MITBIH_BeatClassification_Ozal Like.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MITBIH_BeatClassification_OzalLike.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/M-torki/ECG-Classification/blob/main/MITBIH_BeatClassification_Ozal%20Like.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "d65f2bDCQ8R7",
        "outputId": "ae2183ab-7eef-4246-dbf6-907699bd40d9"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/gdrive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /gdrive; to attempt to forcibly remount, call drive.mount(\"/gdrive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GzMqLdqqRMYz",
        "outputId": "e55c33c0-85c1-4257-9278-ad1ca500c166"
      },
      "source": [
        "cd /gdrive/MyDrive/"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/gdrive/MyDrive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "adxoI3ObRMHy",
        "outputId": "9aafbbb1-4105-4117-adbc-6de99c18d795"
      },
      "source": [
        "cd mit-bih-arrhythmia-database-1.0.0/"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/gdrive/MyDrive/mit-bih-arrhythmia-database-1.0.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gyiEXw1FPrmb"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.metrics import classification_report\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import f1_score\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from keras.utils.np_utils import to_categorical\n",
        "from sklearn.utils import class_weight\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "from sklearn.utils import shuffle"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1wuOUP3MRXCw",
        "outputId": "85898dfc-c868-4e7f-9605-c4e9bea3f0c5"
      },
      "source": [
        "ls"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "100.atr    108.at_  117.at_  200.xws  210.dat  221.atr  234.atr\n",
            "100.dat    108.atr  117.atr  201.atr  210.hea  221.dat  234.dat\n",
            "100.hea    108.dat  117.dat  201.dat  210.xws  221.hea  234.hea\n",
            "100.xws    108.hea  117.hea  201.hea  212.atr  221.xws  234.xws\n",
            "101.atr    108.xws  117.xws  201.xws  212.dat  222.at_  ANNOTATORS\n",
            "101.dat    109.atr  118.atr  202.atr  212.hea  222.atr  a_train.txt\n",
            "101.hea    109.dat  118.dat  202.dat  212.xws  222.dat  A_train.txt\n",
            "101.xws    109.hea  118.hea  202.hea  213.atr  222.hea  best_model.h5\n",
            "102-0.atr  109.xws  118.xws  202.xws  213.dat  222.xws  best_modelozal.h5\n",
            "102.atr    111.atr  119.at_  203.at_  213.hea  223.atr  F_train.txt\n",
            "102.dat    111.dat  119.atr  203.at-  213.xws  223.dat  J_train.txt\n",
            "102.hea    111.hea  119.dat  203.atr  214.at_  223.hea  L_train.txt\n",
            "102.xws    111.xws  119.hea  203.dat  214.atr  223.xws  MIT_BIH_Record_100.csv\n",
            "103.atr    112.atr  119.xws  203.hea  214.dat  228.atr  MITBIHtest1.csv\n",
            "103.dat    112.dat  121.atr  203.xws  214.hea  228.dat  MITBIHtest2.csv\n",
            "103.hea    112.hea  121.dat  205.atr  214.xws  228.hea  MITBIHtest.csv\n",
            "103.xws    112.xws  121.hea  205.dat  215.at_  228.xws  MITBIHtrain1.csv\n",
            "104.atr    113.atr  121.xws  205.hea  215.atr  230.atr  MITBIHtrain2.csv\n",
            "104.dat    113.dat  122.atr  205.xws  215.dat  230.dat  MITBIHtrain.csv\n",
            "104.hea    113.hea  122.dat  207.atr  215.hea  230.hea  \u001b[0m\u001b[01;34mmitdbdir\u001b[0m/\n",
            "104.xws    113.xws  122.hea  207.dat  215.xws  230.xws  N_train.txt\n",
            "105.atr    114.atr  122.xws  207.hea  217.atr  231.atr  RECORDS\n",
            "105.dat    114.dat  123.atr  207.xws  217.dat  231.dat  R_train.txt\n",
            "105.hea    114.hea  123.dat  208.atr  217.hea  231.hea  sairandata.csv\n",
            "105.xws    114.xws  123.hea  208.dat  217.xws  231.xws  SHA256SUMS.txt\n",
            "106.atr    115.atr  123.xws  208.hea  219.atr  232.atr  slash_train.txt\n",
            "106.dat    115.dat  124.atr  208.xws  219.dat  232.dat  testdata.csv\n",
            "106.hea    115.hea  124.dat  209.at_  219.hea  232.hea  traindata.csv\n",
            "106.xws    115.xws  124.hea  209.atr  219.xws  232.xws  V_train.txt\n",
            "107.atr    116.atr  124.xws  209.dat  220.atr  233.atr  \u001b[01;34mx_mitdb\u001b[0m/\n",
            "107.dat    116.dat  200.atr  209.hea  220.dat  233.dat\n",
            "107.hea    116.hea  200.dat  209.xws  220.hea  233.hea\n",
            "107.xws    116.xws  200.hea  210.atr  220.xws  233.xws\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MQvVQFMiQ5hY"
      },
      "source": [
        "train_df=pd.read_csv('MITBIHtrain2.csv', index_col=0)\n",
        "train_df = shuffle(train_df)\n",
        "\n",
        "test_df=pd.read_csv('MITBIHtest2.csv', index_col=0)\n",
        "test_df = shuffle(test_df)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 444
        },
        "id": "3HjpmUFOSHyr",
        "outputId": "cc690624-04f0-4972-9acc-cec50684dd64"
      },
      "source": [
        "train_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "      <th>12</th>\n",
              "      <th>13</th>\n",
              "      <th>14</th>\n",
              "      <th>15</th>\n",
              "      <th>16</th>\n",
              "      <th>17</th>\n",
              "      <th>18</th>\n",
              "      <th>19</th>\n",
              "      <th>20</th>\n",
              "      <th>21</th>\n",
              "      <th>22</th>\n",
              "      <th>23</th>\n",
              "      <th>24</th>\n",
              "      <th>25</th>\n",
              "      <th>26</th>\n",
              "      <th>27</th>\n",
              "      <th>28</th>\n",
              "      <th>29</th>\n",
              "      <th>30</th>\n",
              "      <th>31</th>\n",
              "      <th>32</th>\n",
              "      <th>33</th>\n",
              "      <th>34</th>\n",
              "      <th>35</th>\n",
              "      <th>36</th>\n",
              "      <th>37</th>\n",
              "      <th>38</th>\n",
              "      <th>39</th>\n",
              "      <th>...</th>\n",
              "      <th>220</th>\n",
              "      <th>221</th>\n",
              "      <th>222</th>\n",
              "      <th>223</th>\n",
              "      <th>224</th>\n",
              "      <th>225</th>\n",
              "      <th>226</th>\n",
              "      <th>227</th>\n",
              "      <th>228</th>\n",
              "      <th>229</th>\n",
              "      <th>230</th>\n",
              "      <th>231</th>\n",
              "      <th>232</th>\n",
              "      <th>233</th>\n",
              "      <th>234</th>\n",
              "      <th>235</th>\n",
              "      <th>236</th>\n",
              "      <th>237</th>\n",
              "      <th>238</th>\n",
              "      <th>239</th>\n",
              "      <th>240</th>\n",
              "      <th>241</th>\n",
              "      <th>242</th>\n",
              "      <th>243</th>\n",
              "      <th>244</th>\n",
              "      <th>245</th>\n",
              "      <th>246</th>\n",
              "      <th>247</th>\n",
              "      <th>248</th>\n",
              "      <th>249</th>\n",
              "      <th>250</th>\n",
              "      <th>251</th>\n",
              "      <th>252</th>\n",
