, #T_951f5_row3_col5, #T_951f5_row3_col17, #T_951f5_row3_col18, #T_951f5_row3_col19, #T_951f5_row4_col0, #T_951f5_row4_col1, #T_951f5_row4_col2, #T_951f5_row4_col3, #T_951f5_row4_col4, #T_951f5_row4_col19, #T_951f5_row5_col0, #T_951f5_row5_col1, #T_951f5_row5_col2, #T_951f5_row5_col3, #T_951f5_row5_col4, #T_951f5_row5_col19, #T_951f5_row6_col0, #T_951f5_row6_col1, #T_951f5_row6_col2, #T_951f5_row6_col3, #T_951f5_row7_col0, #T_951f5_row7_col1, #T_951f5_row7_col2, #T_951f5_row7_col3, #T_951f5_row7_col12, #T_951f5_row8_col0, #T_951f5_row8_col1, #T_951f5_row8_col2, #T_951f5_row8_col3, #T_951f5_row8_col9, #T_951f5_row8_col10, #T_951f5_row8_col11, #T_951f5_row8_col12, #T_951f5_row8_col13, #T_951f5_row8_col14, #T_951f5_row9_col0, #T_951f5_row9_col1, #T_951f5_row9_col2, #T_951f5_row9_col8, #T_951f5_row9_col9, #T_951f5_row9_col10, #T_951f5_row9_col11, #T_951f5_row9_col12, #T_951f5_row9_col13, #T_951f5_row9_col14, #T_951f5_row9_col15, #T_951f5_row10_col0, #T_951f5_row10_col1, #T_951f5_row10_col2, #T_951f5_row10_col8, #T_951f5_row10_col9, #T_951f5_row10_col10, #T_951f5_row10_col11, #T_951f5_row10_col12, #T_951f5_row10_col13, #T_951f5_row10_col14, #T_951f5_row10_col15, #T_951f5_row11_col0, #T_951f5_row11_col1, #T_951f5_row11_col2, #T_951f5_row11_col8, #T_951f5_row11_col9, #T_951f5_row11_col10, #T_951f5_row11_col11, #T_951f5_row11_col12, #T_951f5_row11_col13, #T_951f5_row11_col14, #T_951f5_row11_col15, #T_951f5_row11_col16, #T_951f5_row12_col0, #T_951f5_row12_col1, #T_951f5_row12_col8, #T_951f5_row12_col9, #T_951f5_row12_col10, #T_951f5_row12_col11, #T_951f5_row12_col12, #T_951f5_row12_col13, #T_951f5_row12_col14, #T_951f5_row12_col15, #T_951f5_row12_col16, #T_951f5_row13_col0, #T_951f5_row13_col1, #T_951f5_row13_col8, #T_951f5_row13_col9, #T_951f5_row13_col10, #T_951f5_row13_col11, #T_951f5_row13_col12, #T_951f5_row13_col13, #T_951f5_row13_col14, #T_951f5_row13_col15, #T_951f5_row13_col16, #T_951f5_row14_col0, #T_951f5_row14_col1, #T_951f5_row14_col8, #T_951f5_row14_col9, #T_951f5_row14_col10, #T_951f5_row14_col11, #T_951f5_row14_col12, #T_951f5_row14_col13, #T_951f5_row14_col14, #T_951f5_row14_col15, #T_951f5_row14_col16, #T_951f5_row15_col0, #T_951f5_row15_col1, #T_951f5_row15_col2, #T_951f5_row15_col9, #T_951f5_row15_col10, #T_951f5_row15_col11, #T_951f5_row15_col12, #T_951f5_row15_col13, #T_951f5_row15_col14, #T_951f5_row15_col15, #T_951f5_row16_col0, #T_951f5_row16_col1, #T_951f5_row16_col2, #T_951f5_row16_col10, #T_951f5_row16_col11, #T_951f5_row16_col12, #T_951f5_row16_col13, #T_951f5_row16_col14, #T_951f5_row17_col0, #T_951f5_row17_col1, #T_951f5_row17_col2, #T_951f5_row17_col3 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #ffffff;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row2_col8, #T_951f5_row4_col5, #T_951f5_row15_col16 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #f7f7f7;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row2_col9, #T_951f5_row7_col14, #T_951f5_row9_col7, #T_951f5_row12_col7, #T_951f5_row13_col7, #T_951f5_row14_col7 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #f1f1f1;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row2_col10, #T_951f5_row2_col11 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #8a8a8a;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row2_col12 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #505050;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row2_col13, #T_951f5_row3_col7 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #c5c5c5;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row2_col14, #T_951f5_row5_col5 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #d2d2d2;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row3_col6, #T_951f5_row15_col8 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #f9f9f9;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row3_col8 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #050505;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row3_col9, #T_951f5_row3_col10, #T_951f5_row3_col11, #T_951f5_row3_col12, #T_951f5_row4_col7, #T_951f5_row4_col8, #T_951f5_row4_col9, #T_951f5_row4_col10, #T_951f5_row4_col11, #T_951f5_row4_col12, #T_951f5_row4_col13, #T_951f5_row4_col14, #T_951f5_row4_col15, #T_951f5_row5_col6, #T_951f5_row5_col7, #T_951f5_row5_col8, #T_951f5_row5_col9, #T_951f5_row5_col10, #T_951f5_row5_col11, #T_951f5_row5_col12, #T_951f5_row5_col13, #T_951f5_row5_col14, #T_951f5_row5_col15, #T_951f5_row5_col16, #T_951f5_row5_col17, #T_951f5_row6_col6, #T_951f5_row6_col7, #T_951f5_row6_col8, #T_951f5_row6_col9, #T_951f5_row6_col10, #T_951f5_row6_col14, #T_951f5_row6_col15, #T_951f5_row6_col16, #T_951f5_row6_col17, #T_951f5_row7_col5, #T_951f5_row7_col6, #T_951f5_row7_col7, #T_951f5_row7_col16, #T_951f5_row7_col17, #T_951f5_row7_col18, #T_951f5_row8_col5, #T_951f5_row8_col6, #T_951f5_row8_col17, #T_951f5_row8_col18, #T_951f5_row9_col4, #T_951f5_row9_col5, #T_951f5_row9_col6, #T_951f5_row9_col17, #T_951f5_row9_col18, #T_951f5_row9_col19, #T_951f5_row10_col4, #T_951f5_row10_col5, #T_951f5_row10_col18, #T_951f5_row10_col19, #T_951f5_row11_col4, #T_951f5_row11_col5, #T_951f5_row11_col18, #T_951f5_row11_col19, #T_951f5_row12_col4, #T_951f5_row12_col5, #T_951f5_row12_col6, #T_951f5_row12_col18, #T_951f5_row12_col19, #T_951f5_row13_col2, #T_951f5_row13_col3, #T_951f5_row13_col4, #T_951f5_row13_col5, #T_951f5_row13_col6, #T_951f5_row13_col18, #T_951f5_row13_col19, #T_951f5_row14_col4, #T_951f5_row14_col5, #T_951f5_row14_col6, #T_951f5_row14_col18, #T_951f5_row14_col19, #T_951f5_row15_col4, #T_951f5_row15_col5, #T_951f5_row15_col6, #T_951f5_row15_col18, #T_951f5_row15_col19, #T_951f5_row16_col5, #T_951f5_row16_col6, #T_951f5_row16_col7, #T_951f5_row16_col17, #T_951f5_row16_col18, #T_951f5_row17_col6, #T_951f5_row17_col7, #T_951f5_row17_col8, #T_951f5_row17_col16, #T_951f5_row17_col17, #T_951f5_row17_col18 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #000000;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row3_col13 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #2c2c2c;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row3_col14 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #090909;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row3_col15 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #7b7b7b;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row3_col16 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #f0f0f0;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row4_col6 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #5e5e5e;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row4_col16 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #1b1b1b;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row4_col17 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #e0e0e0;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row4_col18, #T_951f5_row16_col3 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #f8f8f8;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row5_col18 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #b3b3b3;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row6_col4 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #dbdbdb;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row6_col5 