Semi_Supervised_ECG_Classification,_Group_15.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UWInOgGR2Ozj"
      },
      "source": [
        "# Imports and Installations"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XGX6ANSOJxwv",
        "outputId": "10ac4ab7-bb26-4dba-c2a3-54a219015ffc"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "baiAJeOtOy_i",
        "outputId": "537a33aa-b1ad-484e-b476-daf3e37cae42"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Collecting transformers\n",
            "  Downloading transformers-4.18.0-py3-none-any.whl (4.0 MB)\n",
            "\u001b[K     |████████████████████████████████| 4.0 MB 8.4 MB/s \n",
            "\u001b[?25hRequirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from transformers) (2.23.0)\n",
            "Collecting pyyaml>=5.1\n",
            "  Downloading PyYAML-6.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (596 kB)\n",
            "\u001b[K     |████████████████████████████████| 596 kB 55.2 MB/s \n",
            "\u001b[?25hCollecting huggingface-hub<1.0,>=0.1.0\n",
            "  Downloading huggingface_hub-0.5.1-py3-none-any.whl (77 kB)\n",
            "\u001b[K     |████████████████████████████████| 77 kB 7.9 MB/s \n",
            "\u001b[?25hRequirement already satisfied: importlib-metadata in /usr/local/lib/python3.7/dist-packages (from transformers) (4.11.3)\n",
            "Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.7/dist-packages (from transformers) (4.64.0)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.7/dist-packages (from transformers) (21.3)\n",
            "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (2019.12.20)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.7/dist-packages (from transformers) (3.6.0)\n",
            "Collecting sacremoses\n",
            "  Downloading sacremoses-0.0.53.tar.gz (880 kB)\n",
            "\u001b[K     |████████████████████████████████| 880 kB 63.9 MB/s \n",
            "\u001b[?25hCollecting tokenizers!=0.11.3,<0.13,>=0.11.1\n",
            "  Downloading tokenizers-0.12.1-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (6.6 MB)\n",
            "\u001b[K     |████████████████████████████████| 6.6 MB 46.7 MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (1.21.6)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.7/dist-packages (from huggingface-hub<1.0,>=0.1.0->transformers) (4.2.0)\n",
            "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging>=20.0->transformers) (3.0.8)\n",
            "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.7/dist-packages (from importlib-metadata->transformers) (3.8.0)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (3.0.4)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2.10)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (1.24.3)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2021.10.8)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (1.15.0)\n",
            "Requirement already satisfied: click in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (7.1.2)\n",
            "Requirement already satisfied: joblib in /usr/local/lib/python3.7/dist-packages (from sacremoses->transformers) (1.1.0)\n",
            "Building wheels for collected packages: sacremoses\n",
            "  Building wheel for sacremoses (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for sacremoses: filename=sacremoses-0.0.53-py3-none-any.whl size=895260 sha256=bfe27fbed2fd896ec078d8e4cb89c4b3a0a816ffa0b57ae3851e8770f2c2f378\n",
            "  Stored in directory: /root/.cache/pip/wheels/87/39/dd/a83eeef36d0bf98e7a4d1933a4ad2d660295a40613079bafc9\n",
            "Successfully built sacremoses\n",
            "Installing collected packages: pyyaml, tokenizers, sacremoses, huggingface-hub, transformers\n",
            "  Attempting uninstall: pyyaml\n",
            "    Found existing installation: PyYAML 3.13\n",
            "    Uninstalling PyYAML-3.13:\n",
            "      Successfully uninstalled PyYAML-3.13\n",
            "Successfully installed huggingface-hub-0.5.1 pyyaml-6.0 sacremoses-0.0.53 tokenizers-0.12.1 transformers-4.18.0\n",
            "Collecting pyts\n",
            "  Downloading pyts-0.12.0-py3-none-any.whl (2.5 MB)\n",
            "\u001b[K     |████████████████████████████████| 2.5 MB 7.8 MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.17.5 in /usr/local/lib/python3.7/dist-packages (from pyts) (1.21.6)\n",
            "Requirement already satisfied: scipy>=1.3.0 in /usr/local/lib/python3.7/dist-packages (from pyts) (1.4.1)\n",
            "Requirement already satisfied: joblib>=0.12 in /usr/local/lib/python3.7/dist-packages (from pyts) (1.1.0)\n",
            "Requirement already satisfied: numba>=0.48.0 in /usr/local/lib/python3.7/dist-packages (from pyts) (0.51.2)\n",
            "Requirement already satisfied: scikit-learn>=0.22.1 in /usr/local/lib/python3.7/dist-packages (from pyts) (1.0.2)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.7/dist-packages (from numba>=0.48.0->pyts) (57.4.0)\n",
            "Requirement already satisfied: llvmlite<0.35,>=0.34.0.dev0 in /usr/local/lib/python3.7/dist-packages (from numba>=0.48.0->pyts) (0.34.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.22.1->pyts) (3.1.0)\n",
            "Installing collected packages: pyts\n",
            "Successfully installed pyts-0.12.0\n"
          ]
        }
      ],
      "source": [
        "!pip install transformers\n",
        "!pip install pyts"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "tqZhGLhTRguv"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from sklearn.linear_model import LinearRegression\n",
        "plt.style.use(\"ggplot\")\n",
        "\n",
        "from sklearn.cluster import KMeans, AgglomerativeClustering, SpectralClustering\n",
        "\n",
        "from pyts.transformation import ShapeletTransform\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "\n",
        "import requests\n",
        "import tensorflow as tf"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "vGu0YHnmRJPc"
      },
      "outputs": [],
      "source": [
        "#train_path = \"/content/drive/MyDrive/AI Year 3/Coursework/mitbih_train.csv\"\n",
        "#test_path = \"/content/drive/MyDrive/AI Year 3/Coursework/mitbih_test.csv\"\n",
        "\n",
        "train_path = \"/content/drive/MyDrive/Coursework/mitbih_train.csv\"\n",
        "test_path = \"/content/drive/MyDrive/Coursework/mitbih_test.csv\"\n",
        "\n",
        "train_df = pd.read_csv(train_path, header = None)\n",
        "test_df = pd.read_csv(test_path, header = None)\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SzBzLE57XJE1"
      },
      "source": [
        "# Data Augmentation and Resampling"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "id": "mwSOpxKcQGvi",
        "outputId": "2b751b21-a0ae-4148-93f3-5495de17cec0"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1440x720 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# representation of classes % wise\n",
        "per_class = train_df[187].value_counts()\n",
        "plt.figure(figsize=(20,10))\n",
        "my_circle=plt.Circle( (0,0), 0.7, color='white')\n",
        "plt.pie(per_class, labels=['normal beat','unknown Beats','Ventricular ectopic beats','Supraventricular ectopic beats','Fusion Beats'], colors=['tab:blue','tab:orange','tab:purple','tab:olive','tab:green'],autopct='%1.1f%%')\n",
        "p=plt.gcf()\n",
        "p.gca().add_artist(my_circle)\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZETEfWF6bUo1"
      },
      "source": [
        "### DATA AUGMENTATION"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "i1F2PmWVbT59"
      },
      "outputs": [],
      "source": [
        "from sklearn.utils import resample\n",
        "df_1=train_df[train_df[187]==1]\n",
        "df_2=train_df[train_df[187]==2]\n",
        "df_3=train_df[train_df[187]==3]\n",
        "df_4=train_df[train_df[187]==4]\n",
        "df_0=(train_df[train_df[187]==0]).sample(n=20000,random_state=42)\n",
        "\n",
        "df_1_upsample=resample(df_1,replace=True,n_samples=20000,random_state=123)\n",
        "df_2_upsample=resample(df_2,replace=True,n_samples=20000,random_state=124)\n",
        "df_3_upsample=resample(df_3,replace=True,n_samples=20000,random_state=125)\n",
        "df_4_upsample=resample(df_4,replace=True,n_samples=20000,random_state=126)\n",
        "\n",
        "train_df=pd.concat([df_0,df_1_upsample,df_2_upsample,df_3_upsample,df_4_upsample])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "id": "YZwqla2qbevN",
        "outputId": "be9a2ea9-0b15-46d2-fdfe-5666b7fa1acd"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1440x720 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# representation of classes % wise\n",
        "per_class = train_df[187].value_counts()\n",
        "plt.figure(figsize=(20,10))\n",
        "my_circle=plt.Circle( (0,0), 0.7, color='white')\n",
        "plt.pie(per_class, labels=['normal beat','unknown Beats','Ventricular ectopic beats','Supraventricular ectopic beats','Fusion Beats'], colors=['tab:blue','tab:orange','tab:purple','tab:olive','tab:green'],autopct='%1.1f%%')\n",
        "p=plt.gcf()\n",
        "p.gca().add_artist(my_circle)\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "aSRzH_UybnKR"
      },
      "outputs": [],
      "source": [
        "classes=train_df.groupby(187,group_keys=False).apply(lambda train_df : train_df.sample(1))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "Ys2e0Q66bn7v",
        "outputId": "1064880a-12ff-4a69-fe5d-63bdf7afc326"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-d4f127f5-1efe-4903-aa91-d109a5a4c20f\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>...</th>\n",
              "      <th>178</th>\n",
              "      <th>179</th>\n",
              "      <th>180</th>\n",
              "      <th>181</th>\n",
              "      <th>182</th>\n",
              "      <th>183</th>\n",
              "      <th>184</th>\n",
              "      <th>185</th>\n",
              "      <th>186</th>\n",
              "      <th>187</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>34127</th>\n",
              "      <td>0.982249</td>\n",
              "      <td>0.766272</td>\n",
              "      <td>0.068047</td>\n",
              "      <td>0.050296</td>\n",
              "      <td>0.103550</td>\n",
              "      <td>0.079882</td>\n",
              "      <td>0.091716</td>\n",
              "      <td>0.082840</td>\n",
              "      <td>0.085799</td>\n",
              "      <td>0.071006</td>\n",
              "      <td>...</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>72782</th>\n",
              "      <td>0.956349</td>\n",
              "      <td>0.853175</td>\n",
              "      <td>0.218254</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.079365</td>\n",
              "      <td>0.119048</td>\n",
              "      <td>0.099206</td>\n",
              "      <td>0.138889</td>\n",
              "      <td>0.158730</td>\n",
              "      <td>0.134921</td>\n",
              "      <td>...</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75315</th>\n",
              "      <td>0.822547</td>\n",
              "      <td>0.891441</td>\n",
              "      <td>0.843424</td>\n",
              "      <td>0.805845</td>\n",
              "      <td>0.757829</td>\n",
              "      <td>0.743215</td>\n",
              "      <td>0.774530</td>\n",
              "      <td>0.686848</td>\n",
              "      <td>0.440501</td>\n",
              "      <td>0.233820</td>\n",
              "      <td>...</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>80713</th>\n",
              "      <td>0.613757</td>\n",
              "      <td>0.619048</td>\n",
              "      <td>0.706349</td>\n",
              "      <td>0.777778</td>\n",
              "      <td>0.817460</td>\n",
              "      <td>0.825397</td>\n",
              "      <td>0.822751</td>\n",
              "      <td>0.828042</td>\n",
              "      <td>0.846561</td>\n",
              "      <td>0.857143</td>\n",
              "      <td>...</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>85958</th>\n",
              "      <td>0.680441</td>\n",
              "      <td>0.460055</td>\n",
              "      <td>0.435262</td>\n",
              "      <td>0.399449</td>\n",
              "      <td>0.371901</td>\n",
              "      <td>0.319559</td>\n",
              "      <td>0.297521</td>\n",
              "      <td>0.245179</td>\n",
              "      <td>0.201102</td>\n",
              "      <td>0.093664</td>\n",
              "      <td>...</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>4.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 188 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d4f127f5-1efe-4903-aa91-d109a5a4c20f')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-d4f127f5-1efe-4903-aa91-d109a5a4c20f button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-d4f127f5-1efe-4903-aa91-d109a5a4c20f');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "            0         1         2         3         4         5         6    \\\n",
              "34127  0.982249  0.766272  0.068047  0.050296  0.103550  0.079882  0.091716   \n",
              "72782  0.956349  0.853175  0.218254  0.000000  0.079365  0.119048  0.099206   \n",
              "75315  0.822547  0.891441  0.843424  0.805845  0.757829  0.743215  0.774530   \n",
              "80713  0.613757  0.619048  0.706349  0.777778  0.817460  0.825397  0.822751   \n",
              "85958  0.680441  0.460055  0.435262  0.399449  0.371901  0.319559  0.297521   \n",
              "\n",
              "            7         8         9    ...  178  179  180  181  182  183  184  \\\n",
              "34127  0.082840  0.085799  0.071006  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
              "72782  0.138889  0.158730  0.134921  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
              "75315  0.686848  0.440501  0.233820  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
