Onion_Classifier.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Onion Classifier",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RJYgj0CtmWeK"
      },
      "source": [
        "# Init\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AIkFDzxjOGsJ",
        "outputId": "2d363f28-4e6a-4456-c86a-86abdc201472",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 100
        }
      },
      "source": [
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn import metrics\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from sklearn.model_selection import StratifiedKFold\n",
        "from sklearn.model_selection import cross_val_score\n",
        "from sklearn.model_selection import cross_validate\n",
        "\n",
        "from keras.wrappers.scikit_learn import KerasClassifier\n",
        "from keras.layers import  Dropout, Dense\n",
        "from keras.models import Sequential\n",
        "from keras import backend as K\n",
        "import tensorflow as tf\n",
        "import plotly.graph_objects as go\n",
        "\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import plotly.express as px\n",
        "\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import nltk\n",
        "import re\n",
        "\n",
        "from wordcloud import WordCloud\n",
        "\n",
        "nltk.download('punkt')\n",
        "nltk.download('stopwords')"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Unzipping tokenizers/punkt.zip.\n",
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/stopwords.zip.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_URPq7zrzH9B"
      },
      "source": [
        "ENG_SW = set(nltk.corpus.stopwords.words('english'))"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SYwYItFJUYVH"
      },
      "source": [
        "seed = 42\n",
        "np.random.seed(seed)"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wCHmHmwXmcj9"
      },
      "source": [
        "# Pre Proccess"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ak1ERjM6P6-e"
      },
      "source": [
        "df = pd.read_csv(\"onion-or-not.csv\")"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S_YSLrDwQpXt",
        "outputId": "ff4ad3c4-6257-4e94-9fb7-96b1d7a8c9c3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 197
        }
      },
      "source": [
        "df.head(5)"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Entire Facebook Staff Laughs As Man Tightens P...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Muslim Woman Denied Soda Can for Fear She Coul...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Bold Move: Hulu Has Announced That They’re Gon...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Despondent Jeff Bezos Realizes He’ll Have To W...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>For men looking for great single women, online...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                text  label\n",
              "0  Entire Facebook Staff Laughs As Man Tightens P...      1\n",
              "1  Muslim Woman Denied Soda Can for Fear She Coul...      0\n",
              "2  Bold Move: Hulu Has Announced That They’re Gon...      1\n",
              "3  Despondent Jeff Bezos Realizes He’ll Have To W...      1\n",
              "4  For men looking for great single women, online...      1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "s6iCI5fqQDrj",
        "outputId": "5f02892c-ac20-439c-83de-99e96b7b91fb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 66
        }
      },
      "source": [
        "df[\"label\"].value_counts()/len(df[\"label\"])"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    0.625\n",
              "1    0.375\n",
              "Name: label, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kEjsAgt0qXBY",