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #515151;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row6_col11 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #adadad;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row6_col12, #T_951f5_row9_col16 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #c8c8c8;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row6_col13 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #464646;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row6_col18 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #0d0d0d;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row6_col19 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #c6c6c6;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row7_col4 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #656565;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row7_col8, #T_951f5_row16_col8 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #414141;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row7_col9, #T_951f5_row7_col10 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #7f7f7f;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row7_col11 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #e7e7e7;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row7_col13 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #fafafa;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row7_col15, #T_951f5_row15_col7 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #444444;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row7_col19, #T_951f5_row17_col19 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #666666;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row8_col4 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #636363;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row8_col7 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #c3c3c3;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row8_col8 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #d6d6d6;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row8_col15 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #949494;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row8_col16 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #020202;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row8_col19, #T_951f5_row17_col5 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #333333;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row9_col3 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #e5e5e5;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row10_col3 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #cbcbcb;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row10_col6, #T_951f5_row10_col17 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #101010;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row10_col7 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #f4f4f4;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row10_col16 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #f3f3f3;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row11_col3 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #d9d9d9;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row11_col6 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #565656;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row11_col7 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #fcfcfc;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row11_col17 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #6b6b6b;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row12_col2 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #474747;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row12_col3 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #1a1a1a;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row12_col17 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #cfcfcf;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row13_col17, #T_951f5_row14_col17, #T_951f5_row17_col10, #T_951f5_row17_col11, #T_951f5_row17_col14 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #e4e4e4;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row14_col2 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #f2f2f2;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row14_col3 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #6e6e6e;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row15_col3 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #bebebe;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row15_col17, #T_951f5_row17_col15 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #383838;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row16_col4 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #3c3c3c;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row16_col9 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #d7d7d7;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row16_col15 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #f5f5f5;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row16_col16 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #707070;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row16_col19 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #0a0a0a;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row17_col4 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #dcdcdc;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row17_col9 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #4d4d4d;\n",
       "  color: #f1f1f1;\n",
       "}\n",
       "#T_951f5_row17_col12 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #eeeeee;\n",
       "  color: #000000;\n",
       "}\n",
       "#T_951f5_row17_col13 {\n",
       "  font-size: 6pt;\n",
       "  background-color: #e9e9e9;\n",
       "  color: #000000;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_951f5\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_951f5_level0_col0\" class=\"col_heading level0 col0\" >0</th>\n",
       "      <th id=\"T_951f5_level0_col1\" class=\"col_heading level0 col1\" >1</th>\n",
       "      <th id=\"T_951f5_level0_col2\" class=\"col_heading level0 col2\" >2</th>\n",
       "      <th id=\"T_951f5_level0_col3\" class=\"col_heading level0 col3\" >3</th>\n",
       "      <th id=\"T_951f5_level0_col4\" class=\"col_heading level0 col4\" >4</th>\n",
       "      <th id=\"T_951f5_level0_col5\" class=\"col_heading level0 col5\" >5</th>\n",
       "      <th id=\"T_951f5_level0_col6\" class=\"col_heading level0 col6\" >6</th>\n",
       "      <th id=\"T_951f5_level0_col7\" class=\"col_heading level0 col7\" >7</th>\n",
       "      <th id=\"T_951f5_level0_col8\" class=\"col_heading level0 col8\" >8</th>\n",
       "      <th id=\"T_951f5_level0_col9\" class=\"col_heading level0 col9\" >9</th>\n",
       "      <th id=\"T_951f5_level0_col10\" class=\"col_heading level0 col10\" >10</th>\n",
       "      <th id=\"T_951f5_level0_col11\" class=\"col_heading level0 col11\" >11</th>\n",
       "      <th id=\"T_951f5_level0_col12\" class=\"col_heading level0 col12\" >12</th>\n",