              "80713  0.828042  0.846561  0.857143  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
              "85958  0.245179  0.201102  0.093664  ...  0.0  0.0  0.0  0.0  0.0  0.0  0.0   \n",
              "\n",
              "       185  186  187  \n",
              "34127  0.0  0.0  0.0  \n",
              "72782  0.0  0.0  1.0  \n",
              "75315  0.0  0.0  2.0  \n",
              "80713  0.0  0.0  3.0  \n",
              "85958  0.0  0.0  4.0  \n",
              "\n",
              "[5 rows x 188 columns]"
            ]
          },
          "execution_count": 9,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "classes"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "id": "nQ5VTXIqbr_q",
        "outputId": "3bfb83ae-0a71-4d85-fcb3-44fb0463e7fd"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1152x576 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# plotting classes ECG\n",
        "plt.figure(figsize=(16,8))\n",
        "# normal\n",
        "plt.subplot(2, 3, 1)  \n",
        "plt.plot(classes.iloc[0,:186])\n",
        "plt.title('normal beat')\n",
        "# unknown\n",
        "plt.subplot(2, 3, 2)  \n",
        "plt.plot(classes.iloc[1,:186])\n",
        "plt.title('unknown Beat')\n",
        "# veb\n",
        "plt.subplot(2, 3, 3)  \n",
        "plt.plot(classes.iloc[2,:186])\n",
        "plt.title('Ventricular ectopic beat')\n",
        "# seb\n",
        "plt.subplot(2, 3, 4)  \n",
        "plt.plot(classes.iloc[3,:186])\n",
        "plt.title('Supraventricular ectopic beat')\n",
        "# fusion\n",
        "plt.subplot(2, 3, 5)  \n",
        "plt.plot(classes.iloc[4,:186])\n",
        "plt.title('Fusion Beat')\n",
        "# show plot\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "yqEj7v_vbx3M"
      },
      "outputs": [],
      "source": [
        "def plot_hist(class_number,ma,mi):\n",
        "    # get the class data\n",
        "    img=train_df.loc[train_df[187]==class_number].values\n",
        "    # crop it\n",
        "    img=img[:,mi:ma]\n",
        "    # creates a continous steam of data\n",
        "    img_flatten=img.flatten()\n",
        "    final1=list(range(mi,ma))*(img.shape[0])\n",
        "    print(img.shape[0])\n",
        "    print(len(final1))\n",
        "    print(len(img_flatten))\n",
        "    plt.hist2d(final1,img_flatten, bins=(ma-mi,ma-mi),cmap=plt.cm.jet)\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 109
        },
        "id": "3tzy4qGwb1cX",
        "outputId": "8904a947-48cc-4a21-afe1-7c1386bc4576"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "20000\n",
            "1300000\n",
            "1300000\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x288 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(12,4))\n",
        "plt.subplot(1, 2, 1)  \n",
        "plt.plot(classes.iloc[0,:186])\n",
        "plt.title('Normal beat')\n",
        "plt.subplot(1, 2, 2)  \n",
        "plot_hist(0,70,5)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BxS0VqQE3z1w"
      },
      "source": [
        "### Adding Guassian Noise"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "id": "LNH5DCa5b7yn"
      },
      "outputs": [],
      "source": [
        "def add_gaussian_noise(signal):\n",
        "    noise=np.random.normal(0,0.05,186)\n",
        "    return (signal+noise)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "id": "W3io2fENb8Tb",
        "outputId": "9d8767d1-b43e-4ac3-fa85-f43da5b63efb"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "tempo=classes.iloc[0,:186]\n",
        "#print(tempo)\n",
        "bruiter=add_gaussian_noise(tempo)\n",
        "\n",
        "plt.subplot(2,1,1)\n",
        "plt.plot(classes.iloc[0,:186])\n",
        "\n",
        "plt.subplot(2,1,2)\n",
        "plt.plot(bruiter)\n",
        "\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "IfslKAOqcM_r"
      },
      "outputs": [],
      "source": [
        "target_train=train_df[187]\n",
        "target_test=test_df[187]\n",
        "y_train=tf.keras.utils.to_categorical(target_train)\n",
        "y_test=tf.keras.utils.to_categorical(target_test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "lbGOVfzKczfh"
      },
      "outputs": [],
      "source": [
        "X_train=train_df.iloc[:,:186].values\n",
        "X_test=test_df.iloc[:,:186].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)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WUNmBWfnc7vF",
        "outputId": "53013547-c32c-4d3f-c3f5-388a7a65a03d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(100000, 186, 1)\n",
            "(100000, 5)\n"
          ]
        }
      ],
      "source": [
        "print(X_train.shape)\n",
        "print(y_train.shape)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xT_LaiptSs6k"
      },
      "source": [
        "# Clustering of Data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "H2W8O_nbtDKm"
      },
      "source": [
        "## Label Processing for Clustering"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Nmw7FLJ0XA0O",
        "outputId": "bd2c0f51-4654-48a9-fe5f-6c28afb6e126"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(100000, 186, 1)"
            ]
          },
          "execution_count": 18,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "X_train.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "v1AimDifKYKo"
      },
      "outputs": [],
      "source": [
        "randomRows = np.random.randint(len(X_train), size=int(len(X_train)/10))\n",
        "newarr=[]\n",
        "for i in randomRows:\n",
        "    newarr.append(X_train[i])\n",
        "\n",
        "clust_xtrain = np.asarray(newarr)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vqDt-iJBYPbm",
        "outputId": "af481cf4-4cac-4064-aeb2-ad7c4ae67bb7"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(10000, 186, 1)"
            ]
          },
          "execution_count": 20,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "clust_xtrain.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fFnDF7X6KLfW"
      },
      "outputs": [],
      "source": [
        "#shapelet = ShapeletTransform(window_sizes=np.arange(10, 186, 1), random_state=42)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "id": "eVYfRsgILBpr"
      },
      "outputs": [],
      "source": [
        "nsamples, nx, ny = clust_xtrain.shape\n",
        "clust_xtrain2 = clust_xtrain.reshape((nsamples,nx*ny))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wBfgyj_LGvae",
        "outputId": "37ffd602-47bc-42cc-d53e-0e5f3c5bcaa0"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(10000, 186)"
            ]
          },
          "execution_count": 22,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "clust_xtrain2.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uzUg5ztJLhHc",
        "outputId": "1e40f454-e736-4702-a913-9ed2c37b7d35"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(10000, 186, 1)\n"
          ]
        }
      ],
      "source": [
        "print(clust_xtrain.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "4HN9DhmYMNWp"
      },
      "outputs": [],
      "source": [
        "#nsamples, nx, ny = X_test.shape\n",
        "#d2_X_test_dataset = X_test.reshape((nsamples,nx*ny))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ckyWNkD1Gk6L"
      },
      "source": [
        "## Deciding Number of Clusters using Silhouette Score and Davies-Bouldin Index\n",
        "\n",
        "Experiments: Do Not Need to Run"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-yjkRX50eBX3"
      },
      "source": [
        "### K-Means\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 55,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 336
        },
        "id": "WFDTk7N0Kvlq",
        "outputId": "7c4f640a-8be0-4592-b1f7-a4b902fd5e18"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "range(6, 21)\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 15/15 [01:34<00:00,  6.33s/it]\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "from sklearn.metrics import silhouette_score\n",
        "\n",
        "silhouette_avg = []\n",
        "\n",
        "K = range(6,21)\n",
        "\n",
        "print(K)\n",
        "for num_clusters in tqdm(K):\n",
        "  kmeans = KMeans(n_clusters=num_clusters)\n",
        "  kmeans.fit(clust_xtrain2)\n",
        "  cluster_labels = kmeans.labels_\n",
        "  #silhouette_avg.append(silhouette_score(clust_xtrain2, cluster_labels))\n",
        "  score = silhouette_score(clust_xtrain2, cluster_labels)\n",
        "  plt.plot(num_clusters,score,'bx-')\n",
        "\n",
        "plt.xlabel(\"Values of K\") \n",
        "plt.ylabel(\"Silhouette score\") \n",
        "plt.title(\"Silhouette analysis For Number of Cluster with K Means\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 54,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 336
        },
        "id": "VWD-ssV6TMrz",
        "outputId": "d6bf2de3-e395-462c-da21-4285221ba50b"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "range(6, 21)\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 15/15 [01:05<00:00,  4.38s/it]\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "from sklearn.metrics import davies_bouldin_score\n",
        "\n",
        "K = range(6,21)\n",
        "\n",
        "print(K)\n",
        "for num_clusters in tqdm(K):\n",
        "  kmeans = KMeans(n_clusters=num_clusters)\n",
        "  kmeans.fit(clust_xtrain2)\n",
        "  cluster_labels = kmeans.labels_\n",
        "  #silhouette_avg.append(silhouette_score(clust_xtrain2, cluster_labels))\n",
        "  score = davies_bouldin_score(clust_xtrain2, cluster_labels)\n",
        "  plt.plot(num_clusters,score,'bx-')\n",
        "\n",
        "plt.xlabel(\"Number of Clusters\") \n",
        "plt.ylabel(\"Davies Boulding Score\") \n",
        "plt.title(\"Davies Boulding Score For Number of Cluster with K Means\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hCqs69YUeIzW"
      },
      "source": [
        "### Ward Hierachical"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 66,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 336
        },
        "id": "8k5psQjiPw5i",
        "outputId": "fae41337-09aa-4e92-89b3-64c4280dd8de"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "range(6, 25)\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 19/19 [03:21<00:00, 10.60s/it]\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "from sklearn.metrics import silhouette_score\n",
        "\n",
        "\n",
        "K = range(6,25)\n",
        "\n",
        "for num_clusters in tqdm(K):\n",
        "  ward = AgglomerativeClustering(n_clusters=num_clusters)\n",
        "  ward.fit(clust_xtrain2)\n",
        "  cluster_labels = ward.labels_\n",
        "  #silhouette_avg.append(silhouette_score(clust_xtrain2, cluster_labels))\n",
        "  sil_score = silhouette_score(clust_xtrain2, cluster_labels)\n",
        "  #db_score = davies_bouldin_score(clust_xtrain2, cluster_labels)\n",
        "  plt.plot(num_clusters,sil_score,'bx-')\n",
        "  #axs[1].plot(num_clusters,db_score,'bx-')\n",
        "\n",
        "plt.xlabel(\"Number of Clusters\") \n",
        "plt.ylabel(\"Silhouette score\") \n",
        "plt.title(\"Silhouette analysis For Ward Number of Clusters\")\n",
        "plt.show()\n",
        "\n",
        "#Davies-Bouldin\n",
        "#axs[1].xlabel(\"Number of Clusters\") \n",
        "#axs[1].ylabel(\"Davies-Bouldin score\") \n",
        "#axs[1].title(\"Davies-Bouldin analysis For Ward Number of Clusters\")\n",
        "#axs[1].show()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 67,
      "metadata": {
        "id": "b_RksTrGTFFK"
      },
      "outputs": [],
      "source": [
        "from sklearn.metrics import davies_bouldin_score"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 68,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 317
        },
        "id": "8vYkx06BTI-D",
        "outputId": "3f2c9d70-f238-4ec0-f7c9-95391654a7e7"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 19/19 [02:55<00:00,  9.24s/it]\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "K = range(6,25)\n",
        "\n",
        "for num_clusters in tqdm(K):\n",
        "  ward = AgglomerativeClustering(n_clusters=num_clusters)\n",
        "  ward.fit(clust_xtrain2)\n",
        "  cluster_labels = ward.labels_\n",
        "  db_score = davies_bouldin_score(clust_xtrain2, cluster_labels)\n",
        "  #plt.plot(num_clusters,sil_score,'bx-')\n",
        "  plt.plot(num_clusters,db_score,'bx-')\n",
        "\n",
        "#plt.xlabel(\"Number of Clusters\") \n",
        "#plt.ylabel(\"Silhouette score\") \n",
        "#plt.title(\"Silhouette analysis For Ward Number of Clusters\")\n",
        "#plt.show()\n",
        "\n",
        "#Davies-Bouldin\n",
        "plt.xlabel(\"Number of Clusters\") \n",
        "plt.ylabel(\"Davies-Bouldin score\") \n",
        "plt.title(\"Davies-Bouldin analysis For Ward Number of Clusters\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "valStK2weNS2"
      },
      "source": [
        "### Spectral"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 71,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 317
        },
        "id": "l7yd2yjbeTX-",
        "outputId": "f12ce0e2-cb04-4af0-b0eb-840c8eeb4f28"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 19/19 [16:02<00:00, 50.66s/it]\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "from sklearn.metrics import silhouette_score\n",
        "\n",
        "K = range(6,25)\n",
        "\n",
        "for num_clusters in tqdm(K):\n",
        "  spectral = SpectralClustering(n_clusters=num_clusters)\n",
        "  spectral.fit(clust_xtrain2)\n",
        "  cluster_labels = spectral.labels_\n",
        "  #silhouette_avg.append(silhouette_score(clust_xtrain2, cluster_labels))\n",
        "  sil_score = silhouette_score(clust_xtrain2, cluster_labels)\n",
        "  #db_score = davies_bouldin_score(clust_xtrain2, cluster_labels)\n",
        "  plt.plot(num_clusters,sil_score,'bx-')\n",
        "  #axs[1].plot(num_clusters,db_score,'bx-')\n",