        "outputId": "5f6e287e-5c4d-47fe-f4b3-c39f747e2149",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 107
        }
      },
      "source": [
        "pd.DataFrame(df[\"label\"].replace({0:\"Not Onion\", 1:\"Onion\"}).value_counts())"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Not Onion</th>\n",
              "      <td>15000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Onion</th>\n",
              "      <td>9000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "           label\n",
              "Not Onion  15000\n",
              "Onion       9000"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gTswmT4hqH67",
        "outputId": "c03b13f7-c2dd-4e5e-a472-f4ab012bd97b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        }
      },
      "source": [
        "fig = px.pie(pd.DataFrame(df[\"label\"].value_counts()), values='label', title='Source Distribution',names=[\"Not Onion\",\"Onion\"])\n",
        "fig.show()"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<html>\n",
              "<head><meta charset=\"utf-8\" /></head>\n",
              "<body>\n",
              "    <div>\n",
              "            <script src=\"https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-AMS-MML_SVG\"></script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script>\n",
              "                <script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script>\n",
              "        <script src=\"https://cdn.plot.ly/plotly-latest.min.js\"></script>    \n",
              "            <div id=\"22745629-91d6-40c1-aeb7-a3f3353810b1\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>\n",
              "            <script type=\"text/javascript\">\n",
              "                \n",
              "                    window.PLOTLYENV=window.PLOTLYENV || {};\n",
              "                    \n",
              "                if (document.getElementById(\"22745629-91d6-40c1-aeb7-a3f3353810b1\")) {\n",
              "                    Plotly.newPlot(\n",
              "                        '22745629-91d6-40c1-aeb7-a3f3353810b1',\n",
              "                        [{\"domain\": {\"x\": [0.0, 1.0], \"y\": [0.0, 1.0]}, \"hoverlabel\": {\"namelength\": 0}, \"hovertemplate\": \"label=%{value}\", \"labels\": [\"Not Onion\", \"Onion\"], \"legendgroup\": \"\", \"name\": \"\", \"showlegend\": true, \"type\": \"pie\", \"values\": [15000, 9000]}],\n",
              "                        {\"legend\": {\"tracegroupgap\": 0}, \"template\": {\"data\": {\"bar\": [{\"error_x\": {\"color\": \"#2a3f5f\"}, \"error_y\": {\"color\": \"#2a3f5f\"}, \"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"bar\"}], \"barpolar\": [{\"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"barpolar\"}], \"carpet\": [{\"aaxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"baxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"type\": \"carpet\"}], \"choropleth\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"choropleth\"}], \"contour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"contour\"}], \"contourcarpet\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"contourcarpet\"}], \"heatmap\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmap\"}], \"heatmapgl\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmapgl\"}], \"histogram\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"histogram\"}], \"histogram2d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2d\"}], \"histogram2dcontour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2dcontour\"}], \"mesh3d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"mesh3d\"}], \"parcoords\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"parcoords\"}], \"pie\": [{\"automargin\": true, \"type\": \"pie\"}], \"scatter