       "      <th id=\"T_951f5_level0_col13\" class=\"col_heading level0 col13\" >13</th>\n",
       "      <th id=\"T_951f5_level0_col14\" class=\"col_heading level0 col14\" >14</th>\n",
       "      <th id=\"T_951f5_level0_col15\" class=\"col_heading level0 col15\" >15</th>\n",
       "      <th id=\"T_951f5_level0_col16\" class=\"col_heading level0 col16\" >16</th>\n",
       "      <th id=\"T_951f5_level0_col17\" class=\"col_heading level0 col17\" >17</th>\n",
       "      <th id=\"T_951f5_level0_col18\" class=\"col_heading level0 col18\" >18</th>\n",
       "      <th id=\"T_951f5_level0_col19\" class=\"col_heading level0 col19\" >19</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_951f5_row0_col0\" class=\"data row0 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col1\" class=\"data row0 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col2\" class=\"data row0 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col3\" class=\"data row0 col3\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col4\" class=\"data row0 col4\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col5\" class=\"data row0 col5\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col6\" class=\"data row0 col6\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col7\" class=\"data row0 col7\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col8\" class=\"data row0 col8\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col9\" class=\"data row0 col9\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col10\" class=\"data row0 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col11\" class=\"data row0 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col12\" class=\"data row0 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col13\" class=\"data row0 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col14\" class=\"data row0 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col15\" class=\"data row0 col15\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col16\" class=\"data row0 col16\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col17\" class=\"data row0 col17\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col18\" class=\"data row0 col18\" >0</td>\n",
       "      <td id=\"T_951f5_row0_col19\" class=\"data row0 col19\" >0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_951f5_row1_col0\" class=\"data row1 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col1\" class=\"data row1 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col2\" class=\"data row1 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col3\" class=\"data row1 col3\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col4\" class=\"data row1 col4\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col5\" class=\"data row1 col5\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col6\" class=\"data row1 col6\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col7\" class=\"data row1 col7\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col8\" class=\"data row1 col8\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col9\" class=\"data row1 col9\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col10\" class=\"data row1 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col11\" class=\"data row1 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col12\" class=\"data row1 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col13\" class=\"data row1 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col14\" class=\"data row1 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col15\" class=\"data row1 col15\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col16\" class=\"data row1 col16\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col17\" class=\"data row1 col17\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col18\" class=\"data row1 col18\" >0</td>\n",
       "      <td id=\"T_951f5_row1_col19\" class=\"data row1 col19\" >0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_951f5_row2_col0\" class=\"data row2 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col1\" class=\"data row2 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col2\" class=\"data row2 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col3\" class=\"data row2 col3\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col4\" class=\"data row2 col4\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col5\" class=\"data row2 col5\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col6\" class=\"data row2 col6\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col7\" class=\"data row2 col7\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col8\" class=\"data row2 col8\" >18</td>\n",
       "      <td id=\"T_951f5_row2_col9\" class=\"data row2 col9\" >30</td>\n",
       "      <td id=\"T_951f5_row2_col10\" class=\"data row2 col10\" >137</td>\n",
       "      <td id=\"T_951f5_row2_col11\" class=\"data row2 col11\" >137</td>\n",
       "      <td id=\"T_951f5_row2_col12\" class=\"data row2 col12\" >192</td>\n",
       "      <td id=\"T_951f5_row2_col13\" class=\"data row2 col13\" >86</td>\n",
       "      <td id=\"T_951f5_row2_col14\" class=\"data row2 col14\" >72</td>\n",
       "      <td id=\"T_951f5_row2_col15\" class=\"data row2 col15\" >1</td>\n",
       "      <td id=\"T_951f5_row2_col16\" class=\"data row2 col16\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col17\" class=\"data row2 col17\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col18\" class=\"data row2 col18\" >0</td>\n",
       "      <td id=\"T_951f5_row2_col19\" class=\"data row2 col19\" >0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_951f5_row3_col0\" class=\"data row3 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row3_col1\" class=\"data row3 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row3_col2\" class=\"data row3 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row3_col3\" class=\"data row3 col3\" >0</td>\n",
       "      <td id=\"T_951f5_row3_col4\" class=\"data row3 col4\" >0</td>\n",
       "      <td id=\"T_951f5_row3_col5\" class=\"data row3 col5\" >0</td>\n",
       "      <td id=\"T_951f5_row3_col6\" class=\"data row3 col6\" >13</td>\n",
       "      <td id=\"T_951f5_row3_col7\" class=\"data row3 col7\" >86</td>\n",
       "      <td id=\"T_951f5_row3_col8\" class=\"data row3 col8\" >250</td>\n",
       "      <td id=\"T_951f5_row3_col9\" class=\"data row3 col9\" >254</td>\n",
       "      <td id=\"T_951f5_row3_col10\" class=\"data row3 col10\" >254</td>\n",
       "      <td id=\"T_951f5_row3_col11\" class=\"data row3 col11\" >254</td>\n",
       "      <td id=\"T_951f5_row3_col12\" class=\"data row3 col12\" >254</td>\n",
       "      <td id=\"T_951f5_row3_col13\" class=\"data row3 col13\" >217</td>\n",
       "      <td id=\"T_951f5_row3_col14\" class=\"data row3 col14\" >246</td>\n",
       "      <td id=\"T_951f5_row3_col15\" class=\"data row3 col15\" >151</td>\n",
       "      <td id=\"T_951f5_row3_col16\" class=\"data row3 col16\" >32</td>\n",
       "      <td id=\"T_951f5_row3_col17\" class=\"data row3 col17\" >0</td>\n",
       "      <td id=\"T_951f5_row3_col18\" class=\"data row3 col18\" >0</td>\n",