              "      <th>253</th>\n",
              "      <th>254</th>\n",
              "      <th>255</th>\n",
              "      <th>256</th>\n",
              "      <th>257</th>\n",
              "      <th>258</th>\n",
              "      <th>259</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>18337</th>\n",
              "      <td>0.082707</td>\n",
              "      <td>0.071429</td>\n",
              "      <td>0.060150</td>\n",
              "      <td>0.063910</td>\n",
              "      <td>0.067669</td>\n",
              "      <td>0.075188</td>\n",
              "      <td>0.078947</td>\n",
              "      <td>0.075188</td>\n",
              "      <td>0.075188</td>\n",
              "      <td>0.071429</td>\n",
              "      <td>0.075188</td>\n",
              "      <td>0.078947</td>\n",
              "      <td>0.078947</td>\n",
              "      <td>0.078947</td>\n",
              "      <td>0.071429</td>\n",
              "      <td>0.071429</td>\n",
              "      <td>0.075188</td>\n",
              "      <td>0.086466</td>\n",
              "      <td>0.090226</td>\n",
              "      <td>0.090226</td>\n",
              "      <td>0.090226</td>\n",
              "      <td>0.086466</td>\n",
              "      <td>0.086466</td>\n",
              "      <td>0.097744</td>\n",
              "      <td>0.101504</td>\n",
              "      <td>0.093985</td>\n",
              "      <td>0.090226</td>\n",
              "      <td>0.086466</td>\n",
              "      <td>0.093985</td>\n",
              "      <td>0.097744</td>\n",
              "      <td>0.093985</td>\n",
              "      <td>0.082707</td>\n",
              "      <td>0.086466</td>\n",
              "      <td>0.082707</td>\n",
              "      <td>0.093985</td>\n",
              "      <td>0.101504</td>\n",
              "      <td>0.097744</td>\n",
              "      <td>0.090226</td>\n",
              "      <td>0.075188</td>\n",
              "      <td>0.075188</td>\n",
              "      <td>...</td>\n",
              "      <td>0.101504</td>\n",
              "      <td>0.105263</td>\n",
              "      <td>0.109023</td>\n",
              "      <td>0.109023</td>\n",
              "      <td>0.112782</td>\n",
              "      <td>0.109023</td>\n",
              "      <td>0.124060</td>\n",
              "      <td>0.131579</td>\n",
              "      <td>0.139098</td>\n",
              "      <td>0.150376</td>\n",
              "      <td>0.154135</td>\n",
              "      <td>0.169173</td>\n",
              "      <td>0.191729</td>\n",
              "      <td>0.221805</td>\n",
              "      <td>0.266917</td>\n",
              "      <td>0.342105</td>\n",
              "      <td>0.417293</td>\n",
              "      <td>0.503759</td>\n",
              "      <td>0.590226</td>\n",
              "      <td>0.661654</td>\n",
              "      <td>0.755639</td>\n",
              "      <td>0.830827</td>\n",
              "      <td>0.879699</td>\n",
              "      <td>0.887218</td>\n",
              "      <td>0.860902</td>\n",
              "      <td>0.804511</td>\n",
              "      <td>0.706767</td>\n",
              "      <td>0.593985</td>\n",
              "      <td>0.451128</td>\n",
              "      <td>0.312030</td>\n",
              "      <td>0.233083</td>\n",
              "      <td>0.218045</td>\n",
              "      <td>0.214286</td>\n",
              "      <td>0.184211</td>\n",
              "      <td>0.146617</td>\n",
              "      <td>0.127820</td>\n",
              "      <td>0.127820</td>\n",
              "      <td>0.127820</td>\n",
              "      <td>0.112782</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>52926</th>\n",
              "      <td>0.347107</td>\n",
              "      <td>0.343802</td>\n",
              "      <td>0.345455</td>\n",
              "      <td>0.343802</td>\n",
              "      <td>0.347107</td>\n",
              "      <td>0.350413</td>\n",
              "      <td>0.348760</td>\n",
              "      <td>0.352066</td>\n",
              "      <td>0.353719</td>\n",
              "      <td>0.350413</td>\n",
              "      <td>0.350413</td>\n",
              "      <td>0.347107</td>\n",
              "      <td>0.348760</td>\n",
              "      <td>0.343802</td>\n",
              "      <td>0.345455</td>\n",
              "      <td>0.347107</td>\n",
              "      <td>0.348760</td>\n",
              "      <td>0.350413</td>\n",
              "      <td>0.353719</td>\n",
              "      <td>0.357025</td>\n",
              "      <td>0.357025</td>\n",
              "      <td>0.353719</td>\n",
              "      <td>0.355372</td>\n",
              "      <td>0.352066</td>\n",
              "      <td>0.350413</td>\n",
              "      <td>0.355372</td>\n",
              "      <td>0.355372</td>\n",
              "      <td>0.360331</td>\n",
              "      <td>0.365289</td>\n",
              "      <td>0.366942</td>\n",
              "      <td>0.368595</td>\n",
              "      <td>0.376860</td>\n",
              "      <td>0.380165</td>\n",
              "      <td>0.380165</td>\n",
              "      <td>0.388430</td>\n",
              "      <td>0.385124</td>\n",
              "      <td>0.390083</td>\n",
              "      <td>0.390083</td>\n",
              "      <td>0.390083</td>\n",
              "      <td>0.390083</td>\n",
              "      <td>...</td>\n",
              "      <td>0.357025</td>\n",
              "      <td>0.357025</td>\n",
              "      <td>0.358678</td>\n",
              "      <td>0.358678</td>\n",
              "      <td>0.363636</td>\n",
              "      <td>0.366942</td>\n",
              "      <td>0.365289</td>\n",
              "      <td>0.366942</td>\n",
              "      <td>0.363636</td>\n",
              "      <td>0.360331</td>\n",
              "      <td>0.360331</td>\n",
              "      <td>0.357025</td>\n",
              "      <td>0.355372</td>\n",
              "      <td>0.357025</td>\n",
              "      <td>0.358678</td>\n",
              "      <td>0.365289</td>\n",
              "      <td>0.363636</td>\n",
              "      <td>0.366942</td>\n",
              "      <td>0.365289</td>\n",
              "      <td>0.365289</td>\n",
              "      <td>0.366942</td>\n",
              "      <td>0.366942</td>\n",
              "      <td>0.363636</td>\n",
              "      <td>0.361983</td>\n",
              "      <td>0.360331</td>\n",
              "      <td>0.365289</td>\n",
              "      <td>0.370248</td>\n",
              "      <td>0.373554</td>\n",
              "      <td>0.373554</td>\n",
              "      <td>0.373554</td>\n",
              "      <td>0.376860</td>\n",
              "      <td>0.371901</td>\n",
              "      <td>0.370248</td>\n",
              "      <td>0.368595</td>\n",
              "      <td>0.370248</td>\n",
              "      <td>0.368595</td>\n",
              "      <td>0.370248</td>\n",
              "      <td>0.371901</td>\n",
              "      <td>0.373554</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>74631</th>\n",
              "      <td>0.468202</td>\n",
              "      <td>0.468202</td>\n",
              "      <td>0.468202</td>\n",
              "      <td>0.467105</td>\n",