        "\n",
        "plt.xlabel(\"Number of Clusters\") \n",
        "plt.ylabel(\"Silhouette score\") \n",
        "plt.title(\"Silhouette analysis For Spectral Number of Clusters\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 129,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 317
        },
        "id": "zdqBij6teTnT",
        "outputId": "8cbb429b-8c6c-4b6e-c446-b3da7c512189"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 19/19 [11:58<00:00, 37.80s/it]\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaMAAAEaCAYAAAC8UDhJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deVRT1/o38G8SDAEDSAiIKCCKoDigoliBKipSrq1WrUXrgLQ/b6VqFV+1otLS61CplooDONUB0FrUOvW23lacRa0DggMOqFWRQWSQoQgI2e8f3pxLhGDAhCTwfNZyLXLGb05inpyzd87mMcYYCCGEEC3iazsAIYQQQsWIEEKI1lExIoQQonVUjAghhGgdFSNCCCFaR8WIEEKI1jWbYrR9+3YYGBhoO4ZaqfKcHjx4AB6PhzNnztT6WB8EBgbCx8dHbdtriu8FTTpx4gR4PB4eP36s7Sh10tX3dlFREUaNGgUzMzPweDw8ePCgwdvS1eeoDlorRoGBgeDxeODxeGjRogWkUim8vLywYsUK/P3332rf39ixY5GRkaH27b5K/px4PB4EAgHatGmDiRMnIjMzU+P7VoWtrS2ysrLQr18/bUfRGnW8F6q/zvJ/IpFITQn/x87ODvPnz1eYtn//fvB4PKXTHz16pPYcryP///zFF18oTH/8+DF4PB5OnDjR6Jl0xfr163Hu3DmcOXMGWVlZsLW1Vbrsb7/9hnfeeQcWFhYwMjKCs7MzgoKCcOfOHY3l27FjB3g8nsa2ryqtnhm9/fbbyMrKwsOHD3H8+HFMmDAB69atQ+/evfHkyRO17svIyAitW7dW6zaVWbduHbKysvDo0SPEx8cjJSUFkyZNapR9v45AIIC1tTVatGih7Shao673gvx1lv97+PBhg7dVUVFR6/QhQ4bg2LFjCtOOHj0KOzu7Wqd36tQJdnZ2as2gKpFIhDVr1rzRcdBVL168aPC6aWlp6Nq1K7p37w5ra2sIBIJal1u8eDGGDx8OR0dH7Nu3D7du3cLWrVshFAoRGhra4P03pjd6DzEtmTx5MhsyZEiN6Y8fP2bm5uYsMDCQm/bHH3+wgQMHMnNzc2ZqasoGDBjA/vzzT27++PHj2dChQ2tsy8/Pj02YMIExxti2bduYQCBQmH/p0iU2dOhQ1rJlSyaVStmoUaPYgwcPuPnp6els9OjRzMLCghkaGjIHBwe2YsWKOp8XABYXF6cwbc2aNczU1FRh2q+//sp69+7NhEIhs7S0ZJ999hkrKSmp8/jExcWx6i9Zbc8pPj6edezYkRkaGrL+/fuzgwcPMgDs9OnTjDHG/vrrr1ofx8fHs3fffZcZGRkxBwcHtm3btjqfZ35+PpswYQKztbVlIpGIOTk5se+++47JZLIaz2Hjxo3Mzs6OmZiYsOHDh7Ps7Gxumfv377NRo0axNm3aMCMjI9atWzcWGxursK/qx+L48eOMz+ezR48eKSwTExPDTE1NuWO4bNky5uDgwIRCIZNKpczX15eVlpbWetwKCwtZYGAga926NRMKhaxdu3Zs9uzZdT7/2l7n6rZv3866dOnCWrRowdq2bcsWLVrEXrx4wc0fOHAg++STT1hoaCiztrZmrVu3rnU7cXFxjM/ns/z8fG5a586d2aZNm5iBgUGN6Z999hljjLGdO3cyd3d3ZmpqyiwsLNiwYcPY7du3uWXlr/uOHTvYP/7xD2ZsbMy++OILxtjL92vbtm2ZkZER8/X1ZTExMQwAS09PV/p8J0+ezAYPHszc3d3Z+PHjuenp6ekMADt+/LjCfuXvP7mOHTuysLAw7jEAtmbNGubv78+MjY2Zra0t27NnD3v27BkbP348E4vFzMHBge3du7fGc4qLi2ODBw9mIpGIOTg4sF27dinsKzs7m02ePJlJpVImFouZh4cHO3nyJDf/+PHjDAD797//zTw9PZmhoSGLjo6u9XlXVFSw+fPnMxsbG9aiRQvWpUsXtnPnTm6+vb09A8D9GzhwYK3buXTpEgPAli9fXut8+eus7P/v647n5s2bWefOnZmhoSEzNzdnb7/9NktPT+eea/V/kydP5tZbs2YNc3Z2ZoaGhszR0ZEtXbpU4X1sb2/PFi1axD777DMmkUiYu7t7nfuri84VI8YYmzFjBjM1NWVVVVWMMcb27dvH4uPj2a1bt9j169fZ//3f/zFzc3OWm5vLGGPs999/Z3w+n2VkZHDbyMzMZAKBgP3++++MsZofQDdu3GAtW7ZkX331Fbt58ya7evUqGzNmDOvUqRN7/vw5Y4yx4cOHsyFDhrArV66wv/76ix07doz9+OOPdT6vVz+kMjMzmbe3N/Pz8+OmpaSkMIFAwIKDg9nNmzfZb7/9xmxtbdnEiRPrPD6vK0ZJSUmMz+ezkJAQduvWLfbzzz+z9u3bq1SMHBwcWHx8PEtLS2MLFixgAoFA4cPrVVlZWWz58uXs8uXL7P79+ywuLo61bNmSbd26VeE5mJqasnHjxrFr166xs2fPsvbt2ys8z6tXr7K1a9ey5ORkdvfuXbZmzRomEAjYsWPHlB4LZ2dn9vXXXyvk8fLyYkFBQYwxxn7++WdmYmLCDh06xB4+fMiuXLnCVq1apbQYff7556xHjx7s/Pnz7OHDhywxMZFt2rRJ6XNnrO5i9O9//5vx+Xz2zTffsNu3b7OffvqJtWrVioWGhnLLDBw4kInFYjZ16lR248YNdvXq1Vq3lZGRwQCwffv2cY/lxal79+4K0wFwH85bt25lhw4dYnfv3mVJSUls+PDhzNHRkZWXlzPG/ve6t23blu3YsYPdv3+f3b9/nx04cIAJBAIWERHBbt++zX744QdmZWWlUjEaMmQIO3XqFOPxeOzixYuMsTcrRq1bt2bbt29naWlp7LPPPmMikYj5+fmxbdu2sbS0NDZjxgxmbGzMfQ7It92mTRu2Y8cOduvWLbZo0SLG5/NZUlISY4yx0tJS1qVLFzZ69Gh28eJFlpaWxpYuXcqEQiFLTU1ljP2vGDk7O7NDhw6x+/fvK33uc+fOZRKJhO3evZvdvn2bLVu2jPF4PJaQkMAYYywnJ4f5+/uzt99+m2VlZbG8vLxatzNr1ixmbGzMvT7KNKQYXbp0iQkEAhYTE8MePHjArl69yjZv3szS09NZeXk5W7duHQPAsrKyWFZWFnv27BljjLGwsDBmZ2fH9u3bx+7fv89+/fVXZmtrq/A+tre3ZyYmJiwsLIzdvn2b3bhxo8791UUni9H69esZAPbkyZNa51dVVbFWrVqxHTt2cI9tbGwUzlpWrlzJ2rZtyxW0Vz+AJk+ezMaOHauw3bKyMmZkZMT279/PGGOsR48eCv9BVAGAGRoaspYtWzIjIyMGgPXs2VPhbGDixImsb9++CusdOHCA8Xg87sysIcVowoQJzMPDQ2GdtWvXqlSMIiIiuHUqKyuZWCxmGzZsqNdznzlzJvPx8eEeT548mVlaWrKysjJuWnh4OLO2tq5zOyNGjGBTpkxR2E71YxEREcHs7Oy41/bmzZsMAPeB8/3337NOnTqxioqKWrf/6nEbMWKEwrdBVVR/neX/Fi9ezBh7WRg//PBDheUjIyOZSCTiPmwGDhzIOnXqxD2HunTp0oXNmDGDMfbyPdCrVy/G2MvjXX36q2dQ1eXl5TEA7MyZM4yx/73u8sxynp6eCmc2jDE2Z84clYsRY4yNHDmSOwN4k2I0a9Ys7nFOTg4DwD1fxl6eLQBgv/zyi8K2q39YMsZY//79uS9A27ZtY23btlX4ds8YY4MGDeL2Jy9Gr56hv+rvv/9mQqGQRUVFKUwfOXIkGzRoUK3HRpl//OMfrHv37nUuw1jDitG+ffuYqakpKywsrHWbr36uyJ+bkZERO3z4sML0mJgYZmZmxj22t7dngwcPVljmdftTRid707H/3rtV3qj2119/YdKkSXB0dISpqSlMTU1RWFjIXZvm8/mYOHEi4uLiuG3ExcVhwoQJ4PNrf4oXL17E/v37IRaLuX8WFhYoKytDWloaACA4OBjffPMN+vXrh/nz5+PUqVPc+jt37lRYd+fOndy8ZcuWITk5GSkpKUhISIChoSFGjRqFqqoqAMCNGzcwYMAAhTwDBw4EYwypqakNPm6pqanw8PBQmObl5aXSuj179uT+FggEsLKyqrPdTiaTITw8HD179oRUKoVYLMaGDRtqtBd07twZhoaG3GMbGxuF7ZaWliIkJARdu3aFRCKBWCzGb7/9Vme7w+TJk5GTk4Pff/8dAPDDDz/Azc0NvXr1AgD4+/vjxYsXsLe3R2BgIOLi4lBcXKx0e9OmTcPevXvRrVs3zJo1C4cPH4ZMJlO6vJz8dZb/mz59OgDlr29ZWRnu3bvHTXNzc1P6/qzOx8cHR48eBfCyXWjw4MEAgEGDBilM79WrF8zNzQEAycnJGDVqFBwcHGBiYsK1I716XN3d3RUev8l7SO7bb79FYmIiDh06VK/1XuXq6sr9bWlpCYFAgB49enDTzM3NIRQKkZOTo7Be//79FR57enrixo0bAF7+v8/OzkarVq0U/v+ePn2a+38v9+qxedXdu3dRUVFR62st35+q5J95mjB06FB06NABDg4OGDduHDZt2oTc3Nw617lx4waeP3+ODz74QOE4TZ06FYWFhXj69Cm37KvHqSH7AwCd7N9648YNmJmZwcLCAgDw3nvvQSqVIioqCra2thAKhfDy8lJoLAsICMCKFSuQnJwMALh69Sp27dqldB8ymQyTJk1CSEhIjXny/X788cfw8/PDf/7zHxw/fhz/+Mc/MGrUKOzYsQMjRoxQ6JFWvUG8devWcHR0BAB06tQJLVu2RP/+/XH8+HGVuyjz+fwab9A3aUR9HaFQqPCYx+PV+YEcERGB5cuXY9WqVejVqxdMTEywatUq/Prrr6/dbvXnNW/ePBw8eBDff/89nJ2d0bJlS8yZMweFhYVK921hYYExY8Zg8+bNGDJkCGJjY7F06VJuftu2bXHr1i0cP34cx44dw5IlSzB//nz8+eeftfZkeuedd/Do0SP8/vvvOHHiBCZOnIju3bvj6NGjShubAcXXuSFatmyp0nJDhgzB2rVrkZmZiaNHj2LDhg0AXn7o3b59m5s+fvx4AC8LvK+vL7y8vLBt2zbuvdm1a9caDcyqZqgPJycnTJ06FfPnz8fhw4cV5smLryrv7do62bw67XXv01fJZDJ06dIF+/fvrzHP2NhY4bEmjo0yzs7OOHXqFCoqKmr8n6mLKsdTLBbj0qVLSExMREJCAjZs2IAvvvgCR48ehZubW63blR/TPXv2wMnJqcZ8iUTC/f3qcWrI/gAd/J1RRkYGdu7cidGjR4PP5yMvLw+pqakICQnBO++8AxcXF4hEohrfhrp27Qo3NzfExcUhNjYWbm5ucHFxUbqfPn364OrVq+jYsSMcHR0V/sm/XQJAmzZt8PHHHyM2NhZbtmzBzp07UVRUBBMTE4V1TExMlO5L/oH2/PlzLmv1sywAOHnyJHg8Hrp27QoAsLKyqtEdPCkpqc5j5+LigrNnzypMS0xMrHOdhjp16hT8/PzwySefoFevXnB0dKzxzVLV7UyYMAH+/v5wdXVFhw4dVOrGOnXqVPzyyy/YuHEjnj9/jo8++khhvqGhIfz8/LBixQpcu3YNpaWlOHDggNLtSSQSfPTRR9i4cSN+/fVXnDx5ssFnqcpeXyMjI3Ts2LHe2/P29oZAIMDGjRuRlZXFfRM3NzeHq6srNm7ciPT0dO6Lzs2bN/H06VMsW7YM3t7e6NKlCwoKClT69q2u91BYWBgyMzOxadMmhemWlpYAoPDezsnJUevPLs6fP6/w+OzZs9xnQZ8+fXD//n2YmprW+H9vY2NTr/04OjrC0NCw1te6W7du9drWxIkTUVpaiu+//77W+QUFBbVOV/V4CgQCDBgwAIsXL8bly5fRpk0b/PjjjwD+94VRfuUGePkeFolEuH//fo3j5OjoWOeXtNftTxmtnhlVVFQgOzsbMpkMeXl5OHPmDJYvXw4rKyssX74cwMv/cJaWlti8eTM6duyIvLw8fPHFFzAyMqqxvYCAAG69hQsX1rnvhQsXwt3dHRMnTsSsWbNgaWmJBw8e4MCBA5g1axY6dOiAGTNmYNiwYXB2dkZZWRn27dsHW1vbOgsPABQWFiI7OxuMMTx69Ajz58+HlZUVPD09Abw8G+jduzdmz56NqVOn4sGDB/j8888xYcIE7nKKj48Pvv32W0RFRcHPzw/Hjh3D7t2769zv7Nmz0bdvXyxatAiTJ0/GjRs3EBERUec6DeXs7Iy4uDgcP34cbdu2RWxsLP7880+FQq7qdg4ePMhdDvj++++RmZn52q7XXl5ecHZ2xty5cxEQEKDwmmzZsgUymQzu7u5o1aoVjh49iuLiYqVfThYtWgQ3Nzd07doVfD6fuwTb0C7SCxYswPDhwxEeHo7Ro0cjOTkZX3/9NebMmVOvb71yZmZmcHNzw6pVq9C3b1+IxWJu3qBBg7Bq1SoYGhpy7y97e3sYGhpi7dq1mDNnDh48eICQkBCVfksyZ84cfPjhh3B3d8ewYcNw5swZhcvfqrK0tERISAiWLFmiMN3IyAienp5YsWIFOnfujMrKSixatEjhUu6b2rJlCzp37ow+ffpgx44dOHfuHNauXQsAmDBhAlatWoV3330Xy5Ytg5OTE548eYJjx46hS5cuGDlypMr7MTY2xsyZM/Hll1/C0tISrq6u2Lt3Lw4ePIgjR47UK3OfPn3w1VdfYdGiRUhPT8fYsWNhb2+PzMxM7N69GxkZGbX+/1fleB48eBD379/HgAEDYGlpicuXLyM9PZ37/+Dg4AAAOHToELy8vGBkZASxWIyFCxdi4cKF4PF48PHxQWVlJa5du4YrV67g22+/VfpcXrc/perVwqRGkydP5roSCgQCJpFImKenJ/v2228VujgzxtiJEydYjx49mKGhIXNycmJ79+6t0eDJGGNPnz5lLVq0YC1atGBPnz5VmFdbN+irV6+yESNGsFatWjGRSMQ6duzI/vnPf3I9XqZNm8Y6derERCIRk0gkbNiwYez69et1Pi/5cwLAeDwes7KyYsOHD2cpKSkKy1Xv2i2VSllQUFCN57106VJmY2PDWrZsycaNG8f1eqnrOe3atYt16NCBCYVC5u7uzg4cOKBSB4bXNSi/6tmzZ+zDDz9kJiYmTCKRsGnTprHQ0FBmb2/PLaNKJ4xHjx4xX19fZmxszKytrdlXX33FPvnkE4UusMoagCMjIxkAduHCBYXpP//8M+vfvz9r1aoVMzIyYl27dmU//PCD0uO2ePFi1rVrV9ayZUvupwOvHo9XQYWu3Z07d2YtWrRgNjY2bOHChTW6dv/f//1fnfuobsGCBQwAW7RokcL0X375hQGo0Yi8Z88e5ujoyAwNDVnPnj3ZiRMnmEAg4LrsK3vdGXt5XG1sbJhIJGJDhgxh27dvr1cHBrnnz58zW1tbhQ4MjDF2+/ZtNmDAAGZsbMwcHR3Zzz//XGsHhlePb/X8coaGhmzz5s0Kzyk2NpYNHDiQGRoasvbt2yt0tWaMsdzcXBYUFMR1x7axsWEjR47kOsDIOzC8rvcXY6/v2q3s2Chz6NAhNnToUGZubs593n322WcsLS1N4TlWf91edzxPnjzJBg0axKRSKddF+9Uu5LNmzWKWlpY1unZv3ryZubq6MkNDQ9aqVSvm7u6u0M3d3t6eLVmyRGFbquyvNjzGaKRXop+++OILHDlyBFeuXNF2FELIG9LJDgyE1KWwsBB37tzBpk2bsGbNGm3HIYSoAZ0ZEb3j7e2NP//8E+PGjcOWLVtU6h5NCNFtVIwIIYRoHX2lJIQQonVUjAghhGhdk+7AoCtjCNVGKpWqdIsMbdOXnID+ZKWc6qcvWXU9Z31/+KtOdGZECCFE66gYEUII0ToqRoQQQrSOihEhhBCto2JECCFE66gY/Vd0tBiJiYp3VE5MFCI6WqxkDUIIIepCxei/XF0rEBRkzhWkxEQhgoLM4epa8Zo1CSGEvKkm/Tuj+vD0rMCGDQUICjJHQEApYmONsWFDATw9qRgRQoim0ZlRNZ6eFQgIKEVkpAkCAkqpEBFCSCOhYlRNYqIQsbHGCA4uRmyscY02JEIIIZpBxei/5G1EGzYUYN68Yu6SHRUkQgjRPCpG/5WSIlRoI5K3IaWkUDEihBBNow4M/zVtWkmNaZ6eFdRuRAghjYDOjAghhGgdFSNCCCFaR8WIEEKI1lExIoQQonVUjAghhGhdo/Smi46ORlJSEszMzBAREVFj/o0bN7BixQpYWVkBAPr164cxY8YAAJKTk7Ft2zbIZDIMGTIEI0eObIzIhBBCGlGjFCNvb2/4+fkhKipK6TJdunRBSEiIwjSZTIYtW7YgNDQUFhYWWLBgAfr06YN27dppOjIhhJBG1CiX6VxcXCAW138ohrt378La2hqtW7eGgYEBPDw8cPHiRQ0kJIQQok0686PXO3fuYN68eTA3N8ekSZNga2uL/Px8WFhYcMtYWFggLS1N6TYSEhKQkJAAAAgPD4dUKtV47oYyMDDQ6Xxy+pIT0J+slFP99CWrvuTUBp0oRg4ODoiOjoZIJEJSUhJWrlyJNWvW1Hs7Pj4+8PHx4R7n5uaqM6ZaSaVSnc4npy85Af3JSjnVT1+y6npOGxsbre1bJ3rTGRsbQyQSAQB69+6NqqoqFBUVQSKRIC8vj1suLy8PEolEWzEJIYRoiE4Uo2fPnoExBuBlO5FMJoOJiQk6duyIrKws5OTkoLKyEmfPnkWfPn20nJYQQoi6NcplusjISKSmpqK4uBhBQUHw9/dHZWUlAMDX1xfnz5/HH3/8AYFAAKFQiODgYPB4PAgEAnzyySdYtmwZZDIZBg0aBFtb28aITAghpBHxmPyUpAnKzMzUdgSldP3asZy+5AT0JyvlVD99yarrOZt9mxEhhJDmjYoRIYQQraNiRAghROuoGBFCCNE6KkaEvEZ0tBiJiUKFaYmJQkRH1/8WV4SQ2lExIuQ1XF0rEBRkzhWkxEQhgoLM4epaoeVkhDQdOnE7IEJ0madnBTZsKEBQkDkCAkoRG2uMDRsK4OlJxYgQdaEzI0JU4OlZgYCAUkRGmiAgoJQKESFqRsVIjahtoelKTBQiNtYYwcHFiI01rvE6E0LeDBUjNaK2haZJ/jpu2FCAefOKuUt2VJAIUR8qRmpUvW1h5UoT7gOMLunot5QUocLrKH+dU1KoGBGiLtSBQc2qty0EBxdTIWoCpk0rqTHN07OCXltC1IjOjNSM2hYIIaT+qBipEbUtEEJIw1AxUiNqWyCEkIahNiM1orYFQghpGDozIoQQonVUjAghhGgdFSNCCCFap3KbUUZGBs6dO4dnz55hypQpyMjIQGVlJezt7TWZjxBCSDOg0pnRuXPnEBYWhvz8fJw+fRoAUFZWhtjYWI2GI4QQ0jyodGa0e/duhIaGon379jh37hwAwN7eHg8ePNBkNkIIIc2ESmdGhYWFNS7H8Xg88Hg8jYQihBDSvKhUjDp06IBTp04pTEtMTISjo6NGQhFCCGleVLpM9/HHH2Pp0qU4duwYysvLsWzZMmRmZiI0NFTT+QghhDQDry1GjDEYGBggIiICycnJcHNzg4WFBdzc3CASiRojIyGEkCbutcWIx+Nh7ty5iImJgYeHR4N2Eh0djaSkJJiZmSEiIkLpcnfv3kVoaCiCg4Px1ltvAQDGjh0LOzs7AIBUKsX8+fMblIEQQojuUukyXfv27ZGVlYW2bds2aCfe3t7w8/NDVFSU0mVkMhl27twJV1dXhelCoRArV65s0H4JIYToB5WKUdeuXfHNN99g4MCBkEqlCvMGDx782vVdXFyQk5NT5zKHDx9Gv379cO/ePVUiEUIIaUJUKka3b9+GlZUVbt68WWOeKsXodfLz83HhwgWEhYVh/fr1CvNevHiBkJAQCAQCvP/++3B3d1e6nYSEBCQkJAAAwsPDaxROXWJgYKDT+eT0JSegP1kpp/rpS1Z9yakNKhWjsLAwjYbYvn07JkyYAD6/Zk/z6OhoSCQSPHnyBIsXL4adnR2sra1r3Y6Pjw98fHy4x7m5uRrL/KakUmmt+aKjxXB1VRx2IjFRiJQUYa1DVGiaspy6SF+yUk7105esup7TxsZGa/tW+d50JSUluHz5MvLz8yGRSODm5gaxWKyWEPfu3cPq1asBAEVFRbhy5Qr4fD7c3d0hkUgAAK1bt4aLiwsePHigtBg1Ba6uFdxosZ6eFQqjxxJCSFOlUjG6c+cOli9fjrZt20IqlSIpKQnbt2/HggUL4OTk9MYhqndsiIqKgpubG9zd3VFSUgJDQ0O0aNECRUVFuH37Nt5///033p8uk48OGxRkjoCAUsTGGiuMHksIIU2RSsVo+/btmDJlCjw9PblpZ8+exbZt27B8+fLXrh8ZGYnU1FQUFxcjKCgI/v7+qKysBAD4+voqXS8jIwObNm0Cn8+HTCbDyJEj0a5dO1Ui6zVPzwoEBJQiMtIEwcHFVIgIIU2eSsUoKysL/fv3V5j21ltvYfPmzSrtJDg4WOVA06dP5/52dnau83dJTVViohCxscYIDi5GbKwxPDzKqSARQpo0le5NZ21tjbNnzypMO3fuHFq3bq2RUM1Z9TaiefOKuUt2iYlCbUcjhBCNUenMKDAwEOHh4Th8+DCkUimePn2KrKwshISEaDpfs5OSIlRoI5K3IaWkCOnsiBDSZKlUjJydnbF27VokJSWhoKAAbm5u6N27t9p605H/qa37tqdnBRUiQkiTplIxys/Ph1AoxIABA7hpJSUlXDdvQggh5E2o1Ga0cuVK5OfnK0zLz8/Hd999p5FQhBBCmheVilFmZiZ352w5Ozs7ZGRkaCQUIYSQ5kWlYmRqaors7GyFadnZ2TAxMdFIKEIIIc2LSm1GgwYNQkREBMaNG4fWrVsjOzsb8fHxarlJKiGEEKJSMRo5ciQMDAwQFxeHvLw8WFhYYPDgwXjvvfc0nY8QQkgzoFIx4vP5GDFiBEaMGKHpPIQQQpohlYrR9evXYWVlBSsrKzx79gw7duwAn8/H+PHj0apVK01nJIQQ0sSp1LvDkCEAACAASURBVIFhy5Yt3FhDMTExqKqqAo/Hw8aNGzUajhBCSPOg8o9epVIpqqqqkJKSgujoaBgYGGDq1KmazkcIIaQZUKkYGRkZ4dmzZ0hPT0e7du0gEolQWVnJDQNBCCGEvAmVipGfnx8WLFiAyspKBAYGAgBu3bqFtm3bajIbIYSQZkLlrt3u7u7g8/nckN8SiQRBQUEaDUcIIaR5UKkYAYCNjU2djwkhhJCGUqk3HSGEEKJJVIwIIYRoHRUjQgghWqdymxEAFBYWoqysTGFa69at1RqIEEJI86NSMUpOTsb69evx7NmzGvPi4+PVHooQ0rRFR4vh6loBT88KblpiohApKUJMm1aixWREW1QqRlu2bMEHH3wAb29vCIVCTWcihDRxrq4VCAoyx4YNBfD0rEBiopB7TJonlYpRSUkJhg4dCh6Pp+k8hJBmwNOzAhs2FCAoyBwBAaWIjTXmChNpnlTqwDB48GAcP35c01kIIc2Ip2cFAgJKERlpgoCAUipEzZxKZ0ZpaWk4fPgwDh48WGPIiH/9618aCUYIadoSE4WIjTVGcHAxYmON4eFRTgWpGVOpGA0ePPiNhxiPjo5GUlISzMzMEBERoXS5u3fvIjQ0FMHBwXjrrbcAACdOnMC+ffsAAKNHj4a3t/cbZSGEaFf1NiJPzwp4eJQrPCbNj0rFSB0f/t7e3vDz80NUVJTSZWQyGXbu3AlXV1duWklJCfbu3Yvw8HAAQEhICPr06QOxWPzGmQgh2pGSIlQoPPI2pJQUYaMVI+rRp1uUFqNTp05hwIABAIBjx44p3YCqZ0wuLi7Iycmpc5nDhw+jX79+uHfvHjctOTkZPXr04IpPjx49kJycDC8vL5X2SwjRPbV92Ht6VjTqWRH16NMtSotRYmIiV4xOnz6tdANvevlOLj8/HxcuXEBYWBjWr1+vMN3CwoJ7LJFIkJ+fX+s2EhISkJCQAAAIDw+HVCpVSzZNMDAw0Ol8cvqSE9CfrJRT/RqS9f33ATMzGSZMsMCnn8qwaRMfu3ZVwtvbVEMp9euYNjalxWjBggXc32FhYRoPsn37dkyYMIEb3rwhfHx84OPjwz3Ozc1VRzSNkEqlOp1PTl9yAvqTlXKqX0OzdusGTJxogm++MUFwcDG6dSuGJp+yrh9TbY7GoLQYyWQylTbwJsWjunv37mH16tUAgKKiIly5cgV8Ph8SiQSpqanccvn5+XBxcVHLPgkhzRv16NMdSovRRx99pNIG1HU7oOodG6KiouDm5gZ3d3eUlJRg165dKCl5eY05JSUF48ePV8s+CSHNF/Xo0y1Ki9G6deu4v5OSknD+/HmMGjWKO808ePAg+vXrp/KOIiMjkZqaiuLiYgQFBcHf3x+VlZUAAF9fX6XricVifPDBB9xlwzFjxlBPOkLIG9OFHn3kf3iMMfa6hT7//HOEh4ejZcuW3LSSkhIsWLAAa9eu1WjAN5GZmantCErp+rVjOX3JCehPVsqpfvqSVddzarPNSKUGn9LSUpSXlytMq6ioQGlpqUZCEUIIaV5U+tHrwIEDsWTJErz77ruwsLBAXl4eDh8+jIEDB2o6HyGEkGZApWI0ceJEWFtb4+zZsygoKECrVq3wzjvvKHSjJoQQQhpKpWLE5/Ph6+tbZ0cDQgghpKGUFqO6bgFUnbruwEAIIaT5UlqM6roFUHVUjAghhLwppcWoMW4BRAghhAAqthnVdWsgdd0OiBBCSPOlUjGq69ZA6rodECGEkOZLpWJU/dZAAFBQUIADBw6gT58+GglFCCGkeVHpGpulpaXCPycnJ8yYMQMHDx7UdD5CCCHNQIMbfEpLS1FUVKTOLIQQQpoplS7TrV27Fjwej3tcXl6Omzdv4u2339ZYMEIIIc2HSsXI2tpa4bGhoSGGDh2KHj16aCQUIYSQ5kWlYvThhx9qOgchpBFER4vh6lqhMF5PYqIQKSlCTJtWosVkpLlTqRgBwPHjx3Hq1Cnk5+dDIpFgwIABGDRokCazEULUzNW1QmE00+qjnRKiTSoVo3379uHkyZMYPnw4NzjUoUOHUFBQgNGjR2s6IyFETeSjmQYFmSMgoBSxscb1Hmabzq6IJqjUm+7o0aMIDQ2Fj48PevbsCR8fHyxcuBAJCQmazkcIUTNPzwoEBJQiMtIEAQGl9R5iW352lZgoBADu7MrVtfkN1R0dLeaOg1xiohDR0WItJdJfKhWj8vJymJqaKkwzMTFBRUXze/MRou8SE4WIjTVGcHAxYmONa3yYvk71s6uVK00ULvs1N1SY1UelYtSzZ0+sWbMGmZmZqKioQEZGBtatWwdXV1dN5yOEqFH1NqJ584q5otKQgvQmZ1dNBRVm9VGpGH3yyScwMjLC3LlzMWnSJMybNw8ikQiffPKJpvMRQtQoJUWo8GEp/zBNSalfMXrTs6umhAqzevAYY0zVhWUyGYqLi2FiYqIXd+vOzMzUdgSl5B1BdJ2+5AT0J6u+56x+dvVqjzxtfRBr85jKn78qHUJ0/bW3sbHR2r7rVVHkBejixYvIyMjQSCBCiG5T19lVU6Cuy57kNV278/PzsXXrVjx+/BhOTk4YPnw4wsLCwOfz8ffff2PGjBnw9PRsrKyEEB1QW/dtT8+KZnl5qq7C3ByPx5uosxht2rQJZmZmmDx5Ms6ePYtly5YhKCgI7u7uuHjxIuLj46kYEUKaLSrM6lPnZbo7d+7gn//8J3r16oUpU6agsLAQffv2BQD07dsXT58+bZSQhBBCmrY6z4yqqqpgYPByEUNDQ4hEIoW7d6sqOjoaSUlJMDMzQ0RERI358rMsHo8HgUCAwMBAdO7cGQAwduxY2NnZAXjZ+Dd//vx6758QQohue20xun79OvdYJpPVeKwKb29v+Pn5ISoqqtb53bt3R58+fcDj8fDw4UOsWrUKkZGRAAChUIiVK1eqtB9CCCH6qc5iZGZmhvXr13OPxWKxwuNX78qgjIuLC3JycpTOF4lE3N/l5eUNOvsihBCiv+osRsrOZDThwoUL+PHHH1FYWIgFCxZw01+8eIGQkBAIBAK8//77cHd3V7qNhIQE7n554eHhkEqlGs/dUAYGBjqdT05fcgL6k5Vyqp++ZNWXnNqg8hAScsuXL1coFuri7u4Od3d3pKamIj4+Hl9++SWAl+1NEokET548weLFi2FnZ1djsD85Hx8f+Pj4cI91+cdluv7jNzl9yQnoT1bKqX76klXXc+rNj14B4NatW5rIwXFxccGTJ09QVFQEAJBIJACA1q1bw8XFBQ8ePNDo/gkhhDS+ehejetw9SGXZ2dncdu/fv48XL17AxMQEJSUlePHiBQCgqKgIt2/fRrt27dS+f0L0AQ1XQJqyel+m+/TTT+u9k8jISKSmpqK4uBhBQUHw9/dHZWUlAMDX1xfnz5/HqVOnIBAIIBQKMXv2bPB4PGRkZGDTpk3g8/mQyWQYOXIkFSPSbNEoraQpq9eNUuWuX78OPp8PFxcXTWRSG7pR6pvTl5yA/mR9k5z1uSnnm9KX4wnoT1Zdz6nzbUZhYWFcW9GBAwewevVqrF69Gvv27dNoOEKIIhqugDRVKhWj9PR0ODk5AXg5BHlYWBiWLVuGI0eOaDQcIUQRjSNEmiqV2ozkV/Kys7MBgGu3+fvvvzUUixDyqlfHDfLwKNf6OEKEqItKZ0bOzs7YunUr4uLiuBulZmdnw8TERKPhiP6inl/qR+MIkaZMpWI0ffp0GBsbw97eHv7+/gBedg4YNmyYRsMR/SXv+SUvSPJv9a6u9A2+oaZNK6lxBuTpWVHrMAaE6BuVLtOZmJhg/PjxCtN69+6tkUCkaZB/a2+snl+EEP2mUjF68eIF9u7di8TERBQXFyMmJgYpKSnIysqCn5+fpjMSPVW951dwcDEVIkKIUipdpouJiUF6ejpmzpzJ3VHb1tYWf/zxh0bDEf1GPb8IIapSqRhduHABM2fOhJOTE1eMJBIJ8vPzNRqO6K/qPb/mzSvmLtlRQSJEEXX2eUmlYmRgYFBjIL2ioiLqTUeUop5fhKiGOvu8pFKb0VtvvYV169YhMDAQAFBQUIDt27fDw8NDk9mIHquth5enZwW1GxHyCurs85JKZ0bjx4+HlZUV5syZg9LSUsycORPm5ub48MMPNZ2PEEKaPLrNk4pnRgYGBggMDERgYCB3eY6GBieEaFN0tBiuropn24mJQqSkCPXut1evdvbx8ChvdgVJ6ZlRTk4O9/eTJ0+4f8+fP0dOTg73mBBCtKGptLVQZ5+XlJ4ZzZ07F7GxsQCAmTNnKt1AfHy8+lMRQshrVG9rSU4GNmzQz/v01dXZR9+ey5to0HhG+oLGM3pz2sxZ38swdEzVS19yrlxpwv2wet68Ym3HqZOuH1OdH8/owoULqKqq0nQWQhQ0lcswRHPkbS0LF1bRD6v1nEodGPbs2YP169ejX79+8PLyQrdu3TSdixDq8krqVL2t5f33TdGzZxENqaHHVCpGK1euxOPHj3HmzBls3LgRL168gIeHB7y8vNChQwdNZyTNGN3fjihDbS1NS4PajO7cuYPdu3fj2rVrOt2BgdqM3py2c8q//apyZqTtrKqinOqnL1l1Pac224xUOjOSy83NxdmzZ3HmzBk8ffoUgwYN0lQuQmhkU0KaEZWK0e+//44zZ87g4cOH6NWrF8aMGYPevXvDwKBetYyQeqHLMIQ0HypVk6SkJAwdOhTu7u4QiUSazkQIALq/HSHNiUrFaMGCBZrOQQghpBlT+TrbpUuXkJqaiqKiIoXpM2bMUHsoQgghzYtKP3rds2cPNm3aBJlMhvPnz0MsFiMlJQXGxsaazkcIIaQZUOnM6Pjx4wgNDYWdnR1OnDiBwMBAeHl54eeff1Z5R9HR0UhKSoKZmRkiIiJqzL948SLi4+PB4/EgEAgQGBiIzp07AwBOnDiBffv2AQBGjx4Nb29vlfdLCCFE96lUjP7++2/Y2dm9XMHAAJWVlXB0dERqaqrKO/L29oafnx+ioqJqnd+9e3f06dMHPB4PDx8+xKpVqxAZGYmSkhLs3bsX4eHhAICQkBD06dMHYnHzGpKXEEKaMpUu01lbWyM9PR0AYGtriz/++AOnTp2qV0FwcXGpc3mRSMSNkVReXs79nZycjB49ekAsFkMsFqNHjx5ITk5Web+EEEJ0n0pnRmPHjkVx8cu74Y4fPx5r1qxBWVkZpkyZotYwFy5cwI8//ojCwkKuB19+fj4sLCy4ZSQSCfLz82tdPyEhAQkJCQCA8PBwSKVSteZTJwMDA53OJ6cvOQH9yUo51U9fsupLTm1QqRj17t2b+7tTp05Yu3atRsK4u7vD3d0dqampiI+Px5dfflmv9X18fODj48M91uXbbuj6bUHk9CUnoD9ZKaf66UtWXc+p80NIPH78GEeOHMH+/ftx5MgRPH78WKOhXFxc8OTJExQVFUEikSAvL4+bl5+fD4lEotH9k6YjOlpcY1iBxEQhoqOpzZEQXVJnMWKMITo6GnPmzMH+/ftx6dIl7Nu3D3PnzkV0dDTUOS5fdnY2t7379+/jxYsXMDExQc+ePZGSkoKSkhKUlJQgJSUFPXv2VNt+SdNGYyKR5qApfOmq8zJdQkICUlNTsWzZMjg6OnLT7969i9WrV+PIkSPw9fVVaUeRkZFITU1FcXExgoKC4O/vj8rKSgCAr68vzp8/j1OnTkEgEEAoFGL27Nng8XgQi8X44IMPuDakMWPGUE86ojIaE4k0B/IvXfL3dvWbDOuLOoeQ+PLLLzFy5Ei4ubnVmHf58mUcOHAAS5Ys0WjAN0FDSLw5fckJ1J1Vl4am1pdjqi85Af3Jqsmc9RluRRmdbTN6/PgxXFxcap3n4uKi8bYjQtRBPjR1cHAxDU1NmqzqA1EGBJTq3dl/ncVIJpPByMio1nlGRkaQyWQaCUWIulS/XDFvXjF3yY4KEmlq9P1LV51tRlVVVbh+/brS+VSMiK6jMZFIc9AUBqKssxiZmZlh/fr1SuebmpqqPRAh6kRjIpHmoCl86aqzGCm7jxwhhBDd0RS+dKn0o1dCCCFEk6gYEUII0ToqRoQQQrSOihEhhBCto2JEamgK97kihOgXKkakBrq5KCGksak0nhFpXqrfXDQ5GdiwQb9+PEcI0T90ZkRqJb/P1TffCPTyPleEEP1CxYjUSn6fq4ULq/TyPleEEP1CxYjUUP0+V2FhVXRzUUKIxlExIjXUdZ8rQgjRBOrAQGpoCve5IoToFzozIoQQonVUjAghhGgdFSNCCCFaR8WIEEKI1lExIoQQonVUjAghhGgdFSNCCCFaR8WIEEKI1lExIoQQonWNcgeG6OhoJCUlwczMDBERETXmnz59GgcPHgRjDEZGRpgyZQrat28PAJg+fTpEIhH4fD4EAgHCw8MbIzIhhJBG1CjFyNvbG35+foiKiqp1vpWVFb7++muIxWJcuXIFmzZtwjfffMPNDwsLg6mpaWNEJYQQogWNcpnOxcUFYrHyIaudnZ25+Z06dUJeXl5jxGqyaNhwQoi+0bk2o2PHjqFXr14K05YtW4b58+cjISFBS6n0Cw0bTgjRNzp11+7r16/j+PHjWLx4MTdtyZIlkEgkKCwsxNKlS2FjYwMXF5da109ISOAKVnh4OKRSaaPkbggDAwON5Xv/fcDMTIYJEyzw6acybNrEx65dlfD2rv+lTk3mVDd9yUo51U9fsupLTm3QmWL08OFDbNy4EQsWLICJiQk3XSKRAADMzMzQt29f3L17V2kx8vHxgY+PD/c4NzdXs6HfgFQq1Wi+bt2AiRNN8M03JggOLka3bsVoyO40nVOd9CUr5VQ/fcmq6zltbGy0tm+duEyXm5uL7777DjNmzFA4GGVlZXj+/Dn399WrV2FnZ6etmHpFPmx4cHAxDRtOCNF5jXJmFBkZidTUVBQXFyMoKAj+/v6orKwEAPj6+mLv3r0oKSnBDz/8AABcF+7CwkJ89913AICqqip4eXmhZ8+ejRFZr1UfNtzTswIeHuUKjwkhRNfwGGNM2yE0JTMzU9sRlNLk6Xp0tBiuroojsyYmCpGSIqx1FNe66Pplher0JSvlVD99yarrObV5mU5n2oyI+tCw4YQQfaMTbUaENHX02y9C6kbFiJBGQL/9IqRudJmOkEbg6VmBDRsKEBRkjuRkYMMG6lBCSHV0ZkRII/H0rEBAQCm++UaAgIBSKkSEVEPFiJBGIv/t18KFVfTbL0JeQcWIkEZQ/bdfYWFV3CU7KkiEvETFiJBGkJIiVGgjkrchpaRQMSIEoA4MhDQK+u0XIXWjMyNCCCFaR8WIEEKI1lExIoQQonVUjAghhGgdFSNCCCFa16SHkCCEEKIf6MxIS0JCQrQdQSX6khPQn6yUU/30Jau+5NQGKkaEEEK0jooRIYQQrRN8/fXXX2s7RHPVoUMHbUdQib7kBPQnK+VUP33Jqi85Gxt1YCCEEKJ1dJmOEEKI1lExIoQQonV0124NyszMxKpVq7jHOTk58Pf3x7vvvstNu3HjBlasWAErKysAQL9+/TBmzBiNZ4uOjkZSUhLMzMwQEREBACgpKcGqVavw9OlTWFpaYvbs2RCLxTXWPXHiBPbt2wcAGD16NLy9vRs9a1xcHC5fvgwDAwO0bt0a06ZNQ8uWLWusO336dIhEIvD5fAgEAoSHhzdqzt27d+Po0aMwNTUFAHz00Ufo3bt3jXWTk5Oxbds2yGQyDBkyBCNHjmzUnKtWrUJmZiYAoLS0FMbGxli5cmWNdRvzeObm5iIqKgrPnj0Dj8eDj48Phg0bpnPvU2U5dfE9qtMYaRRVVVVsypQpLCcnR2H69evX2fLlyxs9z40bN9i9e/fY//t//4+bFhcXx/bv388YY2z//v0sLi6uxnrFxcVs+vTprLi4WOHvxs6anJzMKisrudy1ZWWMsWnTprHCwkKN5qsrZ3x8PDt48GCd61VVVbEZM2aw7Oxs9uLFCzZ37lyWnp7eqDmri4mJYXv27Kl1XmMez/z8fHbv3j3GGGOlpaVs5syZLD09Xefep8py6uJ7VJfRZbpGcu3aNVhbW8PS0lLbUQAALi4uNb5NXrx4EQMHDgQADBw4EBcvXqyxXnJyMnr06AGxWAyxWIwePXogOTm50bO6urpCIBAAAJycnJCfn6/RDKqoLacq7t69C2tra7Ru3RoGBgbw8PCo9dirS105GWM4d+4cPD09NbZ/VZmbm3M9z4yMjNC2bVvk5+fr3PtUWU5dfI/qMrpM10gSExOV/ge/c+cO5s2bB3Nzc0yaNAm2traNnO6lwsJCmJubAwBatWqFwsLCGsvk5+fDwsKCeyyRSLT+n+zYsWPw8PBQOn/ZsmUAgKFDh8LHx6exYnF+//13nDp1Ch06dEBAQECNQvDqMbWwsEBaWlpjxwQA3Lx5E2ZmZmjTpo3SZbRxPHNycvDXX3/B0dFRp9+n1XNWp+vvUV1AxagRVFZW4vLlyxg/fnyNeQ4ODoiOjoZIJEJSUhJWrlyJNWvWaCGlIh6PBx6Pp+0Yr7Vv3z4IBAK8/fbbtc5fsmQJJBIJCgsLsXTpUtjY2MDFxaXR8vn6+nJtgPHx8YiNjcW0adMabf/1VdeXJkA7x7OsrAwREREIDAyEsbGxwjxdep8qy6nr71FdQZfpGsGVK1fg4OCAVq1a1ZhnbGwMkUgEAOjduzeqqqpQVFTU2BEBAGZmZigoKAAAFBQUcI3u1UkkEuTl5XGP8/PzIZFIGi1jdSdOnMDly5cxc+ZMpR9I8mxmZmbo27cv7t6925gR0apVK/D5fPD5fAwZMgT37t2rNWP1Y5qXl6eVY1pVVYULFy7U+Q2+sY9nZWUlIiIi8Pbbb6Nfv37cvnXtfVpbTkA/3qO6gopRI6jr2+azZ8/A/vu747t370Imk8HExKQx43H69OmDkydPAgBOnjyJvn371limZ8+eSElJQUlJCUpKSpCSkoKePXs2dlQkJyfj4MGDmD9/PgwNDWtdpqysDM+fP+f+vnr1Kuzs7BozJvehCQAXLlyo9RJsx44dkZWVhZycHFRWVuLs2bPo06dPY8YE8LJd08bGRuHyVnWNfTwZY9iwYQPatm2L9957j5uua+9TZTn15T2qK+gODBpWVlaGadOmYd26ddyp+x9//AHg5SWc//znP/jjjz8gEAggFAoREBAAZ2dnjeeKjIxEamoqiouLYWZmBn9/f/Tt2xerVq1Cbm6uQpfZe/fu4ciRIwgKCgLw8vr3/v37AbzsMjto0KBGz7p//35UVlZy7S+dOnXCp59+ivz8fGzcuBELFizAkydP8N133wF4+a3fy8sLo0ePbtScN27cwIMHD8Dj8WBpaYlPP/0U5ubmCjkBICkpCTExMZDJZBg0aFCj5xw8eDCioqLQqVMn+Pr6cstq83jeunULX331Fezs7Lizio8++gidOnXSqfepspzbtm3TufeoLqNiRAghROvoMh0hhBCto2JECCFE66gYEUII0ToqRoQQQrSOihEhhBCto2JEmrWoqCj89NNPWtk3YwzR0dH4+OOPuS7e9eXv74/s7Gw1JyOk8dHtgIhOmT59OsrLy7Fu3TruzhRHjx7F6dOn8fXXX2s3nJrdunULV69exfr167nn+qqCggL89NNPuHLlCsrKyiCRSODh4YERI0YoXae+oqKiYGFhgXHjxqlle4Q0BJ0ZEZ0jk8nw22+/aTtGvclksnotLx+PR1lRKSkpQWhoKCoqKrB06VLExsYiNDQUf//9N548eaKOyGpRVVWl7QikCaAzI6JzRowYgYMHD+Kdd96pMRhZTk4OZsyYgV27dnG35//666/x9ttvY8iQIThx4gSOHj2Kjh074sSJExCLxfj888+RlZWF+Ph4vHjxAhMnTlQYaK2oqAhLlixBWloaHBwcMGPGDG6oj4yMDGzduhX379+Hqakpxo4dy927LSoqCkKhELm5uUhNTcW8efPQo0cPhbz5+fnYvHkzbt26BbFYjPfffx8+Pj44duwYtmzZgsrKSkyaNAnDhw+Hv7+/wrr//ve/IRKJ8Pnnn4PPf/m9USqV4uOPP671uFU/DgC4Y7FkyRIwxhATE4MzZ87gxYsXkEqlmDVrFu7cuYMzZ84AAH799Vd07doVISEhyM/Px9atW3Hz5k2IRCK8++67GDZsGICXAwamp6ejRYsWuHz5MgICAmBvb48ffvgBWVlZEAqF8PLywuTJk+v92pPmi4oR0TkdOnRA165d8csvvzTo0lFaWhoGDx6MrVu3Yvfu3YiMjISbmxvWrFmD1NRURERE4K233uLOSM6cOYOQkBB06tQJO3bswJo1a7BkyRKUlZVh6dKl8Pf3x8KFC/Ho0SMsXboUdnZ2aNeuHbfuggULMH/+fFRWVtbIsnr1atja2mLjxo3IzMzEkiVLYG1tjcGDB4PP53PFojbXrl1Dv379uEL0JlJSUnDz5k2sXr0axsbGyMjIQMuWLeHj44Pbt28rXKaTyWT49ttv0bdvXwQHByMvLw9LliyBjY0Nd3+3S5cuYfbs2ZgxYwYqKyvxr3/9C8OGDcOAAQNQVlaGR48evXFm0rzQZTqik/z9/XH48OEG3cHcysoKgwYNAp/Ph4eHB/Ly8jBmzBi0aNECrq6uMDAwUGj07927N1xcXNCiRQt89NFHuHPnDnJzc5GUlARLS0sMGjQIAoEADg4O6NevH86dO8et27dvX3Tu3Bl8Ph9CoVAhR25uLm7duoUJEyZAKBSiffv2GDJkCHeTz9cpLi6u9U7vDWFgYICysjJkZGSAMYZ27dpxYwK96t69eygqKsKYMWO4IbOHDBmCs2fPcss4OTnB3d2de97yY1pUVASRSAQnJye15CbNB50ZEZ1kZ2cHNzc3HDhwAG3btq3XumZmZtzf8gJR/UNdKBSirKyMe1z9LtUikQhisRgFBQV4+vQp0tLSEBgYyM2vqqrCgAEDvw6pvgAAAyBJREFUal33VQUFBRCLxTAyMuKmSaXSWoeRqI2JiQmePXum0rKv061bN7zzzjvYsmULcnNz4e7ujkmTJtUYHwh42ZZVUFCg8LxlMhm6dOnCPX71eQcFBSE+Ph6zZ8+GlZUVxowZAzc3N7VkJ80DFSOis/z9/TF//nyF2/LLL62Vl5dzH6Rv+oFdfdybsrIylJSUwNzcHBYWFnBxccGXX36pdN26BnYzNzdHSUkJnj9/zhWk3NxclcfV6d69Oy5cuIAxY8aodKnO0NAQ5eXl3ONXj8uwYcMwbNgwFBYWYtWqVTh06BDGjRtX4zlIpVJYWVnVa5DHNm3aIDg4GDKZDBcuXMD333+PLVu2qK3HH2n66DId0VnW1tbo378/Dh8+zE0zNTWFRCLB6dOnIZPJcOzYsTfuWXblyhXcunULlZWV+Omnn+Dk5ASpVAo3NzdkZWXh1KlTqKysRGVlJe7evYvHjx+rtF2pVApnZ2f8+OOPqKiowMOHD3H8+HGlI36+6r333sPz588RFRWFp0+fAnjZISImJgYPHz6ssXz79u1x4cIFlJeXIzs7G8eOHePm3b17F2lpaaisrIShoSFatGjBFTgzMzOFY+jo6AgjIyMcOHAAFRUVkMlkePToUZ2Dvp06dQpFRUXg8/nclwR1tHWR5oPOjIhOGzNmDE6fPq0wberUqfjhhx+wa9cuDB48+I3bJzw9PbFnzx7cuXMHHTp0wOeffw4AMDIyQmhoKGJiYhATEwPGGOzt7evVS2zWrFnYvHkzpk6dCrFYjA8//LBGjztlxGIxlixZgp9++gkLFy5EeXk5JBIJPD09YW1tXWP5d999F/fu3cM///lP2Nvbw8vLC9euXQMAPH/+HDExMXjy5AmEQiFcXV0xYsQIAMDgwYPx/fffIzAwEC4uLvjiiy8wf/58xMbGYvr06aisrISNjQ3Gjh2rNGtycjJiY2NRXl4OS0tLzJo1q0YbGiF1ofGMCCGEaB2dRxNCCNE6KkaEEEK0jooRIYQQraNiRAghROuoGBFCCNE6KkaEEEK0jooRIYQQraNiRAghROv+P5i6SOXzKTrIAAAAAElFTkSuQmCC",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "K = range(6,25)\n",
        "\n",
        "for num_clusters in tqdm(K):\n",
        "  spectral = SpectralClustering(n_clusters=num_clusters)\n",
        "  spectral.fit(clust_xtrain2)\n",
        "  cluster_labels = spectral.labels_\n",
        "  db_score = davies_bouldin_score(clust_xtrain2, cluster_labels)\n",
        "  #plt.plot(num_clusters,sil_score,'bx-')\n",
        "  plt.plot(num_clusters,db_score,'bx-')\n",
        "\n",
        "#plt.xlabel(\"Number of Clusters\") \n",
        "#plt.ylabel(\"Silhouette score\") \n",
        "#plt.title(\"Silhouette analysis For Ward Number of Clusters\")\n",
        "#plt.show()\n",
        "\n",
        "#Davies-Bouldin\n",
        "plt.xlabel(\"Number of Clusters\") \n",
        "plt.ylabel(\"Davies-Bouldin score\") \n",
        "plt.title(\"Davies-Bouldin analysis For Ward Number of Clusters\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HnW1yP6R59Ie"
      },
      "source": [
        "## K-Means Clustering \n",
        "\n",
        "Best Performing Clustering Algorithm\n",
        "--- Run only this"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 72,
      "metadata": {
        "id": "fFrDTSKDfHDU"
      },
      "outputs": [],
      "source": [
        "NUM_CLUSTERS = 13"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 73,
      "metadata": {
        "id": "_9j1IhJ1OZcp"
      },
      "outputs": [],
      "source": [
        "kmeans = KMeans(n_clusters=NUM_CLUSTERS, random_state=0).fit(clust_xtrain2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aPivD0RIO3n6",
        "outputId": "c16f5476-f5a5-4682-fa4c-5dc442338a9b"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(10000,)\n"
          ]
        }
      ],
      "source": [
        "print(kmeans.labels_.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 74,
      "metadata": {
        "id": "rlKClkoPRpzC"
      },
      "outputs": [],
      "source": [
        "unlablled_labels = kmeans.labels_"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SZZxya-o6DMg"
      },
      "source": [
        "## Ward Hierarchical Clustering"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "puZEQ646lxAi"
      },
      "outputs": [],
      "source": [
        "NUM_CLUSTERS = 6"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "id": "lQQHY7gE7RAE"
      },
      "outputs": [],
      "source": [
        "ward = AgglomerativeClustering(n_clusters=10).fit(clust_xtrain2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dzK6UlIm7Svt",
        "outputId": "6ff0e4e5-f06e-4416-8c56-f1a8dec4a638"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(10000,)\n"
          ]
        }
      ],
      "source": [
        "print(ward.labels_.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "id": "pAAPTlvb7XWa"
      },
      "outputs": [],
      "source": [
        "unlablled_labels = ward.labels_"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ug6kJS4R6Li4"
      },
      "source": [
        "## Spectral Clustering"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Nwy_5B7GlyyV"
      },
      "outputs": [],
      "source": [
        "NUM_CLUSTERS = 13"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 89,
      "metadata": {
        "id": "hciNGtU985xs"
      },
      "outputs": [],
      "source": [
        "spectral = SpectralClustering(n_clusters = 10).fit(clust_xtrain2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 90,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "myqGbr6y-XmJ",
        "outputId": "3d03bda5-e408-423d-bb14-a4a7f850d99f"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(10000,)\n"
          ]
        }
      ],
      "source": [
        "print(spectral.labels_.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "0GHf1Ea5-eBm"
      },
      "outputs": [],
      "source": [
        "unlablled_labels = spectral.labels_"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "m2JHpvgUslAD"
      },
      "source": [
        "## Label Processing for models"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 75,
      "metadata": {
        "id": "XyHuiVALglW_"
      },
      "outputs": [],
      "source": [
        "cat_unlabel = tf.keras.utils.to_categorical(\n",
        "    unlablled_labels, num_classes=NUM_CLUSTERS, dtype='float32'\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 76,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gSuRewWfO5tL",
        "outputId": "d40434f8-ec0b-40cc-f67b-d12f4cb81861"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0.]]\n",
            "[10  9 11 10  5]\n"
          ]
        }
      ],
      "source": [
        "cat_unlabel.shape\n",
        "print(cat_unlabel[:5])\n",
        "print(unlablled_labels[:5])\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 77,