       "      <td id=\"T_951f5_row3_col19\" class=\"data row3 col19\" >0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_951f5_row4_col0\" class=\"data row4 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row4_col1\" class=\"data row4 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row4_col2\" class=\"data row4 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row4_col3\" class=\"data row4 col3\" >0</td>\n",
       "      <td id=\"T_951f5_row4_col4\" class=\"data row4 col4\" >0</td>\n",
       "      <td id=\"T_951f5_row4_col5\" class=\"data row4 col5\" >16</td>\n",
       "      <td id=\"T_951f5_row4_col6\" class=\"data row4 col6\" >179</td>\n",
       "      <td id=\"T_951f5_row4_col7\" class=\"data row4 col7\" >254</td>\n",
       "      <td id=\"T_951f5_row4_col8\" class=\"data row4 col8\" >254</td>\n",
       "      <td id=\"T_951f5_row4_col9\" class=\"data row4 col9\" >254</td>\n",
       "      <td id=\"T_951f5_row4_col10\" class=\"data row4 col10\" >254</td>\n",
       "      <td id=\"T_951f5_row4_col11\" class=\"data row4 col11\" >254</td>\n",
       "      <td id=\"T_951f5_row4_col12\" class=\"data row4 col12\" >254</td>\n",
       "      <td id=\"T_951f5_row4_col13\" class=\"data row4 col13\" >254</td>\n",
       "      <td id=\"T_951f5_row4_col14\" class=\"data row4 col14\" >254</td>\n",
       "      <td id=\"T_951f5_row4_col15\" class=\"data row4 col15\" >254</td>\n",
       "      <td id=\"T_951f5_row4_col16\" class=\"data row4 col16\" >231</td>\n",
       "      <td id=\"T_951f5_row4_col17\" class=\"data row4 col17\" >54</td>\n",
       "      <td id=\"T_951f5_row4_col18\" class=\"data row4 col18\" >15</td>\n",
       "      <td id=\"T_951f5_row4_col19\" class=\"data row4 col19\" >0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_951f5_row5_col0\" class=\"data row5 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row5_col1\" class=\"data row5 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row5_col2\" class=\"data row5 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row5_col3\" class=\"data row5 col3\" >0</td>\n",
       "      <td id=\"T_951f5_row5_col4\" class=\"data row5 col4\" >0</td>\n",
       "      <td id=\"T_951f5_row5_col5\" class=\"data row5 col5\" >72</td>\n",
       "      <td id=\"T_951f5_row5_col6\" class=\"data row5 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col7\" class=\"data row5 col7\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col8\" class=\"data row5 col8\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col9\" class=\"data row5 col9\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col10\" class=\"data row5 col10\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col11\" class=\"data row5 col11\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col12\" class=\"data row5 col12\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col13\" class=\"data row5 col13\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col14\" class=\"data row5 col14\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col15\" class=\"data row5 col15\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col16\" class=\"data row5 col16\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col17\" class=\"data row5 col17\" >254</td>\n",
       "      <td id=\"T_951f5_row5_col18\" class=\"data row5 col18\" >104</td>\n",
       "      <td id=\"T_951f5_row5_col19\" class=\"data row5 col19\" >0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_951f5_row6_col0\" class=\"data row6 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row6_col1\" class=\"data row6 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row6_col2\" class=\"data row6 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row6_col3\" class=\"data row6 col3\" >0</td>\n",
       "      <td id=\"T_951f5_row6_col4\" class=\"data row6 col4\" >61</td>\n",
       "      <td id=\"T_951f5_row6_col5\" class=\"data row6 col5\" >191</td>\n",
       "      <td id=\"T_951f5_row6_col6\" class=\"data row6 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row6_col7\" class=\"data row6 col7\" >254</td>\n",
       "      <td id=\"T_951f5_row6_col8\" class=\"data row6 col8\" >254</td>\n",
       "      <td id=\"T_951f5_row6_col9\" class=\"data row6 col9\" >254</td>\n",
       "      <td id=\"T_951f5_row6_col10\" class=\"data row6 col10\" >254</td>\n",
       "      <td id=\"T_951f5_row6_col11\" class=\"data row6 col11\" >109</td>\n",
       "      <td id=\"T_951f5_row6_col12\" class=\"data row6 col12\" >83</td>\n",
       "      <td id=\"T_951f5_row6_col13\" class=\"data row6 col13\" >199</td>\n",
       "      <td id=\"T_951f5_row6_col14\" class=\"data row6 col14\" >254</td>\n",
       "      <td id=\"T_951f5_row6_col15\" class=\"data row6 col15\" >254</td>\n",
       "      <td id=\"T_951f5_row6_col16\" class=\"data row6 col16\" >254</td>\n",
       "      <td id=\"T_951f5_row6_col17\" class=\"data row6 col17\" >254</td>\n",
       "      <td id=\"T_951f5_row6_col18\" class=\"data row6 col18\" >243</td>\n",
       "      <td id=\"T_951f5_row6_col19\" class=\"data row6 col19\" >85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_951f5_row7_col0\" class=\"data row7 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row7_col1\" class=\"data row7 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row7_col2\" class=\"data row7 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row7_col3\" class=\"data row7 col3\" >0</td>\n",
       "      <td id=\"T_951f5_row7_col4\" class=\"data row7 col4\" >172</td>\n",
       "      <td id=\"T_951f5_row7_col5\" class=\"data row7 col5\" >254</td>\n",
       "      <td id=\"T_951f5_row7_col6\" class=\"data row7 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row7_col7\" class=\"data row7 col7\" >254</td>\n",
       "      <td id=\"T_951f5_row7_col8\" class=\"data row7 col8\" >202</td>\n",
       "      <td id=\"T_951f5_row7_col9\" class=\"data row7 col9\" >147</td>\n",
       "      <td id=\"T_951f5_row7_col10\" class=\"data row7 col10\" >147</td>\n",
       "      <td id=\"T_951f5_row7_col11\" class=\"data row7 col11\" >45</td>\n",
       "      <td id=\"T_951f5_row7_col12\" class=\"data row7 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row7_col13\" class=\"data row7 col13\" >11</td>\n",
       "      <td id=\"T_951f5_row7_col14\" class=\"data row7 col14\" >29</td>\n",
       "      <td id=\"T_951f5_row7_col15\" class=\"data row7 col15\" >200</td>\n",
       "      <td id=\"T_951f5_row7_col16\" class=\"data row7 col16\" >254</td>\n",
       "      <td id=\"T_951f5_row7_col17\" class=\"data row7 col17\" >254</td>\n",
       "      <td id=\"T_951f5_row7_col18\" class=\"data row7 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row7_col19\" class=\"data row7 col19\" >171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_951f5_row8_col0\" class=\"data row8 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row8_col1\" class=\"data row8 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row8_col2\" class=\"data row8 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row8_col3\" class=\"data row8 col3\" >1</td>\n",
       "      <td id=\"T_951f5_row8_col4\" class=\"data row8 col4\" >174</td>\n",
       "      <td id=\"T_951f5_row8_col5\" class=\"data row8 col5\" >254</td>\n",