              "      <td>0.467105</td>\n",
              "      <td>0.467105</td>\n",
              "      <td>0.467105</td>\n",
              "      <td>0.467105</td>\n",
              "      <td>0.467105</td>\n",
              "      <td>0.468202</td>\n",
              "      <td>0.470395</td>\n",
              "      <td>0.472588</td>\n",
              "      <td>0.472588</td>\n",
              "      <td>0.470395</td>\n",
              "      <td>0.469298</td>\n",
              "      <td>0.470395</td>\n",
              "      <td>0.470395</td>\n",
              "      <td>0.472588</td>\n",
              "      <td>0.476974</td>\n",
              "      <td>0.476974</td>\n",
              "      <td>0.474781</td>\n",
              "      <td>0.472588</td>\n",
              "      <td>0.474781</td>\n",
              "      <td>0.475877</td>\n",
              "      <td>0.479167</td>\n",
              "      <td>0.480263</td>\n",
              "      <td>0.476974</td>\n",
              "      <td>0.476974</td>\n",
              "      <td>0.475877</td>\n",
              "      <td>0.481360</td>\n",
              "      <td>0.481360</td>\n",
              "      <td>0.480263</td>\n",
              "      <td>0.478070</td>\n",
              "      <td>0.479167</td>\n",
              "      <td>0.476974</td>\n",
              "      <td>0.480263</td>\n",
              "      <td>0.482456</td>\n",
              "      <td>0.479167</td>\n",
              "      <td>0.480263</td>\n",
              "      <td>0.478070</td>\n",
              "      <td>...</td>\n",
              "      <td>0.844298</td>\n",
              "      <td>0.843202</td>\n",
              "      <td>0.839912</td>\n",
              "      <td>0.832237</td>\n",
              "      <td>0.821272</td>\n",
              "      <td>0.813596</td>\n",
              "      <td>0.809211</td>\n",
              "      <td>0.803728</td>\n",
              "      <td>0.798246</td>\n",
              "      <td>0.786184</td>\n",
              "      <td>0.777412</td>\n",
              "      <td>0.766447</td>\n",
              "      <td>0.756579</td>\n",
              "      <td>0.748904</td>\n",
              "      <td>0.740132</td>\n",
              "      <td>0.723684</td>\n",
              "      <td>0.710526</td>\n",
              "      <td>0.698465</td>\n",
              "      <td>0.688596</td>\n",
              "      <td>0.679825</td>\n",
              "      <td>0.671053</td>\n",
              "      <td>0.661184</td>\n",
              "      <td>0.649123</td>\n",
              "      <td>0.635965</td>\n",
              "      <td>0.627193</td>\n",
              "      <td>0.616228</td>\n",
              "      <td>0.610746</td>\n",
              "      <td>0.599781</td>\n",
              "      <td>0.589912</td>\n",
              "      <td>0.580044</td>\n",
              "      <td>0.574561</td>\n",
              "      <td>0.566886</td>\n",
              "      <td>0.560307</td>\n",
              "      <td>0.555921</td>\n",
              "      <td>0.547149</td>\n",
              "      <td>0.543860</td>\n",
              "      <td>0.542763</td>\n",
              "      <td>0.542763</td>\n",
              "      <td>0.538377</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>61762</th>\n",
              "      <td>0.135678</td>\n",
              "      <td>0.142379</td>\n",
              "      <td>0.144054</td>\n",
              "      <td>0.145729</td>\n",
              "      <td>0.140704</td>\n",
              "      <td>0.135678</td>\n",
              "      <td>0.135678</td>\n",
              "      <td>0.134003</td>\n",
              "      <td>0.139028</td>\n",
              "      <td>0.139028</td>\n",
              "      <td>0.139028</td>\n",
              "      <td>0.140704</td>\n",
              "      <td>0.142379</td>\n",
              "      <td>0.145729</td>\n",
              "      <td>0.142379</td>\n",
              "      <td>0.149079</td>\n",
              "      <td>0.149079</td>\n",
              "      <td>0.149079</td>\n",
              "      <td>0.149079</td>\n",
              "      <td>0.145729</td>\n",
              "      <td>0.144054</td>\n",
              "      <td>0.145729</td>\n",
              "      <td>0.150754</td>\n",
              "      <td>0.155779</td>\n",
              "      <td>0.159129</td>\n",
              "      <td>0.169179</td>\n",
              "      <td>0.172529</td>\n",
              "      <td>0.172529</td>\n",
              "      <td>0.170854</td>\n",
              "      <td>0.172529</td>\n",
              "      <td>0.167504</td>\n",
              "      <td>0.164154</td>\n",
              "      <td>0.164154</td>\n",
              "      <td>0.162479</td>\n",
              "      <td>0.160804</td>\n",
              "      <td>0.162479</td>\n",
              "      <td>0.162479</td>\n",
              "      <td>0.159129</td>\n",
              "      <td>0.160804</td>\n",
              "      <td>0.162479</td>\n",
              "      <td>...</td>\n",
              "      <td>0.134003</td>\n",
              "      <td>0.137353</td>\n",
              "      <td>0.142379</td>\n",
              "      <td>0.147404</td>\n",
              "      <td>0.152429</td>\n",
              "      <td>0.154104</td>\n",
              "      <td>0.159129</td>\n",
              "      <td>0.169179</td>\n",
              "      <td>0.172529</td>\n",
              "      <td>0.179229</td>\n",
              "      <td>0.179229</td>\n",
              "      <td>0.185930</td>\n",
              "      <td>0.187605</td>\n",
              "      <td>0.182580</td>\n",
              "      <td>0.184255</td>\n",
              "      <td>0.184255</td>\n",
              "      <td>0.180905</td>\n",
              "      <td>0.180905</td>\n",
              "      <td>0.182580</td>\n",
              "      <td>0.184255</td>\n",
              "      <td>0.184255</td>\n",
              "      <td>0.184255</td>\n",
              "      <td>0.182580</td>\n",
              "      <td>0.179229</td>\n",
              "      <td>0.179229</td>\n",
              "      <td>0.170854</td>\n",
              "      <td>0.165829</td>\n",
              "      <td>0.164154</td>\n",
              "      <td>0.155779</td>\n",
              "      <td>0.157454</td>\n",
              "      <td>0.150754</td>\n",
              "      <td>0.149079</td>\n",
              "      <td>0.145729</td>\n",
              "      <td>0.145729</td>\n",
              "      <td>0.140704</td>\n",
              "      <td>0.139028</td>\n",
              "      <td>0.139028</td>\n",
              "      <td>0.134003</td>\n",
              "      <td>0.128978</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>61298</th>\n",
              "      <td>0.132653</td>\n",
              "      <td>0.134354</td>\n",
              "      <td>0.130952</td>\n",
              "      <td>0.127551</td>\n",
              "      <td>0.127551</td>\n",
              "      <td>0.125850</td>\n",
              "      <td>0.127551</td>\n",
              "      <td>0.125850</td>\n",
              "      <td>0.122449</td>\n",
              "      <td>0.130952</td>\n",
              "      <td>0.129252</td>\n",
              "      <td>0.129252</td>\n",
              "      <td>0.132653</td>\n",
              "      <td>0.130952</td>\n",
              "      <td>0.130952</td>\n",
              "      <td>0.130952</td>\n",