\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter\"}], \"scatter3d\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter3d\"}], \"scattercarpet\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattercarpet\"}], \"scattergeo\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergeo\"}], \"scattergl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergl\"}], \"scattermapbox\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattermapbox\"}], \"scatterpolar\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolar\"}], \"scatterpolargl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolargl\"}], \"scatterternary\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterternary\"}], \"surface\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"surface\"}], \"table\": [{\"cells\": {\"fill\": {\"color\": \"#EBF0F8\"}, \"line\": {\"color\": \"white\"}}, \"header\": {\"fill\": {\"color\": \"#C8D4E3\"}, \"line\": {\"color\": \"white\"}}, \"type\": \"table\"}]}, \"layout\": {\"annotationdefaults\": {\"arrowcolor\": \"#2a3f5f\", \"arrowhead\": 0, \"arrowwidth\": 1}, \"coloraxis\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"colorscale\": {\"diverging\": [[0, \"#8e0152\"], [0.1, \"#c51b7d\"], [0.2, \"#de77ae\"], [0.3, \"#f1b6da\"], [0.4, \"#fde0ef\"], [0.5, \"#f7f7f7\"], [0.6, \"#e6f5d0\"], [0.7, \"#b8e186\"], [0.8, \"#7fbc41\"], [0.9, \"#4d9221\"], [1, \"#276419\"]], \"sequential\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"sequentialminus\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"colorway\": [\"#636efa\", \"#EF553B\", \"#00cc96\", \"#ab63fa\", \"#FFA15A\", \"#19d3f3\", \"#FF6692\", \"#B6E880\", \"#FF97FF\", \"#FECB52\"], \"font\": {\"color\": \"#2a3f5f\"}, \"geo\": {\"bgcolor\": \"white\", \"lakecolor\": \"white\", \"landcolor\": \"#E5ECF6\", \"showlakes\": true, \"showland\": true, \"subunitcolor\": \"white\"}, \"hoverlabel\": {\"align\": \"left\"}, \"hovermode\": \"closest\", \"mapbox\": {\"style\": \"light\"}, \"paper_bgcolor\": \"white\", \"plot_bgcolor\": \"#E5ECF6\", \"polar\": {\"angularaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"radialaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"scene\": {\"xaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"yaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"zaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}}, \"shapedefaults\": {\"line\": {\"color\": \"#2a3f5f\"}}, \"ternary\": {\"aaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"baxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"caxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"title\": {\"x\": 0.05}, \"xaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}, \"yaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}}}, \"title\": {\"text\": \"Source Distribution\"}},\n",
              "                        {\"responsive\": true}\n",
              "                    ).then(function(){\n",
              "                            \n",
              "var gd = document.getElementById('22745629-91d6-40c1-aeb7-a3f3353810b1');\n",
              "var x = new MutationObserver(function (mutations, observer) {{\n",
              "        var display = window.getComputedStyle(gd).display;\n",
              "        if (!display || display === 'none') {{\n",
              "            console.log([gd, 'removed!']);\n",
              "            Plotly.purge(gd);\n",
              "            observer.disconnect();\n",
              "        }}\n",
              "}});\n",
              "\n",
              "// Listen for the removal of the full notebook cells\n",
              "var notebookContainer = gd.closest('#notebook-container');\n",
              "if (notebookContainer) {{\n",
              "    x.observe(notebookContainer, {childList: true});\n",