       "      <td id=\"T_951f5_row8_col6\" class=\"data row8 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row8_col7\" class=\"data row8 col7\" >89</td>\n",
       "      <td id=\"T_951f5_row8_col8\" class=\"data row8 col8\" >67</td>\n",
       "      <td id=\"T_951f5_row8_col9\" class=\"data row8 col9\" >0</td>\n",
       "      <td id=\"T_951f5_row8_col10\" class=\"data row8 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row8_col11\" class=\"data row8 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row8_col12\" class=\"data row8 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row8_col13\" class=\"data row8 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row8_col14\" class=\"data row8 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row8_col15\" class=\"data row8 col15\" >128</td>\n",
       "      <td id=\"T_951f5_row8_col16\" class=\"data row8 col16\" >252</td>\n",
       "      <td id=\"T_951f5_row8_col17\" class=\"data row8 col17\" >254</td>\n",
       "      <td id=\"T_951f5_row8_col18\" class=\"data row8 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row8_col19\" class=\"data row8 col19\" >212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_951f5_row9_col0\" class=\"data row9 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col1\" class=\"data row9 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col2\" class=\"data row9 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col3\" class=\"data row9 col3\" >47</td>\n",
       "      <td id=\"T_951f5_row9_col4\" class=\"data row9 col4\" >254</td>\n",
       "      <td id=\"T_951f5_row9_col5\" class=\"data row9 col5\" >254</td>\n",
       "      <td id=\"T_951f5_row9_col6\" class=\"data row9 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row9_col7\" class=\"data row9 col7\" >29</td>\n",
       "      <td id=\"T_951f5_row9_col8\" class=\"data row9 col8\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col9\" class=\"data row9 col9\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col10\" class=\"data row9 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col11\" class=\"data row9 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col12\" class=\"data row9 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col13\" class=\"data row9 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col14\" class=\"data row9 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col15\" class=\"data row9 col15\" >0</td>\n",
       "      <td id=\"T_951f5_row9_col16\" class=\"data row9 col16\" >83</td>\n",
       "      <td id=\"T_951f5_row9_col17\" class=\"data row9 col17\" >254</td>\n",
       "      <td id=\"T_951f5_row9_col18\" class=\"data row9 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row9_col19\" class=\"data row9 col19\" >254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "      <td id=\"T_951f5_row10_col0\" class=\"data row10 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col1\" class=\"data row10 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col2\" class=\"data row10 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col3\" class=\"data row10 col3\" >80</td>\n",
       "      <td id=\"T_951f5_row10_col4\" class=\"data row10 col4\" >254</td>\n",
       "      <td id=\"T_951f5_row10_col5\" class=\"data row10 col5\" >254</td>\n",
       "      <td id=\"T_951f5_row10_col6\" class=\"data row10 col6\" >240</td>\n",
       "      <td id=\"T_951f5_row10_col7\" class=\"data row10 col7\" >24</td>\n",
       "      <td id=\"T_951f5_row10_col8\" class=\"data row10 col8\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col9\" class=\"data row10 col9\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col10\" class=\"data row10 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col11\" class=\"data row10 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col12\" class=\"data row10 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col13\" class=\"data row10 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col14\" class=\"data row10 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col15\" class=\"data row10 col15\" >0</td>\n",
       "      <td id=\"T_951f5_row10_col16\" class=\"data row10 col16\" >25</td>\n",
       "      <td id=\"T_951f5_row10_col17\" class=\"data row10 col17\" >240</td>\n",
       "      <td id=\"T_951f5_row10_col18\" class=\"data row10 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row10_col19\" class=\"data row10 col19\" >254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "      <td id=\"T_951f5_row11_col0\" class=\"data row11 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col1\" class=\"data row11 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col2\" class=\"data row11 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col3\" class=\"data row11 col3\" >64</td>\n",
       "      <td id=\"T_951f5_row11_col4\" class=\"data row11 col4\" >254</td>\n",
       "      <td id=\"T_951f5_row11_col5\" class=\"data row11 col5\" >254</td>\n",
       "      <td id=\"T_951f5_row11_col6\" class=\"data row11 col6\" >186</td>\n",
       "      <td id=\"T_951f5_row11_col7\" class=\"data row11 col7\" >7</td>\n",
       "      <td id=\"T_951f5_row11_col8\" class=\"data row11 col8\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col9\" class=\"data row11 col9\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col10\" class=\"data row11 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col11\" class=\"data row11 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col12\" class=\"data row11 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col13\" class=\"data row11 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col14\" class=\"data row11 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col15\" class=\"data row11 col15\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col16\" class=\"data row11 col16\" >0</td>\n",
       "      <td id=\"T_951f5_row11_col17\" class=\"data row11 col17\" >166</td>\n",
       "      <td id=\"T_951f5_row11_col18\" class=\"data row11 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row11_col19\" class=\"data row11 col19\" >254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "      <td id=\"T_951f5_row12_col0\" class=\"data row12 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col1\" class=\"data row12 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col2\" class=\"data row12 col2\" >14</td>\n",
       "      <td id=\"T_951f5_row12_col3\" class=\"data row12 col3\" >232</td>\n",
       "      <td id=\"T_951f5_row12_col4\" class=\"data row12 col4\" >254</td>\n",
       "      <td id=\"T_951f5_row12_col5\" class=\"data row12 col5\" >254</td>\n",
       "      <td id=\"T_951f5_row12_col6\" class=\"data row12 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row12_col7\" class=\"data row12 col7\" >29</td>\n",