              "      <td>0.125850</td>\n",
              "      <td>0.125850</td>\n",
              "      <td>0.130952</td>\n",
              "      <td>0.130952</td>\n",
              "      <td>0.130952</td>\n",
              "      <td>0.129252</td>\n",
              "      <td>0.129252</td>\n",
              "      <td>0.130952</td>\n",
              "      <td>0.136054</td>\n",
              "      <td>0.134354</td>\n",
              "      <td>0.134354</td>\n",
              "      <td>0.139456</td>\n",
              "      <td>0.132653</td>\n",
              "      <td>0.129252</td>\n",
              "      <td>0.127551</td>\n",
              "      <td>0.125850</td>\n",
              "      <td>0.136054</td>\n",
              "      <td>0.141156</td>\n",
              "      <td>0.144558</td>\n",
              "      <td>0.149660</td>\n",
              "      <td>0.156463</td>\n",
              "      <td>0.158163</td>\n",
              "      <td>0.154762</td>\n",
              "      <td>0.151361</td>\n",
              "      <td>...</td>\n",
              "      <td>0.110544</td>\n",
              "      <td>0.117347</td>\n",
              "      <td>0.115646</td>\n",
              "      <td>0.117347</td>\n",
              "      <td>0.124150</td>\n",
              "      <td>0.127551</td>\n",
              "      <td>0.134354</td>\n",
              "      <td>0.137755</td>\n",
              "      <td>0.142857</td>\n",
              "      <td>0.144558</td>\n",
              "      <td>0.146259</td>\n",
              "      <td>0.144558</td>\n",
              "      <td>0.142857</td>\n",
              "      <td>0.141156</td>\n",
              "      <td>0.139456</td>\n",
              "      <td>0.141156</td>\n",
              "      <td>0.139456</td>\n",
              "      <td>0.139456</td>\n",
              "      <td>0.139456</td>\n",
              "      <td>0.141156</td>\n",
              "      <td>0.136054</td>\n",
              "      <td>0.134354</td>\n",
              "      <td>0.130952</td>\n",
              "      <td>0.124150</td>\n",
              "      <td>0.115646</td>\n",
              "      <td>0.110544</td>\n",
              "      <td>0.105442</td>\n",
              "      <td>0.103741</td>\n",
              "      <td>0.098639</td>\n",
              "      <td>0.100340</td>\n",
              "      <td>0.095238</td>\n",
              "      <td>0.098639</td>\n",
              "      <td>0.096939</td>\n",
              "      <td>0.088435</td>\n",
              "      <td>0.091837</td>\n",
              "      <td>0.086735</td>\n",
              "      <td>0.079932</td>\n",
              "      <td>0.079932</td>\n",
              "      <td>0.078231</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>70372</th>\n",
              "      <td>0.372124</td>\n",
              "      <td>0.366712</td>\n",
              "      <td>0.362652</td>\n",
              "      <td>0.359946</td>\n",
              "      <td>0.359946</td>\n",
              "      <td>0.350474</td>\n",
              "      <td>0.343708</td>\n",
              "      <td>0.332882</td>\n",
              "      <td>0.332882</td>\n",
              "      <td>0.336942</td>\n",
              "      <td>0.332882</td>\n",
              "      <td>0.331529</td>\n",
              "      <td>0.328823</td>\n",
              "      <td>0.328823</td>\n",
              "      <td>0.326116</td>\n",
              "      <td>0.326116</td>\n",
              "      <td>0.319350</td>\n",
              "      <td>0.312585</td>\n",
              "      <td>0.307172</td>\n",
              "      <td>0.300406</td>\n",
              "      <td>0.299053</td>\n",
              "      <td>0.300406</td>\n",
              "      <td>0.297700</td>\n",
              "      <td>0.300406</td>\n",
              "      <td>0.294993</td>\n",
              "      <td>0.300406</td>\n",
              "      <td>0.293640</td>\n",
              "      <td>0.296346</td>\n",
              "      <td>0.290934</td>\n",
              "      <td>0.286874</td>\n",
              "      <td>0.284168</td>\n",
              "      <td>0.285521</td>\n",
              "      <td>0.284168</td>\n",
              "      <td>0.284168</td>\n",
              "      <td>0.284168</td>\n",
              "      <td>0.271989</td>\n",
              "      <td>0.262517</td>\n",
              "      <td>0.263870</td>\n",
              "      <td>0.261164</td>\n",
              "      <td>0.273342</td>\n",
              "      <td>...</td>\n",
              "      <td>0.248985</td>\n",
              "      <td>0.253045</td>\n",
              "      <td>0.257104</td>\n",
              "      <td>0.258457</td>\n",
              "      <td>0.265223</td>\n",
              "      <td>0.265223</td>\n",
              "      <td>0.270636</td>\n",
              "      <td>0.267930</td>\n",
              "      <td>0.266576</td>\n",
              "      <td>0.267930</td>\n",
              "      <td>0.273342</td>\n",
              "      <td>0.277402</td>\n",
              "      <td>0.278755</td>\n",
              "      <td>0.282815</td>\n",
              "      <td>0.280108</td>\n",
              "      <td>0.282815</td>\n",
              "      <td>0.282815</td>\n",
              "      <td>0.290934</td>\n",
              "      <td>0.294993</td>\n",
              "      <td>0.296346</td>\n",
              "      <td>0.293640</td>\n",
              "      <td>0.300406</td>\n",
              "      <td>0.304465</td>\n",
              "      <td>0.308525</td>\n",
              "      <td>0.305819</td>\n",
              "      <td>0.299053</td>\n",
              "      <td>0.296346</td>\n",
              "      <td>0.294993</td>\n",
              "      <td>0.294993</td>\n",
              "      <td>0.293640</td>\n",
              "      <td>0.296346</td>\n",
              "      <td>0.299053</td>\n",
              "      <td>0.297700</td>\n",
              "      <td>0.301759</td>\n",
              "      <td>0.303112</td>\n",
              "      <td>0.301759</td>\n",
              "      <td>0.300406</td>\n",
              "      <td>0.305819</td>\n",
              "      <td>0.301759</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6129</th>\n",
              "      <td>0.231076</td>\n",
              "      <td>0.229084</td>\n",
              "      <td>0.229084</td>\n",
              "      <td>0.233068</td>\n",
              "      <td>0.233068</td>\n",
              "      <td>0.232072</td>\n",
              "      <td>0.232072</td>\n",
              "      <td>0.231076</td>\n",
              "      <td>0.235060</td>\n",
              "      <td>0.230080</td>\n",
              "      <td>0.230080</td>\n",
              "      <td>0.230080</td>\n",
              "      <td>0.226096</td>\n",
              "      <td>0.226096</td>\n",
              "      <td>0.226096</td>\n",
              "      <td>0.230080</td>\n",
              "      <td>0.230080</td>\n",
              "      <td>0.233068</td>\n",
              "      <td>0.235060</td>\n",
              "      <td>0.233068</td>\n",
              "      <td>0.229084</td>\n",
              "      <td>0.229084</td>\n",
              "      <td>0.227092</td>\n",
              "      <td>0.225100</td>\n",
              "      <td>0.227092</td>\n",
              "      <td>0.228088</td>\n",
              "      <td>0.229084</td>\n",
              "      <td>0.230080</td>\n",