              "}}\n",
              "\n",
              "// Listen for the clearing of the current output cell\n",
              "var outputEl = gd.closest('.output');\n",
              "if (outputEl) {{\n",
              "    x.observe(outputEl, {childList: true});\n",
              "}}\n",
              "\n",
              "                        })\n",
              "                };\n",
              "                \n",
              "            </script>\n",
              "        </div>\n",
              "</body>\n",
              "</html>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A3d7jO3zqdHW"
      },
      "source": [
        "def prepossessing(df):\n",
        "\n",
        "    df[\"text\"] = df[\"text\"].apply(lambda x : x.lower())\n",
        "    df[\"text\"] = df[\"text\"].apply(lambda x :  re.sub('[^a-zA-z0-9\\s]','',x)) \n",
        "    df[\"text_list\"] = df[\"text\"].apply(lambda x : nltk.word_tokenize(x))\n",
        "    df[\"cleaned_list\"] = df[\"text_list\"].apply(lambda x: [word for word in x if word not in ENG_SW])\n",
        "    df[\"stemmed_cleaned_list\"] = df[\"cleaned_list\"].apply(lambda x : [nltk.PorterStemmer().stem(i) for i in x])\n",
        "    df['text_edited'] = df['stemmed_cleaned_list'].apply(lambda x : \" \".join(x))\n",
        "\n",
        "    return df['text_edited']"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z9ZgEaHmzMhx"
      },
      "source": [
        "df_X = prepossessing(df)"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JzRDa71m3iTb"
      },
      "source": [
        "df_Y = df['label']"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EIJxGUWwiz1y",
        "outputId": "4095501f-d615-4c3a-b10f-816c88d749a7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        }
      },
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(df_X, df_Y, test_size=0.25, random_state=seed)\n",
        "print(X_train.shape, y_train.shape)\n",
        "print(X_test.shape, y_test.shape)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(18000,) (18000,)\n",
            "(6000,) (6000,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ULhYDihtmo-q"
      },
      "source": [
        "# TF idf"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wR6Ix2Jklzsl"
      },
      "source": [
        "vectorizer_x = TfidfVectorizer()"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eAXx5TsHlpwa"
      },
      "source": [
        "X_train_tfidf = vectorizer_x.fit_transform(X_train).toarray()\n",
        "X_test_tfidf = vectorizer_x.transform(X_test).toarray()"
      ],
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "un4PMMhGmtni"
      },
      "source": [
        "# Model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rnlhf58sUKn6"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m9oUAGfRkFjn"
      },
      "source": [
        "def Build_Model_DNN_Text(shape, optimizer='Adam', nLayers= 1, neurons=16, dropout=0.85):\n",
        "    \"\"\"\n",
        "    Build_Model_DNN_Tex(shape, nClasses,dropout)\n",
        "    Build Deep neural networks Model for text classification\n",
        "    Shape is input feature space\n",
        "    \"\"\"\n",
        "    model = Sequential()\n",
        "     # number of  hidden layer\n",
        "    model.add(Dense(64,input_dim=shape,activation='elu'))\n",
        "    model.add(Dropout(dropout))\n",
        "    for i in range(0,nLayers):\n",
        "        model.add(Dense(neurons,input_dim=neurons,activation='elu'))\n",
        "        model.add(Dropout(dropout))\n",
        "    model.add(Dense(1, activation='sigmoid'))\n",
        "    model.compile(loss='binary_crossentropy',\n",
        "                  optimizer=optimizer,\n",