       "      <td id=\"T_951f5_row12_col8\" class=\"data row12 col8\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col9\" class=\"data row12 col9\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col10\" class=\"data row12 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col11\" class=\"data row12 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col12\" class=\"data row12 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col13\" class=\"data row12 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col14\" class=\"data row12 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col15\" class=\"data row12 col15\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col16\" class=\"data row12 col16\" >0</td>\n",
       "      <td id=\"T_951f5_row12_col17\" class=\"data row12 col17\" >75</td>\n",
       "      <td id=\"T_951f5_row12_col18\" class=\"data row12 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row12_col19\" class=\"data row12 col19\" >254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "      <td id=\"T_951f5_row13_col0\" class=\"data row13 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col1\" class=\"data row13 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col2\" class=\"data row13 col2\" >18</td>\n",
       "      <td id=\"T_951f5_row13_col3\" class=\"data row13 col3\" >254</td>\n",
       "      <td id=\"T_951f5_row13_col4\" class=\"data row13 col4\" >254</td>\n",
       "      <td id=\"T_951f5_row13_col5\" class=\"data row13 col5\" >254</td>\n",
       "      <td id=\"T_951f5_row13_col6\" class=\"data row13 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row13_col7\" class=\"data row13 col7\" >29</td>\n",
       "      <td id=\"T_951f5_row13_col8\" class=\"data row13 col8\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col9\" class=\"data row13 col9\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col10\" class=\"data row13 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col11\" class=\"data row13 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col12\" class=\"data row13 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col13\" class=\"data row13 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col14\" class=\"data row13 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col15\" class=\"data row13 col15\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col16\" class=\"data row13 col16\" >0</td>\n",
       "      <td id=\"T_951f5_row13_col17\" class=\"data row13 col17\" >48</td>\n",
       "      <td id=\"T_951f5_row13_col18\" class=\"data row13 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row13_col19\" class=\"data row13 col19\" >254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "      <td id=\"T_951f5_row14_col0\" class=\"data row14 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col1\" class=\"data row14 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col2\" class=\"data row14 col2\" >2</td>\n",
       "      <td id=\"T_951f5_row14_col3\" class=\"data row14 col3\" >163</td>\n",
       "      <td id=\"T_951f5_row14_col4\" class=\"data row14 col4\" >254</td>\n",
       "      <td id=\"T_951f5_row14_col5\" class=\"data row14 col5\" >254</td>\n",
       "      <td id=\"T_951f5_row14_col6\" class=\"data row14 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row14_col7\" class=\"data row14 col7\" >29</td>\n",
       "      <td id=\"T_951f5_row14_col8\" class=\"data row14 col8\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col9\" class=\"data row14 col9\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col10\" class=\"data row14 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col11\" class=\"data row14 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col12\" class=\"data row14 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col13\" class=\"data row14 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col14\" class=\"data row14 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col15\" class=\"data row14 col15\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col16\" class=\"data row14 col16\" >0</td>\n",
       "      <td id=\"T_951f5_row14_col17\" class=\"data row14 col17\" >48</td>\n",
       "      <td id=\"T_951f5_row14_col18\" class=\"data row14 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row14_col19\" class=\"data row14 col19\" >254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "      <td id=\"T_951f5_row15_col0\" class=\"data row15 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row15_col1\" class=\"data row15 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row15_col2\" class=\"data row15 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row15_col3\" class=\"data row15 col3\" >94</td>\n",
       "      <td id=\"T_951f5_row15_col4\" class=\"data row15 col4\" >254</td>\n",
       "      <td id=\"T_951f5_row15_col5\" class=\"data row15 col5\" >254</td>\n",
       "      <td id=\"T_951f5_row15_col6\" class=\"data row15 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row15_col7\" class=\"data row15 col7\" >200</td>\n",
       "      <td id=\"T_951f5_row15_col8\" class=\"data row15 col8\" >12</td>\n",
       "      <td id=\"T_951f5_row15_col9\" class=\"data row15 col9\" >0</td>\n",
       "      <td id=\"T_951f5_row15_col10\" class=\"data row15 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row15_col11\" class=\"data row15 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row15_col12\" class=\"data row15 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row15_col13\" class=\"data row15 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row15_col14\" class=\"data row15 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row15_col15\" class=\"data row15 col15\" >0</td>\n",
       "      <td id=\"T_951f5_row15_col16\" class=\"data row15 col16\" >16</td>\n",
       "      <td id=\"T_951f5_row15_col17\" class=\"data row15 col17\" >209</td>\n",
       "      <td id=\"T_951f5_row15_col18\" class=\"data row15 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row15_col19\" class=\"data row15 col19\" >254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row16\" class=\"row_heading level0 row16\" >16</th>\n",
       "      <td id=\"T_951f5_row16_col0\" class=\"data row16 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row16_col1\" class=\"data row16 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row16_col2\" class=\"data row16 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row16_col3\" class=\"data row16 col3\" >15</td>\n",
       "      <td id=\"T_951f5_row16_col4\" class=\"data row16 col4\" >206</td>\n",
       "      <td id=\"T_951f5_row16_col5\" class=\"data row16 col5\" >254</td>\n",
       "      <td id=\"T_951f5_row16_col6\" class=\"data row16 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row16_col7\" class=\"data row16 col7\" >254</td>\n",
       "      <td id=\"T_951f5_row16_col8\" class=\"data row16 col8\" >202</td>\n",
       "      <td id=\"T_951f5_row16_col9\" class=\"data row16 col9\" >66</td>\n",
       "      <td id=\"T_951f5_row16_col10\" class=\"data row16 col10\" >0</td>\n",
       "      <td id=\"T_951f5_row16_col11\" class=\"data row16 col11\" >0</td>\n",
       "      <td id=\"T_951f5_row16_col12\" class=\"data row16 col12\" >0</td>\n",
       "      <td id=\"T_951f5_row16_col13\" class=\"data row16 col13\" >0</td>\n",
       "      <td id=\"T_951f5_row16_col14\" class=\"data row16 col14\" >0</td>\n",
       "      <td id=\"T_951f5_row16_col15\" class=\"data row16 col15\" >21</td>\n",