              "      <td>0.231076</td>\n",
              "      <td>0.233068</td>\n",
              "      <td>0.236056</td>\n",
              "      <td>0.240040</td>\n",
              "      <td>0.244024</td>\n",
              "      <td>0.246016</td>\n",
              "      <td>0.247012</td>\n",
              "      <td>0.249004</td>\n",
              "      <td>0.250000</td>\n",
              "      <td>0.247012</td>\n",
              "      <td>0.246016</td>\n",
              "      <td>0.248008</td>\n",
              "      <td>...</td>\n",
              "      <td>0.260956</td>\n",
              "      <td>0.259960</td>\n",
              "      <td>0.264940</td>\n",
              "      <td>0.264940</td>\n",
              "      <td>0.265936</td>\n",
              "      <td>0.263944</td>\n",
              "      <td>0.263944</td>\n",
              "      <td>0.263944</td>\n",
              "      <td>0.264940</td>\n",
              "      <td>0.264940</td>\n",
              "      <td>0.265936</td>\n",
              "      <td>0.267928</td>\n",
              "      <td>0.270916</td>\n",
              "      <td>0.270916</td>\n",
              "      <td>0.270916</td>\n",
              "      <td>0.270916</td>\n",
              "      <td>0.266932</td>\n",
              "      <td>0.271912</td>\n",
              "      <td>0.271912</td>\n",
              "      <td>0.270916</td>\n",
              "      <td>0.268924</td>\n",
              "      <td>0.265936</td>\n",
              "      <td>0.267928</td>\n",
              "      <td>0.266932</td>\n",
              "      <td>0.269920</td>\n",
              "      <td>0.267928</td>\n",
              "      <td>0.273904</td>\n",
              "      <td>0.272908</td>\n",
              "      <td>0.275896</td>\n",
              "      <td>0.271912</td>\n",
              "      <td>0.269920</td>\n",
              "      <td>0.267928</td>\n",
              "      <td>0.264940</td>\n",
              "      <td>0.264940</td>\n",
              "      <td>0.270916</td>\n",
              "      <td>0.272908</td>\n",
              "      <td>0.272908</td>\n",
              "      <td>0.271912</td>\n",
              "      <td>0.271912</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>77595</th>\n",
              "      <td>0.502698</td>\n",
              "      <td>0.501799</td>\n",
              "      <td>0.504496</td>\n",
              "      <td>0.502698</td>\n",
              "      <td>0.501799</td>\n",
              "      <td>0.504496</td>\n",
              "      <td>0.503597</td>\n",
              "      <td>0.501799</td>\n",
              "      <td>0.498201</td>\n",
              "      <td>0.499101</td>\n",
              "      <td>0.500899</td>\n",
              "      <td>0.500899</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.498201</td>\n",
              "      <td>0.501799</td>\n",
              "      <td>0.503597</td>\n",
              "      <td>0.507194</td>\n",
              "      <td>0.505396</td>\n",
              "      <td>0.502698</td>\n",
              "      <td>0.498201</td>\n",
              "      <td>0.492806</td>\n",
              "      <td>0.491007</td>\n",
              "      <td>0.489209</td>\n",
              "      <td>0.488309</td>\n",
              "      <td>0.481115</td>\n",
              "      <td>0.477518</td>\n",
              "      <td>0.474820</td>\n",
              "      <td>0.474820</td>\n",
              "      <td>0.473921</td>\n",
              "      <td>0.473921</td>\n",
              "      <td>0.473022</td>\n",
              "      <td>0.469424</td>\n",
              "      <td>0.467626</td>\n",
              "      <td>0.467626</td>\n",
              "      <td>0.469424</td>\n",
              "      <td>0.468525</td>\n",
              "      <td>0.469424</td>\n",
              "      <td>0.467626</td>\n",
              "      <td>0.466727</td>\n",
              "      <td>...</td>\n",
              "      <td>0.513489</td>\n",
              "      <td>0.516187</td>\n",
              "      <td>0.516187</td>\n",
              "      <td>0.513489</td>\n",
              "      <td>0.510791</td>\n",
              "      <td>0.510791</td>\n",
              "      <td>0.512590</td>\n",
              "      <td>0.512590</td>\n",
              "      <td>0.512590</td>\n",
              "      <td>0.511691</td>\n",
              "      <td>0.508993</td>\n",
              "      <td>0.505396</td>\n",
              "      <td>0.507194</td>\n",
              "      <td>0.507194</td>\n",
              "      <td>0.504496</td>\n",
              "      <td>0.502698</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.499101</td>\n",
              "      <td>0.499101</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.496403</td>\n",
              "      <td>0.493705</td>\n",
              "      <td>0.489209</td>\n",
              "      <td>0.486511</td>\n",
              "      <td>0.488309</td>\n",
              "      <td>0.487410</td>\n",
              "      <td>0.485612</td>\n",
              "      <td>0.482914</td>\n",
              "      <td>0.480216</td>\n",
              "      <td>0.479317</td>\n",
              "      <td>0.480216</td>\n",
              "      <td>0.482014</td>\n",
              "      <td>0.482014</td>\n",
              "      <td>0.482014</td>\n",
              "      <td>0.480216</td>\n",
              "      <td>0.479317</td>\n",
              "      <td>0.479317</td>\n",
              "      <td>0.482014</td>\n",
              "      <td>0.481115</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23621</th>\n",
              "      <td>0.529851</td>\n",
              "      <td>0.534115</td>\n",
              "      <td>0.531983</td>\n",
              "      <td>0.528785</td>\n",
              "      <td>0.527719</td>\n",
              "      <td>0.533049</td>\n",
              "      <td>0.537313</td>\n",
              "      <td>0.535181</td>\n",
              "      <td>0.530917</td>\n",
              "      <td>0.530917</td>\n",
              "      <td>0.529851</td>\n",
              "      <td>0.526652</td>\n",
              "      <td>0.525586</td>\n",
              "      <td>0.533049</td>\n",
              "      <td>0.529851</td>\n",
              "      <td>0.526652</td>\n",
              "      <td>0.519190</td>\n",
              "      <td>0.519190</td>\n",
              "      <td>0.524520</td>\n",
              "      <td>0.521322</td>\n",
              "      <td>0.519190</td>\n",
              "      <td>0.514925</td>\n",
              "      <td>0.518124</td>\n",
              "      <td>0.520256</td>\n",
              "      <td>0.520256</td>\n",
              "      <td>0.521322</td>\n",
              "      <td>0.526652</td>\n",
              "      <td>0.526652</td>\n",
              "      <td>0.525586</td>\n",
              "      <td>0.523454</td>\n",
              "      <td>0.523454</td>\n",
              "      <td>0.527719</td>\n",
              "      <td>0.529851</td>\n",
              "      <td>0.524520</td>\n",
              "      <td>0.521322</td>\n",
              "      <td>0.519190</td>\n",
              "      <td>0.524520</td>\n",
              "      <td>0.525586</td>\n",
              "      <td>0.523454</td>\n",
              "      <td>0.521322</td>\n",
              "      <td>...</td>\n",
              "      <td>0.531983</td>\n",
              "      <td>0.536247</td>\n",