        "                  metrics=['accuracy',tf.keras.metrics.AUC(),tf.keras.metrics.Precision(),tf.keras.metrics.Recall()])\n",
        "    return model"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QpFkl2wQll_Z"
      },
      "source": [
        "#Val"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QKfAFKI2ljRd",
        "outputId": "0746b29f-c348-4094-8c59-7d592fd743c5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "model_DNN = Build_Model_DNN_Text(X_train_tfidf.shape[1])\n",
        "history = model_DNN.fit(X_train_tfidf, y_train,\n",
        "                            validation_data=(X_test_tfidf, y_test),\n",
        "                            epochs=100,\n",
        "                            batch_size=4096,\n",
        "                            verbose=2,\n",
        "                            callbacks=[tf.keras.callbacks.EarlyStopping(monitor='val_accuracy', mode = \"max\", patience=10)]\n",
        "                            )"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "5/5 - 1s - loss: 0.6907 - accuracy: 0.5343 - auc: 0.5052 - precision: 0.3813 - recall: 0.3808 - val_loss: 0.6779 - val_accuracy: 0.6292 - val_auc: 0.7134 - val_precision: 0.0000e+00 - val_recall: 0.0000e+00\n",
            "Epoch 2/100\n",
            "5/5 - 0s - loss: 0.6766 - accuracy: 0.5921 - auc: 0.5422 - precision: 0.4165 - recall: 0.2092 - val_loss: 0.6643 - val_accuracy: 0.6292 - val_auc: 0.8055 - val_precision: 0.0000e+00 - val_recall: 0.0000e+00\n",
            "Epoch 3/100\n",
            "5/5 - 0s - loss: 0.6672 - accuracy: 0.6031 - auc: 0.5535 - precision: 0.4042 - recall: 0.1153 - val_loss: 0.6525 - val_accuracy: 0.6292 - val_auc: 0.8453 - val_precision: 0.0000e+00 - val_recall: 0.0000e+00\n",
            "Epoch 4/100\n",
            "5/5 - 0s - loss: 0.6585 - accuracy: 0.6120 - auc: 0.5679 - precision: 0.4215 - recall: 0.0828 - val_loss: 0.6423 - val_accuracy: 0.6295 - val_auc: 0.8642 - val_precision: 1.0000 - val_recall: 8.9888e-04\n",
            "Epoch 5/100\n",
            "5/5 - 0s - loss: 0.6494 - accuracy: 0.6187 - auc: 0.5974 - precision: 0.4573 - recall: 0.0695 - val_loss: 0.6328 - val_accuracy: 0.6303 - val_auc: 0.8769 - val_precision: 1.0000 - val_recall: 0.0031\n",
            "Epoch 6/100\n",
            "5/5 - 0s - loss: 0.6383 - accuracy: 0.6269 - auc: 0.6322 - precision: 0.5356 - recall: 0.0666 - val_loss: 0.6229 - val_accuracy: 0.6342 - val_auc: 0.8831 - val_precision: 1.0000 - val_recall: 0.0135\n",
            "Epoch 7/100\n",
            "5/5 - 0s - loss: 0.6282 - accuracy: 0.6438 - auc: 0.6691 - precision: 0.6862 - recall: 0.0987 - val_loss: 0.6122 - val_accuracy: 0.6502 - val_auc: 0.8877 - val_precision: 0.9701 - val_recall: 0.0584\n",
            "Epoch 8/100\n",
            "5/5 - 0s - loss: 0.6136 - accuracy: 0.6707 - auc: 0.7188 - precision: 0.7969 - recall: 0.1680 - val_loss: 0.6003 - val_accuracy: 0.6862 - val_auc: 0.8911 - val_precision: 0.9647 - val_recall: 0.1596\n",
            "Epoch 9/100\n",
            "5/5 - 0s - loss: 0.5987 - accuracy: 0.6976 - auc: 0.7678 - precision: 0.8320 - recall: 0.2463 - val_loss: 0.5867 - val_accuracy: 0.7260 - val_auc: 0.8933 - val_precision: 0.9204 - val_recall: 0.2858\n",
            "Epoch 10/100\n",
            "5/5 - 0s - loss: 0.5851 - accuracy: 0.7211 - auc: 0.7971 - precision: 0.8434 - recall: 0.3179 - val_loss: 0.5711 - val_accuracy: 0.7555 - val_auc: 0.8952 - val_precision: 0.8948 - val_recall: 0.3861\n",
            "Epoch 11/100\n",
            "5/5 - 0s - loss: 0.5670 - accuracy: 0.7448 - auc: 0.8284 - precision: 0.8574 - recall: 0.3861 - val_loss: 0.5539 - val_accuracy: 0.7773 - val_auc: 0.8966 - val_precision: 0.8882 - val_recall: 0.4571\n",
            "Epoch 12/100\n",
            "5/5 - 0s - loss: 0.5515 - accuracy: 0.7605 - auc: 0.8470 - precision: 0.8749 - recall: 0.4244 - val_loss: 0.5360 - val_accuracy: 0.7898 - val_auc: 0.8978 - val_precision: 0.8777 - val_recall: 0.5034\n",