       "      <td id=\"T_951f5_row16_col16\" class=\"data row16 col16\" >161</td>\n",
       "      <td id=\"T_951f5_row16_col17\" class=\"data row16 col17\" >254</td>\n",
       "      <td id=\"T_951f5_row16_col18\" class=\"data row16 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row16_col19\" class=\"data row16 col19\" >245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_951f5_level0_row17\" class=\"row_heading level0 row17\" >17</th>\n",
       "      <td id=\"T_951f5_row17_col0\" class=\"data row17 col0\" >0</td>\n",
       "      <td id=\"T_951f5_row17_col1\" class=\"data row17 col1\" >0</td>\n",
       "      <td id=\"T_951f5_row17_col2\" class=\"data row17 col2\" >0</td>\n",
       "      <td id=\"T_951f5_row17_col3\" class=\"data row17 col3\" >0</td>\n",
       "      <td id=\"T_951f5_row17_col4\" class=\"data row17 col4\" >60</td>\n",
       "      <td id=\"T_951f5_row17_col5\" class=\"data row17 col5\" >212</td>\n",
       "      <td id=\"T_951f5_row17_col6\" class=\"data row17 col6\" >254</td>\n",
       "      <td id=\"T_951f5_row17_col7\" class=\"data row17 col7\" >254</td>\n",
       "      <td id=\"T_951f5_row17_col8\" class=\"data row17 col8\" >254</td>\n",
       "      <td id=\"T_951f5_row17_col9\" class=\"data row17 col9\" >194</td>\n",
       "      <td id=\"T_951f5_row17_col10\" class=\"data row17 col10\" >48</td>\n",
       "      <td id=\"T_951f5_row17_col11\" class=\"data row17 col11\" >48</td>\n",
       "      <td id=\"T_951f5_row17_col12\" class=\"data row17 col12\" >34</td>\n",
       "      <td id=\"T_951f5_row17_col13\" class=\"data row17 col13\" >41</td>\n",
       "      <td id=\"T_951f5_row17_col14\" class=\"data row17 col14\" >48</td>\n",
       "      <td id=\"T_951f5_row17_col15\" class=\"data row17 col15\" >209</td>\n",
       "      <td id=\"T_951f5_row17_col16\" class=\"data row17 col16\" >254</td>\n",
       "      <td id=\"T_951f5_row17_col17\" class=\"data row17 col17\" >254</td>\n",
       "      <td id=\"T_951f5_row17_col18\" class=\"data row17 col18\" >254</td>\n",
       "      <td id=\"T_951f5_row17_col19\" class=\"data row17 col19\" >171</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f748855a2c0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame(im2[2:20,2:22])\n",
    "df.style.set_properties(**{'font-size':'6pt'}).background_gradient('Greys')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62fa11b3-4de8-4de1-8adb-e00645a9addf",
   "metadata": {},
   "source": [
    "Going back to the raw data, this [post](https://stackoverflow.com/questions/49330639/reading-bytes-for-an-image) nicely explains, this is how we can visualize the image:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "a373bbae-7a66-458d-9db7-87f0d74d45d1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAABwAAAAcCAAAAABXZoBIAAABEUlEQVR4nL2RvUoDURCFj2IhJDFR0S7401mIW2mjjYIGTBdcsLEWxMYn8BkSiFhaWCk2QsBKsNCwhRBIkweQaJWwixYq31osN7m7C5Y5zZ25350zwx1p9BobhjPZE23U/Ycw/SrnfgN0vi4WU6xwj1F3PZ+AJSwdx9nWIwCnlReA4CAGbwGvVltVptgEbmIz38HhThSf/0K7bME1oGiSCnApSRqXJC1J/o+Bz76JItiXvJ656jakvezAderdttU+MD2onJiPjf4Ws+1f6x8dAU3TptCCurWQ/CvgbUuS5q7gc8Eu3WwDT46zvOK0kj8kuQEAH9HhJtqeDZfSczMJmGsYFuymJ54sVwkJq7PJVY9Wf10bnse4xJV1AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<PIL.Image.Image image mode=L size=28x28>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "image = Image.frombytes('L', (28, 28), bytes(im))\n",
    "image"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "951f380a-aa83-4f35-b501-7cd3afbe30b8",
   "metadata": {},
   "source": [
    "## Converting the csv to images\n",
    "\n",
    "Now that we can read the csv and create an image from a line, we can simply convert the csv to image files and afterwards use the same technique as in the first notebook to classify the images (we know how to do that)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6029aa5c-ebbd-432c-a143-78f7ddf9a976",
   "metadata": {},
   "source": [
    "### Converting the training data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5c200f6-ae6b-4b79-b5a4-451577171eea",
   "metadata": {},
   "source": [
    "Let's create the same path structure as in Fast.AI, i.e. collect all images of a digit in one folder:\n",
    "\n",
    "(Again, the vital calls are commented, not to accidentally create too much disk activity)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b74edb71-5900-4af6-bedb-1e850a887e51",
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_paths():\n",
    "    path = Path()\n",
    "    path = (path/'training')\n",
    "\n",
    "    # removing directory\n",
    "    shutil.rmtree(path)\n",
    "\n",
    "    # creating directory\n",
    "    os.mkdir(path)\n",
    "\n",
    "    for i in range(10):\n",
    "        os.mkdir(path/str(i))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "7802341b-5ea5-474a-b042-fe229932028e",
   "metadata": {},
   "outputs": [],
   "source": [
    "#prepare_paths()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ea2e68c-7b8c-4f16-b7ac-d447211f50dd",
   "metadata": {},
   "source": [
    "Let's try one image:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "82723f49-cb5e-4a3c-9764-f04183ba95c8",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_image(index):\n",
    "    number = train.iloc[index,0]\n",
    "    im = torch.tensor(train.iloc[index,1:])\n",
    "    image = Image.frombytes('L', (28, 28), bytes(im))\n",
    "    return number, image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "f08a0443-083f-4935-ba2a-f6bb07cdb4dc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is a 0\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAABwAAAAcCAAAAABXZoBIAAABBUlEQVR4nMXQMUoDQRiG4W+jQUGL2CgJJKJErAQvoY2oYJmgjUiEFILkBFaCHkAiglp6BysbK9Gg2SRFkIAEcgCzhe9ikc3sbnbT6tfM/PPMMP+M9J+x0mfXruve5BJRmy7h5TSiM6+YlEdx0Tfso4mQLdQA567d/gZgNYTnwNe2pM0WgF0MWLIOPEmSjj8B7KyPFaC/NZhnngEakwZdc9BoMoSHpkp3gbykmO9Q15G0PwZ1642xOCupOebOHSfSUGPZK5augLJl8AXgUpKUu+gB1YQkDTakHtcl6lXpYCUl6X2jF2hhj2De5kP9WUWfPgpTw9XhOHeyuybpvtN8+Il73l/lF0mJqOtQ2SOAAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<PIL.Image.Image image mode=L size=28x28>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "number, image = get_image(4)\n",
    "print(f'This is a {number}')\n",
    "image"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b146849e-b239-4ad2-888d-f041b31793ed",
   "metadata": {},
   "source": [
    "Let's save the image:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "cf9099ad-d26b-49da-927c-2c6d18137bea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Path('training')"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = Path()\n",
    "path = (path/'training')\n",
    "path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "48d77644-6f36-40b0-998c-cc1fbccac5a9",
   "metadata": {},
   "outputs": [],
   "source": [
    "image.save(path/'test.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc22dd79-f143-4c66-87b3-5e1c36fd205c",
   "metadata": {},
   "source": [
    "All successful, let's do it for real:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "a6a5f031-905f-4e3a-8336-2b165e8dbf1a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def convert_images():\n",
    "    for i in range(len(train)):\n",