              "      <td>0.535181</td>\n",
              "      <td>0.536247</td>\n",
              "      <td>0.534115</td>\n",
              "      <td>0.534115</td>\n",
              "      <td>0.528785</td>\n",
              "      <td>0.526652</td>\n",
              "      <td>0.524520</td>\n",
              "      <td>0.529851</td>\n",
              "      <td>0.526652</td>\n",
              "      <td>0.524520</td>\n",
              "      <td>0.519190</td>\n",
              "      <td>0.520256</td>\n",
              "      <td>0.522388</td>\n",
              "      <td>0.524520</td>\n",
              "      <td>0.518124</td>\n",
              "      <td>0.514925</td>\n",
              "      <td>0.513859</td>\n",
              "      <td>0.514925</td>\n",
              "      <td>0.514925</td>\n",
              "      <td>0.513859</td>\n",
              "      <td>0.513859</td>\n",
              "      <td>0.511727</td>\n",
              "      <td>0.507463</td>\n",
              "      <td>0.509595</td>\n",
              "      <td>0.512793</td>\n",
              "      <td>0.513859</td>\n",
              "      <td>0.512793</td>\n",
              "      <td>0.507463</td>\n",
              "      <td>0.508529</td>\n",
              "      <td>0.510661</td>\n",
              "      <td>0.508529</td>\n",
              "      <td>0.509595</td>\n",
              "      <td>0.512793</td>\n",
              "      <td>0.513859</td>\n",
              "      <td>0.512793</td>\n",
              "      <td>0.512793</td>\n",
              "      <td>0.514925</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39106</th>\n",
              "      <td>0.328685</td>\n",
              "      <td>0.338645</td>\n",
              "      <td>0.340637</td>\n",
              "      <td>0.340637</td>\n",
              "      <td>0.328685</td>\n",
              "      <td>0.312749</td>\n",
              "      <td>0.314741</td>\n",
              "      <td>0.318725</td>\n",
              "      <td>0.320717</td>\n",
              "      <td>0.320717</td>\n",
              "      <td>0.318725</td>\n",
              "      <td>0.314741</td>\n",
              "      <td>0.320717</td>\n",
              "      <td>0.326693</td>\n",
              "      <td>0.332669</td>\n",
              "      <td>0.328685</td>\n",
              "      <td>0.330677</td>\n",
              "      <td>0.330677</td>\n",
              "      <td>0.338645</td>\n",
              "      <td>0.342629</td>\n",
              "      <td>0.346614</td>\n",
              "      <td>0.342629</td>\n",
              "      <td>0.336653</td>\n",
              "      <td>0.334661</td>\n",
              "      <td>0.338645</td>\n",
              "      <td>0.346614</td>\n",
              "      <td>0.350598</td>\n",
              "      <td>0.344622</td>\n",
              "      <td>0.342629</td>\n",
              "      <td>0.346614</td>\n",
              "      <td>0.350598</td>\n",
              "      <td>0.366534</td>\n",
              "      <td>0.372510</td>\n",
              "      <td>0.376494</td>\n",
              "      <td>0.368526</td>\n",
              "      <td>0.356574</td>\n",
              "      <td>0.352590</td>\n",
              "      <td>0.354582</td>\n",
              "      <td>0.356574</td>\n",
              "      <td>0.350598</td>\n",
              "      <td>...</td>\n",
              "      <td>0.340637</td>\n",
              "      <td>0.336653</td>\n",
              "      <td>0.340637</td>\n",
              "      <td>0.352590</td>\n",
              "      <td>0.362550</td>\n",
              "      <td>0.362550</td>\n",
              "      <td>0.364542</td>\n",
              "      <td>0.366534</td>\n",
              "      <td>0.368526</td>\n",
              "      <td>0.368526</td>\n",
              "      <td>0.366534</td>\n",
              "      <td>0.358566</td>\n",
              "      <td>0.344622</td>\n",
              "      <td>0.338645</td>\n",
              "      <td>0.338645</td>\n",
              "      <td>0.336653</td>\n",
              "      <td>0.336653</td>\n",
              "      <td>0.334661</td>\n",
              "      <td>0.326693</td>\n",
              "      <td>0.320717</td>\n",
              "      <td>0.322709</td>\n",
              "      <td>0.332669</td>\n",
              "      <td>0.336653</td>\n",
              "      <td>0.322709</td>\n",
              "      <td>0.310757</td>\n",
              "      <td>0.304781</td>\n",
              "      <td>0.308765</td>\n",
              "      <td>0.316733</td>\n",
              "      <td>0.320717</td>\n",
              "      <td>0.316733</td>\n",
              "      <td>0.308765</td>\n",
              "      <td>0.294821</td>\n",
              "      <td>0.302789</td>\n",
              "      <td>0.308765</td>\n",
              "      <td>0.310757</td>\n",
              "      <td>0.308765</td>\n",
              "      <td>0.312749</td>\n",
              "      <td>0.312749</td>\n",
              "      <td>0.308765</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>83356 rows × 260 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "              0         1         2  ...       257       258  259\n",
              "18337  0.082707  0.071429  0.060150  ...  0.127820  0.112782    0\n",
              "52926  0.347107  0.343802  0.345455  ...  0.371901  0.373554    0\n",
              "74631  0.468202  0.468202  0.468202  ...  0.542763  0.538377    4\n",
              "61762  0.135678  0.142379  0.144054  ...  0.134003  0.128978    1\n",
              "61298  0.132653  0.134354  0.130952  ...  0.079932  0.078231    1\n",
              "...         ...       ...       ...  ...       ...       ...  ...\n",
              "70372  0.372124  0.366712  0.362652  ...  0.305819  0.301759    3\n",
              "6129   0.231076  0.229084  0.229084  ...  0.271912  0.271912    0\n",
              "77595  0.502698  0.501799  0.504496  ...  0.482014  0.481115    2\n",
              "23621  0.529851  0.534115  0.531983  ...  0.512793  0.514925    0\n",
              "39106  0.328685  0.338645  0.340637  ...  0.312749  0.308765    0\n",
              "\n",
              "[83356 rows x 260 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MTZ-cdJOQ2Z3"
      },
      "source": [
        "##the distribution of labels in train dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wz1q1a5pR2r9",
        "outputId": "ba0fef6c-7074-4df5-f870-890b4d540714"
      },
      "source": [
        "train_df['259']=train_df['259'].astype(int)\n",
        "equilibre=train_df['259'].value_counts()\n",
        "print(equilibre)\n",
        "train_df.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0    59828\n",
            "2     6443\n",
            "1     5791\n",
            "3     5691\n",
            "4     5603\n",
            "Name: 259, dtype: int64\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(83356, 260)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 575
        },
        "id": "6YprxJanSE6j",
        "outputId": "003c7c6b-057c-4f2f-9c77-ab0d25dbe4d2"
      },
      "source": [
        "plt.figure(figsize=(20,10))\n",
        "my_circle=plt.Circle( (0,0), 0.7, color='white')\n",
        "plt.pie(equilibre, labels=['N','R','L','V','PB'], colors=['red','green','blue','skyblue','orange'],autopct='%1.1f%%')\n",
        "p=plt.gcf()\n",
        "p.gca().add_artist(my_circle)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "89XsaSQ1YG34"