            "Epoch 13/100\n",
            "5/5 - 0s - loss: 0.5310 - accuracy: 0.7795 - auc: 0.8664 - precision: 0.8943 - recall: 0.4697 - val_loss: 0.5179 - val_accuracy: 0.7993 - val_auc: 0.8990 - val_precision: 0.8713 - val_recall: 0.5384\n",
            "Epoch 14/100\n",
            "5/5 - 0s - loss: 0.5120 - accuracy: 0.7934 - auc: 0.8792 - precision: 0.9010 - recall: 0.5067 - val_loss: 0.5002 - val_accuracy: 0.8060 - val_auc: 0.9001 - val_precision: 0.8582 - val_recall: 0.5712\n",
            "Epoch 15/100\n",
            "5/5 - 0s - loss: 0.4931 - accuracy: 0.8105 - auc: 0.8917 - precision: 0.9020 - recall: 0.5570 - val_loss: 0.4831 - val_accuracy: 0.8127 - val_auc: 0.9011 - val_precision: 0.8417 - val_recall: 0.6094\n",
            "Epoch 16/100\n",
            "5/5 - 0s - loss: 0.4789 - accuracy: 0.8210 - auc: 0.8983 - precision: 0.8961 - recall: 0.5932 - val_loss: 0.4669 - val_accuracy: 0.8170 - val_auc: 0.9021 - val_precision: 0.8305 - val_recall: 0.6364\n",
            "Epoch 17/100\n",
            "5/5 - 0s - loss: 0.4611 - accuracy: 0.8312 - auc: 0.9052 - precision: 0.8968 - recall: 0.6232 - val_loss: 0.4517 - val_accuracy: 0.8213 - val_auc: 0.9033 - val_precision: 0.8255 - val_recall: 0.6571\n",
            "Epoch 18/100\n",
            "5/5 - 0s - loss: 0.4445 - accuracy: 0.8382 - auc: 0.9127 - precision: 0.9021 - recall: 0.6396 - val_loss: 0.4380 - val_accuracy: 0.8250 - val_auc: 0.9043 - val_precision: 0.8223 - val_recall: 0.6737\n",
            "Epoch 19/100\n",
            "5/5 - 0s - loss: 0.4286 - accuracy: 0.8470 - auc: 0.9202 - precision: 0.9022 - recall: 0.6657 - val_loss: 0.4255 - val_accuracy: 0.8280 - val_auc: 0.9051 - val_precision: 0.8233 - val_recall: 0.6827\n",
            "Epoch 20/100\n",
            "5/5 - 0s - loss: 0.4134 - accuracy: 0.8509 - auc: 0.9241 - precision: 0.9015 - recall: 0.6781 - val_loss: 0.4146 - val_accuracy: 0.8285 - val_auc: 0.9060 - val_precision: 0.8201 - val_recall: 0.6885\n",
            "Epoch 21/100\n",
            "5/5 - 0s - loss: 0.3954 - accuracy: 0.8608 - auc: 0.9323 - precision: 0.9062 - recall: 0.7029 - val_loss: 0.4052 - val_accuracy: 0.8310 - val_auc: 0.9066 - val_precision: 0.8139 - val_recall: 0.7056\n",
            "Epoch 22/100\n",
            "5/5 - 0s - loss: 0.3865 - accuracy: 0.8689 - auc: 0.9355 - precision: 0.9143 - recall: 0.7193 - val_loss: 0.3971 - val_accuracy: 0.8313 - val_auc: 0.9070 - val_precision: 0.8099 - val_recall: 0.7124\n",
            "Epoch 23/100\n",
            "5/5 - 0s - loss: 0.3751 - accuracy: 0.8722 - auc: 0.9370 - precision: 0.9112 - recall: 0.7318 - val_loss: 0.3902 - val_accuracy: 0.8307 - val_auc: 0.9074 - val_precision: 0.8105 - val_recall: 0.7092\n",
            "Epoch 24/100\n",
            "5/5 - 0s - loss: 0.3643 - accuracy: 0.8739 - auc: 0.9413 - precision: 0.9119 - recall: 0.7362 - val_loss: 0.3846 - val_accuracy: 0.8312 - val_auc: 0.9078 - val_precision: 0.8124 - val_recall: 0.7083\n",
            "Epoch 25/100\n",
            "5/5 - 0s - loss: 0.3529 - accuracy: 0.8803 - auc: 0.9428 - precision: 0.9185 - recall: 0.7485 - val_loss: 0.3799 - val_accuracy: 0.8317 - val_auc: 0.9081 - val_precision: 0.8088 - val_recall: 0.7151\n",
            "Epoch 26/100\n",
            "5/5 - 0s - loss: 0.3388 - accuracy: 0.8881 - auc: 0.9483 - precision: 0.9207 - recall: 0.7690 - val_loss: 0.3763 - val_accuracy: 0.8332 - val_auc: 0.9082 - val_precision: 0.8039 - val_recall: 0.7276\n",
            "Epoch 27/100\n",
            "5/5 - 0s - loss: 0.3318 - accuracy: 0.8882 - auc: 0.9496 - precision: 0.9161 - recall: 0.7737 - val_loss: 0.3734 - val_accuracy: 0.8332 - val_auc: 0.9084 - val_precision: 0.7994 - val_recall: 0.7344\n",