      "metadata": {
        "id": "WavuY9R6S4dF"
      },
      "outputs": [],
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras import layers\n",
        "from keras.datasets import imdb\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense\n",
        "from keras.layers import LSTM, Conv1D, MaxPool1D, Flatten, Dropout, BatchNormalization\n",
        "from keras.layers import Bidirectional\n",
        "from keras.layers.embeddings import Embedding\n",
        "from keras.preprocessing import sequence\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oMpjWVeWS0uC"
      },
      "source": [
        "# Bi-LSTM Model Semi Supervised"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 78,
      "metadata": {
        "id": "HvGsyqyVyc6_"
      },
      "outputs": [],
      "source": [
        "model_LSTM_core = Sequential()\n",
        "model_LSTM_core.add(Bidirectional(LSTM(units= 100)))\n",
        "model_LSTM_core.add(Dropout(0.3))\n",
        "#model_LSTM_core.add(Dense(20, activation = 'relu'))\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 80,
      "metadata": {
        "id": "gPVnararS7lm"
      },
      "outputs": [],
      "source": [
        "from keras import activations\n",
        "from keras.layers.normalization.batch_normalization import BatchNormalization\n",
        "model_unsup = Sequential()\n",
        "\n",
        "model_unsup.add(keras.Input(shape=(X_train.shape[1],1)))\n",
        "model_unsup.add(model_LSTM_core)\n",
        "model_unsup.add(Dense(20, activation = 'relu'))\n",
        "model_unsup.add(Dense(NUM_CLUSTERS, activation='softmax'))\n",
        "model_unsup.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "model_unsup.build(input_shape= (10000,186,1))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 81,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xqCPcvhFUO7H",
        "outputId": "4ab9b67c-9d58-49bb-dee9-305d269b64ff"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(100000, 186, 1)\n",
            "(10000, 186, 1)\n"
          ]
        }
      ],
      "source": [
        "print(X_train.shape)\n",
        "print(clust_xtrain.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 82,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Mmx4HUl0TPYg",
        "outputId": "87fb712a-6ac9-4638-d99d-ff6eecc7a7ca"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " sequential (Sequential)     (None, 200)               81600     \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 20)                4020      \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 13)                273       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 85,893\n",
            "Trainable params: 85,893\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "model_unsup.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 109,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sxRPVn6vkyJ0",
        "outputId": "7985e141-fa8f-4c16-e959-1d8f6532684c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(100000, 186, 1)\n",
            "(10000, 10)\n",
            "(21892, 186, 1)\n",
            "(21892, 5)\n"
          ]
        }
      ],
      "source": [
        "print(X_train.shape)\n",
        "print(cat_unlabel.shape)\n",
        "print(X_test.shape)\n",
        "print(y_test.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 84,
      "metadata": {
        "id": "u7bWfNkF2mlf"
      },
      "outputs": [],
      "source": [
        "my_callbacks = [\n",
        "    tf.keras.callbacks.EarlyStopping(patience=3),\n",
        "    #tf.keras.callbacks.ModelCheckpoint(filepath='model.{epoch:02d}-{val_loss:.2f}.h5'),\n",
        "    #tf.keras.callbacks.TensorBoard(log_dir='./logs'),\n",
        "]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 85,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "40mfbv9gTcWt",
        "outputId": "9d36324e-8bfe-46d3-ab00-1c0614f75a3e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/30\n",
            "32/32 [==============================] - 12s 80ms/step - loss: 2.4706 - accuracy: 0.1521 - val_loss: 2.2777 - val_accuracy: 0.2415\n",
            "Epoch 2/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 2.0521 - accuracy: 0.2606 - val_loss: 1.7655 - val_accuracy: 0.3895\n",
            "Epoch 3/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 1.6536 - accuracy: 0.3829 - val_loss: 1.6632 - val_accuracy: 0.3545\n",
            "Epoch 4/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 1.3458 - accuracy: 0.4829 - val_loss: 1.1458 - val_accuracy: 0.5225\n",
            "Epoch 5/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 1.0009 - accuracy: 0.6214 - val_loss: 0.8223 - val_accuracy: 0.6920\n",
            "Epoch 6/30\n",
            "32/32 [==============================] - 1s 36ms/step - loss: 0.7806 - accuracy: 0.7136 - val_loss: 0.8770 - val_accuracy: 0.6845\n",
            "Epoch 7/30\n",
            "32/32 [==============================] - 1s 37ms/step - loss: 0.9433 - accuracy: 0.6605 - val_loss: 0.9823 - val_accuracy: 0.6005\n",
            "Epoch 8/30\n",
            "32/32 [==============================] - 1s 37ms/step - loss: 0.8872 - accuracy: 0.6635 - val_loss: 0.7875 - val_accuracy: 0.7065\n",
            "Epoch 9/30\n",
            "32/32 [==============================] - 1s 37ms/step - loss: 0.7604 - accuracy: 0.7141 - val_loss: 0.6755 - val_accuracy: 0.7540\n",
            "Epoch 10/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.6501 - accuracy: 0.7589 - val_loss: 0.5752 - val_accuracy: 0.7930\n",
            "Epoch 11/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.5580 - accuracy: 0.7947 - val_loss: 0.6360 - val_accuracy: 0.7480\n",
            "Epoch 12/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.6380 - accuracy: 0.7641 - val_loss: 0.6353 - val_accuracy: 0.7440\n",
            "Epoch 13/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.5150 - accuracy: 0.8135 - val_loss: 0.4169 - val_accuracy: 0.8805\n",
            "Epoch 14/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.4071 - accuracy: 0.8600 - val_loss: 0.3812 - val_accuracy: 0.8530\n",
            "Epoch 15/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.3841 - accuracy: 0.8673 - val_loss: 0.3323 - val_accuracy: 0.8920\n",
            "Epoch 16/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.3257 - accuracy: 0.8946 - val_loss: 0.3043 - val_accuracy: 0.8990\n",
            "Epoch 17/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.3734 - accuracy: 0.8681 - val_loss: 0.2750 - val_accuracy: 0.9130\n",
            "Epoch 18/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.3096 - accuracy: 0.8974 - val_loss: 0.3125 - val_accuracy: 0.8890\n",
            "Epoch 19/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.2621 - accuracy: 0.9114 - val_loss: 0.2199 - val_accuracy: 0.9310\n",
            "Epoch 20/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.2589 - accuracy: 0.9111 - val_loss: 0.2549 - val_accuracy: 0.9115\n",
            "Epoch 21/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.2494 - accuracy: 0.9144 - val_loss: 0.3155 - val_accuracy: 0.8840\n",
            "Epoch 22/30\n",
            "32/32 [==============================] - 1s 38ms/step - loss: 0.3003 - accuracy: 0.8964 - val_loss: 0.2587 - val_accuracy: 0.9160\n"
          ]
        }
      ],
      "source": [
        "history=model_unsup.fit(clust_xtrain, cat_unlabel, \n",
        "                        validation_split=0.2, \n",
        "                        batch_size=256, \n",
        "                        epochs = 30, \n",
        "                        callbacks = my_callbacks)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iREXiwCoz9fT"
      },
      "source": [
        "## Final Classification"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 86,
      "metadata": {
        "id": "xySr49Stz9EO"
      },
      "outputs": [],
      "source": [
        "model_sup_core = model_LSTM_core"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 87,
      "metadata": {
        "id": "_4p4bC-U3MEB"
      },
      "outputs": [],
      "source": [
        "model_sup_core.trainable=False"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 92,
      "metadata": {
        "id": "qZw5cEtK0kcv"
      },
      "outputs": [],
      "source": [
        "model_sup = Sequential()\n",
        "\n",
        "model_sup.add(keras.Input(shape=(X_train.shape[1],1)))\n",
        "model_sup.add(model_sup_core)\n",
        "model_sup.add(Dense(30, activation = 'relu'))\n",
        "model_sup.add(Dense(10, activation = 'relu'))\n",
        "model_sup.add(Dense(5, activation='softmax'))\n",
        "model_sup.compile(loss='categorical_crossentropy', optimizer=tf.keras.optimizers.Adam(learning_rate=1e-3), metrics=['accuracy'])\n",
        "model_sup.build(input_shape= (10000,186,1))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 89,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MYqu6M_scDP3",
        "outputId": "9f796c27-4fdf-449d-828c-556683212172"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential_3\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " sequential (Sequential)     (None, 200)               81600     \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 30)                6030      \n",
            "                                                                 \n",
            " dense_5 (Dense)             (None, 10)                310       \n",
            "                                                                 \n",
            " dense_6 (Dense)             (None, 5)                 55        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 87,995\n",
            "Trainable params: 6,395\n",
            "Non-trainable params: 81,600\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "model_sup.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 93,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jQWtSmMa4nNS",
        "outputId": "7046b9b5-2a8d-46d1-eb3f-63a56bb79530"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "391/391 [==============================] - 12s 21ms/step - loss: 0.9633 - accuracy: 0.6156 - val_loss: 0.9503 - val_accuracy: 0.5932\n",
            "Epoch 2/10\n",
            "391/391 [==============================] - 7s 17ms/step - loss: 0.6846 - accuracy: 0.7451 - val_loss: 0.8030 - val_accuracy: 0.7009\n",
            "Epoch 3/10\n",
            "391/391 [==============================] - 7s 17ms/step - loss: 0.6394 - accuracy: 0.7637 - val_loss: 0.7907 - val_accuracy: 0.7144\n",
            "Epoch 4/10\n",
            "391/391 [==============================] - 7s 17ms/step - loss: 0.6142 - accuracy: 0.7755 - val_loss: 0.7216 - val_accuracy: 0.7457\n",
            "Epoch 5/10\n",
            "391/391 [==============================] - 6s 16ms/step - loss: 0.5937 - accuracy: 0.7843 - val_loss: 0.7694 - val_accuracy: 0.7288\n",
            "Epoch 6/10\n",
            "391/391 [==============================] - 6s 17ms/step - loss: 0.5840 - accuracy: 0.7879 - val_loss: 0.7542 - val_accuracy: 0.7347\n",
            "Epoch 7/10\n",
            "391/391 [==============================] - 7s 17ms/step - loss: 0.5711 - accuracy: 0.7918 - val_loss: 0.7581 - val_accuracy: 0.7217\n"
          ]
        }
      ],
      "source": [
        "sup_history=model_sup.fit(X_train, y_train, validation_data=(X_test, y_test), \n",
        "                          batch_size=256, \n",
        "                          epochs = 10, \n",
        "                          callbacks=my_callbacks)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jhiPoIKGsVxl"
      },
      "source": [
        "## Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 133,
      "metadata": {
        "id": "7ULn_C-e6lY_"
      },
      "outputs": [],
      "source": [
        "y_pred = model_sup.predict(X_test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 135,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "70YmMoe66v9A",
        "outputId": "e46cdbcd-7b70-40f2-df19-e9a95c16003f"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "685/685 [==============================] - 7s 10ms/step - loss: 0.7581 - accuracy: 0.7217\n",
            "Test loss: 0.7580764293670654\n",
            "Test accuracy: 0.721724808216095\n"
          ]
        }
      ],
      "source": [
        "loss,accuracy=model_sup.evaluate(X_test,y_test)\n",
        "\n",
        "print('Test loss:', loss)\n",
        "\n",
        "print('Test accuracy:', accuracy)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 136,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 517
        },
        "id": "i2Nd1mAX69kC",
        "outputId": "49176c34-5647-42ac-b060-929d5c924415"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 576x576 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "import seaborn as sns\n",
        "\n",
        "Y_pred_classes = np.argmax(y_pred,axis = 1) \n",
        "Y_true = np.argmax(y_test,axis = 1) \n",
        "\n",
        "confusion_mtx = confusion_matrix(Y_true, Y_pred_classes) \n",
        "\n",
        "f,ax = plt.subplots(figsize=(8, 8))\n",
        "sns.heatmap(confusion_mtx, annot=True, linewidths=0.01,cmap=\"Oranges\",linecolor=\"gray\", fmt= '.1f',ax=ax)\n",
        "plt.xlabel(\"Predicted Label\")\n",
        "plt.ylabel(\"True Label\")\n",
        "plt.title(\"Semi LSTM Confusion Matrix\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 97,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 599
        },
        "id": "zRAnZY9G6o5-",
        "outputId": "f71b1e31-e5d3-4f83-d417-ac859417d7ac"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fb17cfbc810>"
            ]
          },
          "execution_count": 97,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(0)\n",
        "\n",
        "plt.plot(history.history['accuracy'], label='training accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label='val accuracy')\n",
        "plt.title('Accuracy')\n",
        "plt.xlabel('epochs')\n",
        "plt.ylabel('accuracy')\n",
        "plt.legend()\n",
        "\n",