      },
      "source": [
        "target_train=train_df['259']\n",
        "target_test=test_df['259']\n",
        "y_train=to_categorical(target_train)\n",
        "y_test=to_categorical(target_test)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GtGAq4PtYcBE"
      },
      "source": [
        "\n",
        "\n",
        "X_train=train_df.iloc[:,:259].values\n",
        "X_test=test_df.iloc[:,:259].values\n",
        "#for i in range(len(X_train)):\n",
        "#    X_train[i,:186]= add_gaussian_noise(X_train[i,:186])\n",
        "X_train = X_train.reshape(len(X_train), X_train.shape[1],1)\n",
        "X_test = X_test.reshape(len(X_test), X_test.shape[1],1)\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uVRuIv5hYiog",
        "outputId": "040a7e9e-e7c0-4d44-ab05-ef2b94aa9821"
      },
      "source": [
        "X_train.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(83356, 259, 1)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qB9KDPLrOWfj"
      },
      "source": [
        "##network"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R-_V1gt7YpDD"
      },
      "source": [
        "def network(X_train,y_train,X_test,y_test):\n",
        "    \n",
        "\n",
        "    im_shape=(X_train.shape[1],1)\n",
        "    inputs_cnn=Input(shape=(im_shape), name='inputs_cnn')\n",
        "    conv1_1=Convolution1D(16, (6), activation='relu', input_shape=im_shape)(inputs_cnn)\n",
        "    conv1_1=BatchNormalization()(conv1_1)\n",
        "    pool1=MaxPool1D(pool_size=(3), strides=(2), padding=\"same\")(conv1_1)\n",
        "    conv2_1=Convolution1D(32, (3), activation='relu', input_shape=im_shape)(pool1)\n",
        "    conv2_1=BatchNormalization()(conv2_1)\n",
        "    pool2=MaxPool1D(pool_size=(2), strides=(2), padding=\"same\")(conv2_1)\n",
        "    conv3_1=Convolution1D(64, (3), activation='relu', input_shape=im_shape)(pool2)\n",
        "    conv3_1=BatchNormalization()(conv3_1)\n",
        "    pool3=MaxPool1D(pool_size=(2), strides=(2), padding=\"same\")(conv3_1)\n",
        "    flatten=Flatten()(pool3)\n",
        "    dense_end1 = Dense(64, activation='relu')(flatten)\n",
        "    dense_end2 = Dense(32, activation='relu')(dense_end1)\n",
        "    main_output = Dense(5, activation='softmax', name='main_output')(dense_end2)\n",
        "    \n",
        "    \n",
        "    model = Model(inputs= inputs_cnn, outputs=main_output)\n",
        "    model.compile(optimizer='adam', loss='categorical_crossentropy',metrics = ['accuracy'])\n",
        "    model.summary()\n",
        "    \n",
        "    \n",
        "    callbacks = [EarlyStopping(monitor='val_loss', patience=8),\n",
        "             ModelCheckpoint(filepath='best_model.h5', monitor='val_loss', save_best_only=True)]\n",
        "\n",
        "    history=model.fit(X_train, y_train,epochs=10,callbacks=callbacks, batch_size=32,validation_data=(X_test,y_test))\n",
        "    model.load_weights('best_model.h5')\n",
        "    return(model,history)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FlJA0hssYtlE"
      },
      "source": [
        "def evaluate_model(history,X_test,y_test,model):\n",
        "    scores = model.evaluate((X_test),y_test, verbose=0)\n",
        "    print(\"Accuracy: %.2f%%\" % (scores[1]*100))\n",
        "    \n",
        "    print(history)\n",
        "    fig1, ax_acc = plt.subplots()\n",
        "    plt.plot(history.history['accuracy'])\n",
        "    plt.plot(history.history['val_accuracy'])\n",
        "    plt.xlabel('Epoch')\n",
        "    plt.ylabel('Accuracy')\n",
        "    plt.title('Model - Accuracy')\n",
        "    plt.legend(['Training', 'Validation'], loc='lower right')\n",
        "    plt.show()\n",
        "    \n",
        "    fig2, ax_loss = plt.subplots()\n",
        "    plt.xlabel('Epoch')\n",
        "    plt.ylabel('Loss')\n",
        "    plt.title('Model- Loss')\n",
        "    plt.legend(['Training', 'Validation'], loc='upper right')\n",
        "    plt.plot(history.history['loss'])\n",
        "    plt.plot(history.history['val_loss'])\n",
        "    plt.show()\n",
        "    target_names=['0','1','2','3','4']\n",
        "    \n",
        "    y_true=[]\n",
        "    for element in y_test:\n",
        "        y_true.append(np.argmax(element))\n",
        "    prediction_proba=model.predict(X_test)\n",
        "    prediction=np.argmax(prediction_proba,axis=1)\n",
        "    cnf_matrix = confusion_matrix(y_true, prediction)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NpA-_r5cYzzR",
        "outputId": "c0903c24-2ec7-4fb8-89c6-cb45cdc75806"
      },
      "source": [
        "from keras.layers import Dense, Convolution1D, MaxPool1D, Flatten, Dropout\n",
        "from keras.layers import Input\n",
        "from keras.models import Model\n",
        "from keras.layers.normalization import BatchNormalization\n",
        "import keras\n",
        "from keras.callbacks import EarlyStopping, ModelCheckpoint\n",
        "\n",
        "model,history=network(X_train,y_train,X_test,y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "inputs_cnn (InputLayer)      [(None, 259, 1)]          0         \n",
            "_________________________________________________________________\n",
            "conv1d (Conv1D)              (None, 254, 16)           112       \n",
            "_________________________________________________________________\n",
            "batch_normalization (BatchNo (None, 254, 16)           64        \n",
            "_________________________________________________________________\n",
            "max_pooling1d (MaxPooling1D) (None, 127, 16)           0         \n",
            "_________________________________________________________________\n",
            "conv1d_1 (Conv1D)            (None, 125, 32)           1568      \n",
            "_________________________________________________________________\n",
            "batch_normalization_1 (Batch (None, 125, 32)           128       \n",
            "_________________________________________________________________\n",
            "max_pooling1d_1 (MaxPooling1 (None, 63, 32)            0         \n",
            "_________________________________________________________________\n",
            "conv1d_2 (Conv1D)            (None, 61, 64)            6208      \n",
            "_________________________________________________________________\n",
            "batch_normalization_2 (Batch (None, 61, 64)            256       \n",
            "_________________________________________________________________\n",
            "max_pooling1d_2 (MaxPooling1 (None, 31, 64)            0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 1984)              0         \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 64)                127040    \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 32)                2080      \n",
            "_________________________________________________________________\n",
            "main_output (Dense)          (None, 5)                 165       \n",
            "=================================================================\n",
            "Total params: 137,621\n",
            "Trainable params: 137,397\n",
            "Non-trainable params: 224\n",
            "_________________________________________________________________\n",
            "Epoch 1/10\n",