    "        number, image = get_image(i)\n",
    "        image.save(path/str(number)/f'{i}.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "315b5110-5754-4d70-8799-5b0063029dfb",
   "metadata": {},
   "outputs": [],
   "source": [
    "#convert_images()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "70665afb-b442-4e8e-96ae-fa81e31b8e7e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Looking in :  training\n",
      "Looking in :  training/0\n",
      "Looking in :  training/1\n",
      "Looking in :  training/2\n",
      "Looking in :  training/3\n",
      "Looking in :  training/4\n",
      "Looking in :  training/5\n",
      "Looking in :  training/6\n",
      "Looking in :  training/7\n",
      "Looking in :  training/8\n",
      "Looking in :  training/9\n",
      "Number of files 42002\n",
      "Number of Directories 10\n",
      "Total: 42012\n"
     ]
    }
   ],
   "source": [
    "#https://codezup.com/count-files-directories-recursively-python/\n",
    "\n",
    "noOfFiles = 0\n",
    "noOfDir = 0\n",
    "\n",
    "for base, dirs, files in os.walk(path):\n",
    "    print('Looking in : ',base)\n",
    "    for directories in dirs:\n",
    "        noOfDir += 1\n",
    "    for Files in files:\n",
    "        noOfFiles += 1\n",
    "        \n",
    "print('Number of files',noOfFiles)\n",
    "print('Number of Directories',noOfDir)\n",
    "print('Total:',(noOfDir + noOfFiles))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ac679a5-fe18-4105-9a22-b3e79016735c",
   "metadata": {},
   "source": [
    "That looks right: 42000 + 1 dummy."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a3a1bfc-fb60-404d-beef-2967f872543c",
   "metadata": {},
   "source": [
    "### Converting the test data\n",
    "\n",
    "The structure is slightly different, because there is no lable column:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "252c76a2-74e2-4c70-b699-e8a36a8e79f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pixel0</th>\n",
       "      <th>pixel1</th>\n",
       "      <th>pixel2</th>\n",
       "      <th>pixel3</th>\n",
       "      <th>pixel4</th>\n",
       "      <th>pixel5</th>\n",
       "      <th>pixel6</th>\n",
       "      <th>pixel7</th>\n",
       "      <th>pixel8</th>\n",
       "      <th>pixel9</th>\n",
       "      <th>...</th>\n",
       "      <th>pixel774</th>\n",
       "      <th>pixel775</th>\n",
       "      <th>pixel776</th>\n",
       "      <th>pixel777</th>\n",
       "      <th>pixel778</th>\n",
       "      <th>pixel779</th>\n",
       "      <th>pixel780</th>\n",
       "      <th>pixel781</th>\n",
       "      <th>pixel782</th>\n",
       "      <th>pixel783</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 784 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   pixel0  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  pixel8  \\\n",
       "0       0       0       0       0       0       0       0       0       0   \n",
       "1       0       0       0       0       0       0       0       0       0   \n",
       "2       0       0       0       0       0       0       0       0       0   \n",
       "3       0       0       0       0       0       0       0       0       0   \n",
       "4       0       0       0       0       0       0       0       0       0   \n",
       "\n",
       "   pixel9  ...  pixel774  pixel775  pixel776  pixel777  pixel778  pixel779  \\\n",
       "0       0  ...         0         0         0         0         0         0   \n",
       "1       0  ...         0         0         0         0         0         0   \n",
       "2       0  ...         0         0         0         0         0         0   \n",
       "3       0  ...         0         0         0         0         0         0   \n",
       "4       0  ...         0         0         0         0         0         0   \n",
       "\n",
       "   pixel780  pixel781  pixel782  pixel783  \n",
       "0         0         0         0         0  \n",
       "1         0         0         0         0  \n",
       "2         0         0         0         0  \n",
       "3         0         0         0         0  \n",
       "4         0         0         0         0  \n",
       "\n",
       "[5 rows x 784 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test = pd.read_csv(\"test.csv\")\n",
    "test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "46e628b2-a853-4166-994f-3bcbae7ae084",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "28000"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c1c3473-655d-42c2-9c58-43b3b78d1e78",
   "metadata": {},
   "source": [
    "Again, let's have a look at one image:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "63781eae-4335-4c4f-a237-6767775768e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_test_image(index):\n",
    "    im = torch.tensor(test.iloc[index])\n",
    "    image = Image.frombytes('L', (28, 28), bytes(im))\n",
    "    return image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "043f130d-9d33-41d9-8380-e395a59aabbe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAABwAAAAcCAAAAABXZoBIAAABBklEQVR4nGNgoD9gRDC52BkYGBySGfLvYlHX/RcCDLDI2TyBSp4/qochefUvHNw3QZd0ev73798cDY3iL3//tjGjO8jemIFh820GhrMGDAxCH3F4wPLv37/8DAwMDAxMmJIITVgkTXGYyMDAwMBwFm4sC7KwrdrfBQy6wgwMx34ju5abL+BVFoO65L9DDLLKDFc9nyJ0aJWs/YsM7pawIyTL/v79+/frlXsI6QX8cMl/f//+3ZvMoHDq79+/H6qq9v39+3cdXPL/379/3927d+/j37/P3BgYBNc//fsXLjkPZtrZLEcGBgYGBrsvc+GuZROZycDAwJD98fc3qHK+H7/wBcWgBAC12Y+lQlhU6gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<PIL.Image.Image image mode=L size=28x28>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_image = get_test_image(0)\n",
    "test_image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "258adc99-ad1d-4cf8-854f-0bafb8504652",
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_test_path():\n",
    "    path = Path()\n",
    "    path = (path/'test')\n",
    "    #os.mkdir(path)\n",
    "\n",
    "    # removing directory\n",
    "    shutil.rmtree(path)\n",
    "\n",
    "    # creating directory\n",
    "    os.mkdir(path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "5596db58-d182-4ab5-8d82-152bb0c05949",
   "metadata": {},
   "outputs": [],
   "source": [
    "#prepare_test_path()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "a0c21202-cb7e-43d2-852a-94c15ff12ce4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Path('test')"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = Path()\n",
    "path = (path/'test')\n",
    "path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "a209565e-2961-486b-a153-e43668575e0c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def convert_test_data():\n",
    "    for i in range(len(test)):\n",
    "        test_image = get_test_image(i)\n",
    "        test_image.save(path/f'{i}.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "124a533c-98fe-41e6-8442-261e8423e848",
   "metadata": {},
   "outputs": [],
   "source": [
    "#convert_test_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6239d414-caca-4b53-aaeb-2b395cd5fbff",
   "metadata": {},
   "source": [
    "As a result, we have converted the `csv`-file to `png`-images. :)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}