        "plt.figure(1)\n",
        "plt.plot(history.history['loss'], label='training loss')\n",
        "plt.plot(history.history['val_loss'], label='val loss')\n",
        "plt.title('Loss')\n",
        "plt.xlabel('epochs')\n",
        "plt.ylabel('loss')\n",
        "plt.legend()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Pj0agHjn6HvN"
      },
      "outputs": [],
      "source": [
        "model_sup.summary()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Hr4xgNhklPKP"
      },
      "source": [
        "# CNN Model Semi-Supervised"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 98,
      "metadata": {
        "id": "CfS9EIDylTpT"
      },
      "outputs": [],
      "source": [
        "model_CNN_core = Sequential()\n",
        "#model_LSTM_core.add(Bidirectional(LSTM(units= 187)))\n",
        "#model_LSTM_core.add(Dropout(0.5))\n",
        "model_CNN_core.add(Conv1D(filters=32, kernel_size=(3,), padding='same', activation='relu', input_shape = (X_train.shape[1],1)))\n",
        "model_CNN_core.add(Conv1D(filters=64, kernel_size=(3,), padding='same', activation='relu')) \n",
        "model_CNN_core.add(Conv1D(filters=128, kernel_size=(5,), padding='same', activation='relu'))    \n",
        "\n",
        "model_CNN_core.add(MaxPool1D(pool_size=(3,), strides=2, padding='same'))\n",
        "model_CNN_core.add(Dropout(0.5))\n",
        "\n",
        "model_CNN_core.add(Flatten())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 111,
      "metadata": {
        "id": "VyQQqo14nGan"
      },
      "outputs": [],
      "source": [
        "from keras.layers.normalization.batch_normalization import BatchNormalization\n",
        "CNN_model_unsup = Sequential()\n",
        "\n",
        "CNN_model_unsup.add(keras.Input(shape=(X_train.shape[1],1)))\n",
        "#model_LSTM.add(Bidirectional(LSTM(units= 187)))\n",
        "#model_LSTM.add(Dropout(0.5))\n",
        "CNN_model_unsup.add(model_CNN_core)\n",
        "CNN_model_unsup.add(Dense(NUM_CLUSTERS, activation='softmax'))\n",
        "CNN_model_unsup.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "CNN_model_unsup.build(input_shape= (10000,186,1))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 112,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "liiWM84FnTzm",
        "outputId": "afbadd58-b61b-4e28-e16d-6ec113869abc"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential_9\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " sequential_5 (Sequential)   (None, 11904)             47424     \n",
            "                                                                 \n",
            " dense_14 (Dense)            (None, 13)                154765    \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 202,189\n",
            "Trainable params: 154,765\n",
            "Non-trainable params: 47,424\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "CNN_model_unsup.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 113,
      "metadata": {
        "id": "0e7Hi4_0nXdH"
      },
      "outputs": [],
      "source": [
        "history = CNN_model_unsup.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 115,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "unGo2ZMknbee",
        "outputId": "844c0f38-cf99-4c77-e9c5-4ac4827549f8"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/50\n",
            "32/32 [==============================] - 0s 12ms/step - loss: 0.3148 - accuracy: 0.9069 - val_loss: 0.3270 - val_accuracy: 0.9045\n",
            "Epoch 2/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.3062 - accuracy: 0.9094 - val_loss: 0.3171 - val_accuracy: 0.9095\n",
            "Epoch 3/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.3016 - accuracy: 0.9081 - val_loss: 0.3104 - val_accuracy: 0.9110\n",
            "Epoch 4/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2939 - accuracy: 0.9118 - val_loss: 0.3079 - val_accuracy: 0.9115\n",
            "Epoch 5/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2883 - accuracy: 0.9128 - val_loss: 0.3009 - val_accuracy: 0.9130\n",
            "Epoch 6/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2830 - accuracy: 0.9147 - val_loss: 0.2967 - val_accuracy: 0.9120\n",
            "Epoch 7/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2808 - accuracy: 0.9134 - val_loss: 0.2919 - val_accuracy: 0.9145\n",
            "Epoch 8/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.2758 - accuracy: 0.9147 - val_loss: 0.2874 - val_accuracy: 0.9140\n",
            "Epoch 9/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2696 - accuracy: 0.9158 - val_loss: 0.2833 - val_accuracy: 0.9170\n",
            "Epoch 10/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.2681 - accuracy: 0.9187 - val_loss: 0.2802 - val_accuracy: 0.9165\n",
            "Epoch 11/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2642 - accuracy: 0.9154 - val_loss: 0.2794 - val_accuracy: 0.9165\n",
            "Epoch 12/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2609 - accuracy: 0.9190 - val_loss: 0.2744 - val_accuracy: 0.9185\n",
            "Epoch 13/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2573 - accuracy: 0.9186 - val_loss: 0.2704 - val_accuracy: 0.9170\n",
            "Epoch 14/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2555 - accuracy: 0.9174 - val_loss: 0.2697 - val_accuracy: 0.9170\n",
            "Epoch 15/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2530 - accuracy: 0.9215 - val_loss: 0.2652 - val_accuracy: 0.9180\n",
            "Epoch 16/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.2476 - accuracy: 0.9210 - val_loss: 0.2607 - val_accuracy: 0.9190\n",
            "Epoch 17/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2444 - accuracy: 0.9212 - val_loss: 0.2588 - val_accuracy: 0.9195\n",
            "Epoch 18/50\n",
            "32/32 [==============================] - 0s 10ms/step - loss: 0.2407 - accuracy: 0.9230 - val_loss: 0.2583 - val_accuracy: 0.9205\n",
            "Epoch 19/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2414 - accuracy: 0.9206 - val_loss: 0.2559 - val_accuracy: 0.9210\n",
            "Epoch 20/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.2377 - accuracy: 0.9240 - val_loss: 0.2527 - val_accuracy: 0.9220\n",
            "Epoch 21/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.2367 - accuracy: 0.9240 - val_loss: 0.2486 - val_accuracy: 0.9205\n",
            "Epoch 22/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2331 - accuracy: 0.9227 - val_loss: 0.2489 - val_accuracy: 0.9225\n",
            "Epoch 23/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2285 - accuracy: 0.9255 - val_loss: 0.2442 - val_accuracy: 0.9220\n",
            "Epoch 24/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2274 - accuracy: 0.9254 - val_loss: 0.2473 - val_accuracy: 0.9205\n",
            "Epoch 25/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2255 - accuracy: 0.9265 - val_loss: 0.2427 - val_accuracy: 0.9240\n",
            "Epoch 26/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.2245 - accuracy: 0.9246 - val_loss: 0.2398 - val_accuracy: 0.9250\n",
            "Epoch 27/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2226 - accuracy: 0.9249 - val_loss: 0.2388 - val_accuracy: 0.9245\n",
            "Epoch 28/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2205 - accuracy: 0.9274 - val_loss: 0.2375 - val_accuracy: 0.9210\n",
            "Epoch 29/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2210 - accuracy: 0.9270 - val_loss: 0.2338 - val_accuracy: 0.9235\n",
            "Epoch 30/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2174 - accuracy: 0.9274 - val_loss: 0.2343 - val_accuracy: 0.9270\n",
            "Epoch 31/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2166 - accuracy: 0.9268 - val_loss: 0.2317 - val_accuracy: 0.9245\n",
            "Epoch 32/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2161 - accuracy: 0.9269 - val_loss: 0.2304 - val_accuracy: 0.9280\n",
            "Epoch 33/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.2147 - accuracy: 0.9268 - val_loss: 0.2299 - val_accuracy: 0.9235\n",
            "Epoch 34/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2099 - accuracy: 0.9299 - val_loss: 0.2268 - val_accuracy: 0.9280\n",
            "Epoch 35/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2089 - accuracy: 0.9301 - val_loss: 0.2257 - val_accuracy: 0.9290\n",
            "Epoch 36/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.2091 - accuracy: 0.9324 - val_loss: 0.2255 - val_accuracy: 0.9265\n",
            "Epoch 37/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2073 - accuracy: 0.9293 - val_loss: 0.2225 - val_accuracy: 0.9285\n",
            "Epoch 38/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.2065 - accuracy: 0.9291 - val_loss: 0.2233 - val_accuracy: 0.9260\n",
            "Epoch 39/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2040 - accuracy: 0.9291 - val_loss: 0.2204 - val_accuracy: 0.9275\n",
            "Epoch 40/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2018 - accuracy: 0.9330 - val_loss: 0.2198 - val_accuracy: 0.9290\n",
            "Epoch 41/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2017 - accuracy: 0.9296 - val_loss: 0.2165 - val_accuracy: 0.9310\n",
            "Epoch 42/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.2002 - accuracy: 0.9321 - val_loss: 0.2184 - val_accuracy: 0.9285\n",
            "Epoch 43/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.1985 - accuracy: 0.9337 - val_loss: 0.2160 - val_accuracy: 0.9270\n",
            "Epoch 44/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.1978 - accuracy: 0.9310 - val_loss: 0.2153 - val_accuracy: 0.9290\n",
            "Epoch 45/50\n",
            "32/32 [==============================] - 0s 10ms/step - loss: 0.1964 - accuracy: 0.9341 - val_loss: 0.2123 - val_accuracy: 0.9315\n",
            "Epoch 46/50\n",
            "32/32 [==============================] - 0s 10ms/step - loss: 0.1964 - accuracy: 0.9311 - val_loss: 0.2143 - val_accuracy: 0.9305\n",
            "Epoch 47/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.1955 - accuracy: 0.9355 - val_loss: 0.2128 - val_accuracy: 0.9310\n",
            "Epoch 48/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.1964 - accuracy: 0.9334 - val_loss: 0.2112 - val_accuracy: 0.9310\n",
            "Epoch 49/50\n",
            "32/32 [==============================] - 0s 8ms/step - loss: 0.1919 - accuracy: 0.9352 - val_loss: 0.2087 - val_accuracy: 0.9290\n",
            "Epoch 50/50\n",
            "32/32 [==============================] - 0s 9ms/step - loss: 0.1925 - accuracy: 0.9365 - val_loss: 0.2081 - val_accuracy: 0.9295\n"
          ]
        }
      ],
      "source": [
        "history=CNN_model_unsup.fit(clust_xtrain, cat_unlabel, validation_split=0.2, batch_size=256, epochs = 50)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TTDcsW6HsAXd"
      },
      "source": [
        "## CNN Final Classification"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 116,
      "metadata": {
        "id": "WFA4XT2pn35j"
      },
      "outputs": [],
      "source": [
        "CNN_model_sup_core = model_CNN_core"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 117,
      "metadata": {
        "id": "mxoWtLPloEtF"
      },
      "outputs": [],
      "source": [
        "CNN_model_sup_core.trainable=False"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 122,
      "metadata": {
        "id": "nsmyKgtvoQnk"
      },
      "outputs": [],
      "source": [
        "CNN_model_sup = Sequential()\n",
        "\n",
        "CNN_model_sup.add(keras.Input(shape=(X_train.shape[1],1)))\n",
        "CNN_model_sup.add(CNN_model_sup_core)\n",
        "CNN_model_sup.add(Dense(10, activation='relu'))\n",
        "CNN_model_sup.add(Dense(5, activation='softmax'))\n",
        "CNN_model_sup.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "CNN_model_sup.build(input_shape= (10000,186,1))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 119,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sMG1vYf1pvNa",
        "outputId": "88a33724-95f9-47fe-e93a-5ca03e545551"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential_10\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " sequential_5 (Sequential)   (None, 11904)             47424     \n",
            "                                                                 \n",
            " dense_15 (Dense)            (None, 10)                119050    \n",
            "                                                                 \n",
            " dense_16 (Dense)            (None, 5)                 55        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 166,529\n",
            "Trainable params: 119,105\n",
            "Non-trainable params: 47,424\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "CNN_model_sup.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 123,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_lSOoIBWoeoK",
        "outputId": "dc9e7f6b-4bd0-4082-dd9f-dc454b50d9de"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/30\n",
            "391/391 [==============================] - 4s 9ms/step - loss: 0.8072 - accuracy: 0.7181 - val_loss: 0.7959 - val_accuracy: 0.7368\n",
            "Epoch 2/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.5491 - accuracy: 0.8069 - val_loss: 0.6475 - val_accuracy: 0.7760\n",
            "Epoch 3/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.4711 - accuracy: 0.8331 - val_loss: 0.5652 - val_accuracy: 0.8093\n",
            "Epoch 4/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.4302 - accuracy: 0.8461 - val_loss: 0.5798 - val_accuracy: 0.7986\n",
            "Epoch 5/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.4007 - accuracy: 0.8555 - val_loss: 0.5197 - val_accuracy: 0.8184\n",
            "Epoch 6/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.3785 - accuracy: 0.8644 - val_loss: 0.5650 - val_accuracy: 0.7918\n",
            "Epoch 7/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.3629 - accuracy: 0.8693 - val_loss: 0.4831 - val_accuracy: 0.8330\n",