            "2605/2605 [==============================] - 59s 22ms/step - loss: 0.1560 - accuracy: 0.9521 - val_loss: 0.2973 - val_accuracy: 0.9202\n",
            "Epoch 2/10\n",
            "2605/2605 [==============================] - 57s 22ms/step - loss: 0.0354 - accuracy: 0.9885 - val_loss: 1.0692 - val_accuracy: 0.8291\n",
            "Epoch 3/10\n",
            "2605/2605 [==============================] - 57s 22ms/step - loss: 0.0210 - accuracy: 0.9930 - val_loss: 0.7305 - val_accuracy: 0.8779\n",
            "Epoch 4/10\n",
            "2605/2605 [==============================] - 59s 22ms/step - loss: 0.0159 - accuracy: 0.9947 - val_loss: 0.8294 - val_accuracy: 0.8539\n",
            "Epoch 5/10\n",
            "2605/2605 [==============================] - 62s 24ms/step - loss: 0.0140 - accuracy: 0.9954 - val_loss: 0.6550 - val_accuracy: 0.9025\n",
            "Epoch 6/10\n",
            "2605/2605 [==============================] - 57s 22ms/step - loss: 0.0095 - accuracy: 0.9971 - val_loss: 0.7388 - val_accuracy: 0.8892\n",
            "Epoch 7/10\n",
            "2605/2605 [==============================] - 58s 22ms/step - loss: 0.0085 - accuracy: 0.9973 - val_loss: 0.8403 - val_accuracy: 0.8765\n",
            "Epoch 8/10\n",
            "2605/2605 [==============================] - 58s 22ms/step - loss: 0.0090 - accuracy: 0.9975 - val_loss: 0.9643 - val_accuracy: 0.9050\n",
            "Epoch 9/10\n",
            "2605/2605 [==============================] - 58s 22ms/step - loss: 0.0069 - accuracy: 0.9976 - val_loss: 0.7186 - val_accuracy: 0.9031\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ehiciDUlOkzb"
      },
      "source": [
        "##evaluate model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 770
        },
        "id": "cupkp4O4Y6Pz",
        "outputId": "09975121-06d3-4f90-eef2-32220bef3e87"
      },
      "source": [
        "evaluate_model(history,X_test,y_test,model)\n",
        "y_pred=model.predict(X_test)\n",
        "\n",
        "print(y_pred) \n",
        "print(\"Result from real time data included in testing dataset:\")\n",
        "Y_pred_classes = np.argmax(y_pred,axis = 1) \n",
        "print(Y_pred_classes[0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy: 92.02%\n",
            "<tensorflow.python.keras.callbacks.History object at 0x7f87ad289150>\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "[[9.9945146e-01 3.4695564e-04 5.2469059e-05 1.3080981e-08 1.4922177e-04]\n",
            " [6.0846467e-08 6.2877606e-13 6.1631991e-11 2.8080405e-13 9.9999988e-01]\n",
            " [1.0000000e+00 4.0689768e-10 7.1822426e-10 5.6068691e-13 5.8739152e-11]\n",
            " ...\n",
            " [9.9787009e-01 1.6968482e-03 3.5158519e-05 2.7811949e-04 1.1966426e-04]\n",
            " [2.7216126e-07 2.8636562e-11 1.1691975e-09 5.7869955e-11 9.9999976e-01]\n",
            " [9.9997652e-01 2.2958784e-05 1.7458810e-07 4.6119879e-08 3.6570654e-07]]\n",
            "Result from real time data included in testing dataset:\n",
            "0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "A83pksncObVv"
      },
      "source": [
        "##confusion matrix plotter"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 747
        },
        "id": "bjkYCxcMdbiB",
        "outputId": "f47e3827-8eb0-4e09-85b1-3dfcf62266d1"
      },
      "source": [
        "import itertools\n",
        "def plot_confusion_matrix(cm, classes,\n",
        "                          normalize=False,\n",
        "                          title='Confusion matrix',\n",
        "                          cmap=plt.cm.Blues):\n",
        "    \"\"\"\n",
        "    This function prints and plots the confusion matrix.\n",
        "    Normalization can be applied by setting `normalize=True`.\n",
        "    \"\"\"\n",
        "    if normalize:\n",
        "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "        print(\"Normalized confusion matrix\")\n",
        "    else:\n",
        "        print('Confusion matrix, without normalization')\n",
        "\n",
        "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "    plt.title(title)\n",
        "    plt.colorbar()\n",
        "    tick_marks = np.arange(len(classes))\n",
        "    plt.xticks(tick_marks, classes, rotation=45)\n",
        "    plt.yticks(tick_marks, classes)\n",
        "\n",
        "    fmt = '.2f' if normalize else 'd'\n",
        "    thresh = cm.max() / 2.\n",
        "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
        "        plt.text(j, i, format(cm[i, j], fmt),\n",
        "                 horizontalalignment=\"center\",\n",
        "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
        "\n",
        "    plt.tight_layout()\n",
        "    plt.ylabel('True label')\n",
        "    plt.xlabel('Predicted label')\n",
        "\n",
        "# Compute confusion matrix\n",
        "cnf_matrix = confusion_matrix(y_test.argmax(axis=1), y_pred.argmax(axis=1))\n",
        "np.set_printoptions(precision=2)\n",
        "\n",
        "# Plot non-normalized confusion matrix\n",
        "plt.figure(figsize=(10, 10))\n",
        "plot_confusion_matrix(cnf_matrix, classes=['N', 'R', 'L', 'V', 'PB'],normalize=True,\n",
        "                      title='Confusion matrix, with normalization')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Normalized confusion matrix\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "36FdJI3AOexm"
      },
      "source": [
        "##classification report"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z6HrSsgSdg1q",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3f8fa208-36a4-405b-db47-5518506ca37b"
      },
      "source": [
        "classification_report(y_test.argmax(axis=1), y_pred.argmax(axis=1), labels = [0, 1, 2, 3, 4], target_names=['N', 'R', 'L', 'V', 'PB'],\n",
        "                      output_dict = True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'L': {'f1-score': 0.8661894273127754,\n",
              "  'precision': 0.777942631058358,\n",
              "  'recall': 0.9770186335403727,\n",
              "  'support': 1610},\n",
              " 'N': {'f1-score': 0.9468222183527772,\n",
              "  'precision': 0.9880796627416776,\n",
              "  'recall': 0.9088721000200575,\n",
              "  'support': 14957},\n",
              " 'PB': {'f1-score': 0.9140391636242947,\n",
              "  'precision': 0.8536887786732796,\n",
              "  'recall': 0.9835714285714285,\n",
              "  'support': 1400},\n",
              " 'R': {'f1-score': 0.8702337023370235,\n",
              "  'precision': 0.783499446290144,\n",
              "  'recall': 0.9785615491009682,\n",
              "  'support': 1446},\n",
              " 'V': {'f1-score': 0.7931937172774869,\n",
              "  'precision': 0.7412844036697248,\n",
              "  'recall': 0.8529204785362421,\n",
              "  'support': 1421},\n",
              " 'accuracy': 0.9201785542862628,\n",
              " 'macro avg': {'f1-score': 0.8780956457808715,\n",
              "  'precision': 0.8288989844866368,\n",
              "  'recall': 0.9401888379538137,\n",
              "  'support': 20834},\n",
              " 'weighted avg': {'f1-score': 0.9225941217615781,\n",
              "  'precision': 0.9317780924797949,\n",
              "  'recall': 0.9201785542862628,\n",
              "  'support': 20834}}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6nnJCLpsy09C"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}