            "Epoch 8/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.3458 - accuracy: 0.8756 - val_loss: 0.4787 - val_accuracy: 0.8340\n",
            "Epoch 9/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.3345 - accuracy: 0.8801 - val_loss: 0.4355 - val_accuracy: 0.8522\n",
            "Epoch 10/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.3237 - accuracy: 0.8830 - val_loss: 0.3849 - val_accuracy: 0.8734\n",
            "Epoch 11/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.3145 - accuracy: 0.8862 - val_loss: 0.4507 - val_accuracy: 0.8429\n",
            "Epoch 12/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.3081 - accuracy: 0.8891 - val_loss: 0.3866 - val_accuracy: 0.8680\n",
            "Epoch 13/30\n",
            "391/391 [==============================] - 3s 8ms/step - loss: 0.2993 - accuracy: 0.8922 - val_loss: 0.3901 - val_accuracy: 0.8662\n"
          ]
        }
      ],
      "source": [
        "CNN_sup_history=CNN_model_sup.fit(X_train, y_train, validation_data=(X_test, y_test), batch_size=256, epochs = 30, callbacks=my_callbacks)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RUUTGO2JsbNg"
      },
      "source": [
        "## Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 130,
      "metadata": {
        "id": "bvj8aV_sokEp"
      },
      "outputs": [],
      "source": [
        "y_pred = CNN_model_sup.predict(X_test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 131,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-bM1lZVzokx0",
        "outputId": "5bafbfb0-284b-4d4b-ba32-a543cfca008c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "685/685 [==============================] - 3s 5ms/step - loss: 0.3901 - accuracy: 0.8662\n",
            "Test loss: 0.39013975858688354\n",
            "Test accuracy: 0.8662068247795105\n"
          ]
        }
      ],
      "source": [
        "loss,accuracy=CNN_model_sup.evaluate(X_test,y_test)\n",
        "\n",
        "print('Test loss:', loss)\n",
        "\n",
        "print('Test accuracy:', accuracy)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 132,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 517
        },
        "id": "0CTPCmmnowUq",
        "outputId": "a20f7fe5-dfe2-49d8-f0bd-6ba407db10a2"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 576x576 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "import seaborn as sns\n",
        "\n",
        "Y_pred_classes = np.argmax(y_pred,axis = 1) \n",
        "Y_true = np.argmax(y_test,axis = 1) \n",
        "\n",
        "confusion_mtx = confusion_matrix(Y_true, Y_pred_classes) \n",
        "\n",
        "f,ax = plt.subplots(figsize=(8, 8))\n",
        "sns.heatmap(confusion_mtx, annot=True, linewidths=0.01,cmap=\"Oranges\",linecolor=\"gray\", fmt= '.1f',ax=ax)\n",
        "plt.xlabel(\"Predicted Label\")\n",
        "plt.ylabel(\"True Label\")\n",
        "plt.title(\"Semi CNN Confusion Matrix\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 127,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 599
        },
        "id": "7iSZ-sqIo1hd",
        "outputId": "2d20dcdb-d89e-4898-c3e4-215f1091b585"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fb17cff3410>"
            ]
          },
          "execution_count": 127,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(0)\n",
        "\n",
        "plt.plot(CNN_sup_history.history['accuracy'], label='training accuracy')\n",
        "plt.plot(CNN_sup_history.history['val_accuracy'], label='val accuracy')\n",
        "plt.title('Accuracy')\n",
        "plt.xlabel('epochs')\n",
        "plt.ylabel('accuracy')\n",
        "plt.legend()\n",
        "\n",
        "plt.figure(1)\n",
        "plt.plot(CNN_sup_history.history['loss'], label='training loss')\n",
        "plt.plot(CNN_sup_history.history['val_loss'], label='val loss')\n",
        "plt.title('Loss')\n",
        "plt.xlabel('epochs')\n",
        "plt.ylabel('loss')\n",
        "plt.legend()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 128,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9-bLFMB9pEFP",
        "outputId": "c0141641-a8a7-46f5-e4bb-c738af99ffb7"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential_11\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " sequential_5 (Sequential)   (None, 11904)             47424     \n",
            "                                                                 \n",
            " dense_17 (Dense)            (None, 10)                119050    \n",
            "                                                                 \n",
            " dense_18 (Dense)            (None, 5)                 55        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 166,529\n",
            "Trainable params: 119,105\n",
            "Non-trainable params: 47,424\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "CNN_model_sup.summary()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ajgnk6i53U74"
      },
      "source": [
        "# Attention Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 131,
      "metadata": {
        "id": "Chv5odZoQeaz"
      },
      "outputs": [],
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras import layers"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 132,
      "metadata": {
        "id": "Ju8AyowHQepV"
      },
      "outputs": [],
      "source": [
        "def transformer_encoder(inputs, head_size, num_heads, ff_dim, dropout=0):\n",
        "    # Normalization and Attention\n",
        "    x = layers.LayerNormalization(epsilon=1e-6)(inputs)\n",
        "    x = layers.MultiHeadAttention(\n",
        "        key_dim=head_size, num_heads=num_heads, dropout=dropout\n",
        "    )(x, x)\n",
        "    x = layers.Dropout(dropout)(x)\n",
        "    res = x + inputs\n",
        "\n",
        "    # Feed Forward Part\n",
        "    x = layers.LayerNormalization(epsilon=1e-6)(res)\n",
        "    x = layers.Conv1D(filters=ff_dim, kernel_size=1, activation=\"relu\")(x)\n",
        "    x = layers.Dropout(dropout)(x)\n",
        "    x = layers.Conv1D(filters=inputs.shape[-1], kernel_size=1)(x)\n",
        "    return x + res"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 133,
      "metadata": {
        "id": "B3yK68AnQgN3"
      },
      "outputs": [],
      "source": [
        "def build_model(\n",
        "    input_shape,\n",
        "    head_size,\n",
        "    num_heads,\n",
        "    ff_dim,\n",
        "    num_transformer_blocks,\n",
        "    mlp_units,\n",
        "    dropout=0,\n",
        "    mlp_dropout=0,\n",
        "):\n",
        "    inputs = keras.Input(shape=input_shape)\n",
        "    x = inputs\n",
        "    for _ in range(num_transformer_blocks):\n",
        "        x = transformer_encoder(x, head_size, num_heads, ff_dim, dropout)\n",
        "\n",
        "    x = layers.GlobalAveragePooling1D(data_format=\"channels_first\")(x)\n",
        "    for dim in mlp_units:\n",
        "        x = layers.Dense(dim, activation=\"relu\")(x)\n",
        "        x = layers.Dropout(mlp_dropout)(x)\n",
        "    outputs = layers.Dense(5, activation=\"softmax\")(x)\n",
        "    return keras.Model(inputs, outputs)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZrFkFQP8Qh3l",
        "outputId": "b95267c5-4c56-4bf4-fcf6-ec7431819eef"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"model\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_32 (InputLayer)          [(None, 186, 1)]     0           []                               \n",
            "                                                                                                  \n",
            " layer_normalization (LayerNorm  (None, 186, 1)      2           ['input_32[0][0]']               \n",
            " alization)                                                                                       \n",
            "                                                                                                  \n",
            " multi_head_attention (MultiHea  (None, 186, 1)      14337       ['layer_normalization[0][0]',    \n",
            " dAttention)                                                      'layer_normalization[0][0]']    \n",
            "                                                                                                  \n",
            " dropout_8 (Dropout)            (None, 186, 1)       0           ['multi_head_attention[0][0]']   \n",
            "                                                                                                  \n",
            " tf.__operators__.add (TFOpLamb  (None, 186, 1)      0           ['dropout_8[0][0]',              \n",
            " da)                                                              'input_32[0][0]']               \n",
            "                                                                                                  \n",
            " layer_normalization_1 (LayerNo  (None, 186, 1)      2           ['tf.__operators__.add[0][0]']   \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv1d (Conv1D)                (None, 186, 8)       16          ['layer_normalization_1[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_9 (Dropout)            (None, 186, 8)       0           ['conv1d[0][0]']                 \n",
            "                                                                                                  \n",
            " conv1d_1 (Conv1D)              (None, 186, 1)       9           ['dropout_9[0][0]']              \n",
            "                                                                                                  \n",
            " tf.__operators__.add_1 (TFOpLa  (None, 186, 1)      0           ['conv1d_1[0][0]',               \n",
            " mbda)                                                            'tf.__operators__.add[0][0]']   \n",
            "                                                                                                  \n",
            " layer_normalization_2 (LayerNo  (None, 186, 1)      2           ['tf.__operators__.add_1[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " multi_head_attention_1 (MultiH  (None, 186, 1)      14337       ['layer_normalization_2[0][0]',  \n",
            " eadAttention)                                                    'layer_normalization_2[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_10 (Dropout)           (None, 186, 1)       0           ['multi_head_attention_1[0][0]'] \n",
            "                                                                                                  \n",
            " tf.__operators__.add_2 (TFOpLa  (None, 186, 1)      0           ['dropout_10[0][0]',             \n",
            " mbda)                                                            'tf.__operators__.add_1[0][0]'] \n",
            "                                                                                                  \n",
            " layer_normalization_3 (LayerNo  (None, 186, 1)      2           ['tf.__operators__.add_2[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv1d_2 (Conv1D)              (None, 186, 8)       16          ['layer_normalization_3[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_11 (Dropout)           (None, 186, 8)       0           ['conv1d_2[0][0]']               \n",
            "                                                                                                  \n",
            " conv1d_3 (Conv1D)              (None, 186, 1)       9           ['dropout_11[0][0]']             \n",
            "                                                                                                  \n",
            " tf.__operators__.add_3 (TFOpLa  (None, 186, 1)      0           ['conv1d_3[0][0]',               \n",
            " mbda)                                                            'tf.__operators__.add_2[0][0]'] \n",
            "                                                                                                  \n",
            " layer_normalization_4 (LayerNo  (None, 186, 1)      2           ['tf.__operators__.add_3[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " multi_head_attention_2 (MultiH  (None, 186, 1)      14337       ['layer_normalization_4[0][0]',  \n",
            " eadAttention)                                                    'layer_normalization_4[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_12 (Dropout)           (None, 186, 1)       0           ['multi_head_attention_2[0][0]'] \n",
            "                                                                                                  \n",
            " tf.__operators__.add_4 (TFOpLa  (None, 186, 1)      0           ['dropout_12[0][0]',             \n",
            " mbda)                                                            'tf.__operators__.add_3[0][0]'] \n",
            "                                                                                                  \n",
            " layer_normalization_5 (LayerNo  (None, 186, 1)      2           ['tf.__operators__.add_4[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv1d_4 (Conv1D)              (None, 186, 8)       16          ['layer_normalization_5[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_13 (Dropout)           (None, 186, 8)       0           ['conv1d_4[0][0]']               \n",
            "                                                                                                  \n",
            " conv1d_5 (Conv1D)              (None, 186, 1)       9           ['dropout_13[0][0]']             \n",
            "                                                                                                  \n",
            " tf.__operators__.add_5 (TFOpLa  (None, 186, 1)      0           ['conv1d_5[0][0]',               \n",
            " mbda)                                                            'tf.__operators__.add_4[0][0]'] \n",
            "                                                                                                  \n",
            " layer_normalization_6 (LayerNo  (None, 186, 1)      2           ['tf.__operators__.add_5[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " multi_head_attention_3 (MultiH  (None, 186, 1)      14337       ['layer_normalization_6[0][0]',  \n",
            " eadAttention)                                                    'layer_normalization_6[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_14 (Dropout)           (None, 186, 1)       0           ['multi_head_attention_3[0][0]'] \n",
            "                                                                                                  \n",
            " tf.__operators__.add_6 (TFOpLa  (None, 186, 1)      0           ['dropout_14[0][0]',             \n",
            " mbda)                                                            'tf.__operators__.add_5[0][0]'] \n",
            "                                                                                                  \n",
            " layer_normalization_7 (LayerNo  (None, 186, 1)      2           ['tf.__operators__.add_6[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv1d_6 (Conv1D)              (None, 186, 8)       16          ['layer_normalization_7[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_15 (Dropout)           (None, 186, 8)       0           ['conv1d_6[0][0]']               \n",
            "                                                                                                  \n",
            " conv1d_7 (Conv1D)              (None, 186, 1)       9           ['dropout_15[0][0]']             \n",
            "                                                                                                  \n",
            " tf.__operators__.add_7 (TFOpLa  (None, 186, 1)      0           ['conv1d_7[0][0]',               \n",
            " mbda)                                                            'tf.__operators__.add_6[0][0]'] \n",
            "                                                                                                  \n",
            " layer_normalization_8 (LayerNo  (None, 186, 1)      2           ['tf.__operators__.add_7[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " multi_head_attention_4 (MultiH  (None, 186, 1)      14337       ['layer_normalization_8[0][0]',  \n",
            " eadAttention)                                                    'layer_normalization_8[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_16 (Dropout)           (None, 186, 1)       0           ['multi_head_attention_4[0][0]'] \n",
            "                                                                                                  \n",
            " tf.__operators__.add_8 (TFOpLa  (None, 186, 1)      0           ['dropout_16[0][0]',             \n",
            " mbda)                                                            'tf.__operators__.add_7[0][0]'] \n",
            "                                                                                                  \n",
            " layer_normalization_9 (LayerNo  (None, 186, 1)      2           ['tf.__operators__.add_8[0][0]'] \n",
            " rmalization)                                                                                     \n",
            "                                                                                                  \n",
            " conv1d_8 (Conv1D)              (None, 186, 8)       16          ['layer_normalization_9[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_17 (Dropout)           (None, 186, 8)       0           ['conv1d_8[0][0]']               \n",
            "                                                                                                  \n",
            " conv1d_9 (Conv1D)              (None, 186, 1)       9           ['dropout_17[0][0]']             \n",
            "                                                                                                  \n",
            " tf.__operators__.add_9 (TFOpLa  (None, 186, 1)      0           ['conv1d_9[0][0]',               \n",
            " mbda)                                                            'tf.__operators__.add_8[0][0]'] \n",
            "                                                                                                  \n",
            " layer_normalization_10 (LayerN  (None, 186, 1)      2           ['tf.__operators__.add_9[0][0]'] \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " multi_head_attention_5 (MultiH  (None, 186, 1)      14337       ['layer_normalization_10[0][0]', \n",
            " eadAttention)                                                    'layer_normalization_10[0][0]'] \n",
            "                                                                                                  \n",
            " dropout_18 (Dropout)           (None, 186, 1)       0           ['multi_head_attention_5[0][0]'] \n",
            "                                                                                                  \n",
            " tf.__operators__.add_10 (TFOpL  (None, 186, 1)      0           ['dropout_18[0][0]',             \n",
            " ambda)                                                           'tf.__operators__.add_9[0][0]'] \n",
            "                                                                                                  \n",
            " layer_normalization_11 (LayerN  (None, 186, 1)      2           ['tf.__operators__.add_10[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv1d_10 (Conv1D)             (None, 186, 8)       16          ['layer_normalization_11[0][0]'] \n",
            "                                                                                                  \n",
            " dropout_19 (Dropout)           (None, 186, 8)       0           ['conv1d_10[0][0]']              \n",
            "                                                                                                  \n",
            " conv1d_11 (Conv1D)             (None, 186, 1)       9           ['dropout_19[0][0]']             \n",
            "                                                                                                  \n",
            " tf.__operators__.add_11 (TFOpL  (None, 186, 1)      0           ['conv1d_11[0][0]',              \n",
            " ambda)                                                           'tf.__operators__.add_10[0][0]']\n",
            "                                                                                                  \n",
            " layer_normalization_12 (LayerN  (None, 186, 1)      2           ['tf.__operators__.add_11[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " multi_head_attention_6 (MultiH  (None, 186, 1)      14337       ['layer_normalization_12[0][0]', \n",
            " eadAttention)                                                    'layer_normalization_12[0][0]'] \n",
            "                                                                                                  \n",
            " dropout_20 (Dropout)           (None, 186, 1)       0           ['multi_head_attention_6[0][0]'] \n",
            "                                                                                                  \n",
            " tf.__operators__.add_12 (TFOpL  (None, 186, 1)      0           ['dropout_20[0][0]',             \n",
            " ambda)                                                           'tf.__operators__.add_11[0][0]']\n",
            "                                                                                                  \n",
            " layer_normalization_13 (LayerN  (None, 186, 1)      2           ['tf.__operators__.add_12[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv1d_12 (Conv1D)             (None, 186, 8)       16          ['layer_normalization_13[0][0]'] \n",
            "                                                                                                  \n",
            " dropout_21 (Dropout)           (None, 186, 8)       0           ['conv1d_12[0][0]']              \n",
            "                                                                                                  \n",
            " conv1d_13 (Conv1D)             (None, 186, 1)       9           ['dropout_21[0][0]']             \n",
            "                                                                                                  \n",
            " tf.__operators__.add_13 (TFOpL  (None, 186, 1)      0           ['conv1d_13[0][0]',              \n",
            " ambda)                                                           'tf.__operators__.add_12[0][0]']\n",
            "                                                                                                  \n",
            " layer_normalization_14 (LayerN  (None, 186, 1)      2           ['tf.__operators__.add_13[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " multi_head_attention_7 (MultiH  (None, 186, 1)      14337       ['layer_normalization_14[0][0]', \n",
            " eadAttention)                                                    'layer_normalization_14[0][0]'] \n",
            "                                                                                                  \n",
            " dropout_22 (Dropout)           (None, 186, 1)       0           ['multi_head_attention_7[0][0]'] \n",
            "                                                                                                  \n",
            " tf.__operators__.add_14 (TFOpL  (None, 186, 1)      0           ['dropout_22[0][0]',             \n",
            " ambda)                                                           'tf.__operators__.add_13[0][0]']\n",
            "                                                                                                  \n",
            " layer_normalization_15 (LayerN  (None, 186, 1)      2           ['tf.__operators__.add_14[0][0]']\n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " conv1d_14 (Conv1D)             (None, 186, 8)       16          ['layer_normalization_15[0][0]'] \n",
            "                                                                                                  \n",
            " dropout_23 (Dropout)           (None, 186, 8)       0           ['conv1d_14[0][0]']              \n",
            "                                                                                                  \n",
            " conv1d_15 (Conv1D)             (None, 186, 1)       9           ['dropout_23[0][0]']             \n",
            "                                                                                                  \n",
            " tf.__operators__.add_15 (TFOpL  (None, 186, 1)      0           ['conv1d_15[0][0]',              \n",
            " ambda)                                                           'tf.__operators__.add_14[0][0]']\n",
            "                                                                                                  \n",
            " global_average_pooling1d (Glob  (None, 186)         0           ['tf.__operators__.add_15[0][0]']\n",
            " alAveragePooling1D)                                                                              \n",
            "                                                                                                  \n",
            " dense_81 (Dense)               (None, 128)          23936       ['global_average_pooling1d[0][0]'\n",
            "                                                                 ]                                \n",
            "                                                                                                  \n",
            " dropout_24 (Dropout)           (None, 128)          0           ['dense_81[0][0]']               \n",
            "                                                                                                  \n",
            " dense_82 (Dense)               (None, 5)            645         ['dropout_24[0][0]']             \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 139,509\n",
            "Trainable params: 139,509\n",
            "Non-trainable params: 0\n",
            "__________________________________________________________________________________________________\n",
            "Epoch 1/50\n",
            " 52/625 [=>............................] - ETA: 5:05 - loss: 1.6474 - categorical_accuracy: 0.2121"
          ]
        }
      ],
      "source": [
        "input_shape = X_train.shape[1:]\n",
        "\n",
        "model = build_model(\n",
        "    input_shape,\n",
        "    head_size=256,\n",
        "    num_heads=8,\n",
        "    ff_dim=8,\n",
        "    num_transformer_blocks=8,\n",
        "    mlp_units=[128],\n",
        "    mlp_dropout=0.4,\n",
        "    dropout=0.25,\n",
        ")\n",
        "\n",
        "model.compile(\n",
        "    loss=\"categorical_crossentropy\",\n",
        "    optimizer=keras.optimizers.Adam(learning_rate=1e-4),\n",
        "    metrics=[tf.keras.metrics.CategoricalAccuracy()],\n",
        ")\n",
        "\n",
        "#model.compile(optimizer='adam',loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True), metrics=['accuracy'])\n",
        "model.summary()\n",
        "\n",
        "\n",
        "callbacks = [keras.callbacks.EarlyStopping(patience=10, restore_best_weights=True)]\n",
        "\n",
        "shuffle_y = np.random.shuffle(y_train)\n",
        "\n",
        "model.fit(\n",
        "    X_train,\n",
        "    y_train,\n",
        "    validation_split=0.2,\n",
        "    epochs=50,\n",
        "    batch_size=128,\n",
        "    callbacks=callbacks,\n",
        ")\n",
        "\n",
        "model.evaluate(X_test, y_test, verbose=1)"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [
        "SzBzLE57XJE1",
        "ug6kJS4R6Li4",
        "ajgnk6i53U74"
      ],
      "name": "Semi-Supervised ECG Classification, Group 15.ipynb",
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}