            "Epoch 28/100\n",
            "5/5 - 0s - loss: 0.3234 - accuracy: 0.8952 - auc: 0.9532 - precision: 0.9242 - recall: 0.7860 - val_loss: 0.3713 - val_accuracy: 0.8343 - val_auc: 0.9084 - val_precision: 0.8019 - val_recall: 0.7348\n",
            "Epoch 29/100\n",
            "5/5 - 0s - loss: 0.3116 - accuracy: 0.8982 - auc: 0.9559 - precision: 0.9257 - recall: 0.7931 - val_loss: 0.3699 - val_accuracy: 0.8357 - val_auc: 0.9084 - val_precision: 0.7988 - val_recall: 0.7443\n",
            "Epoch 30/100\n",
            "5/5 - 0s - loss: 0.2994 - accuracy: 0.9023 - auc: 0.9598 - precision: 0.9312 - recall: 0.7994 - val_loss: 0.3691 - val_accuracy: 0.8353 - val_auc: 0.9083 - val_precision: 0.7984 - val_recall: 0.7438\n",
            "Epoch 31/100\n",
            "5/5 - 0s - loss: 0.2979 - accuracy: 0.9038 - auc: 0.9587 - precision: 0.9313 - recall: 0.8038 - val_loss: 0.3690 - val_accuracy: 0.8335 - val_auc: 0.9082 - val_precision: 0.8002 - val_recall: 0.7344\n",
            "Epoch 32/100\n",
            "5/5 - 0s - loss: 0.2915 - accuracy: 0.9077 - auc: 0.9605 - precision: 0.9381 - recall: 0.8080 - val_loss: 0.3693 - val_accuracy: 0.8355 - val_auc: 0.9079 - val_precision: 0.7973 - val_recall: 0.7461\n",
            "Epoch 33/100\n",
            "5/5 - 0s - loss: 0.2806 - accuracy: 0.9135 - auc: 0.9638 - precision: 0.9373 - recall: 0.8254 - val_loss: 0.3702 - val_accuracy: 0.8350 - val_auc: 0.9077 - val_precision: 0.7922 - val_recall: 0.7524\n",
            "Epoch 34/100\n",
            "5/5 - 0s - loss: 0.2746 - accuracy: 0.9138 - auc: 0.9657 - precision: 0.9350 - recall: 0.8286 - val_loss: 0.3715 - val_accuracy: 0.8347 - val_auc: 0.9073 - val_precision: 0.7926 - val_recall: 0.7506\n",
            "Epoch 35/100\n",
            "5/5 - 0s - loss: 0.2681 - accuracy: 0.9132 - auc: 0.9665 - precision: 0.9367 - recall: 0.8252 - val_loss: 0.3731 - val_accuracy: 0.8330 - val_auc: 0.9071 - val_precision: 0.7924 - val_recall: 0.7447\n",
            "Epoch 36/100\n",
            "5/5 - 0s - loss: 0.2592 - accuracy: 0.9170 - auc: 0.9690 - precision: 0.9387 - recall: 0.8339 - val_loss: 0.3752 - val_accuracy: 0.8332 - val_auc: 0.9066 - val_precision: 0.7954 - val_recall: 0.7407\n",
            "Epoch 37/100\n",
            "5/5 - 0s - loss: 0.2556 - accuracy: 0.9204 - auc: 0.9694 - precision: 0.9424 - recall: 0.8400 - val_loss: 0.3777 - val_accuracy: 0.8333 - val_auc: 0.9062 - val_precision: 0.7943 - val_recall: 0.7429\n",
            "Epoch 38/100\n",
            "5/5 - 0s - loss: 0.2516 - accuracy: 0.9220 - auc: 0.9705 - precision: 0.9474 - recall: 0.8394 - val_loss: 0.3801 - val_accuracy: 0.8333 - val_auc: 0.9058 - val_precision: 0.7915 - val_recall: 0.7474\n",
            "Epoch 39/100\n",
            "5/5 - 0s - loss: 0.2456 - accuracy: 0.9237 - auc: 0.9721 - precision: 0.9478 - recall: 0.8438 - val_loss: 0.3834 - val_accuracy: 0.8325 - val_auc: 0.9054 - val_precision: 0.7845 - val_recall: 0.7560\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "W8pDOGavCLYy",
        "outputId": "2eed00b1-d052-4084-f41f-8b31ac07902d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 316
        }
      },
      "source": [
        "model_DNN.summary()"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense (Dense)                (None, 64)                1056448   \n",
            "_________________________________________________________________\n",
            "dropout (Dropout)            (None, 64)                0         \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 16)                1040      \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 16)                0         \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 1)                 17        \n",
            "=================================================================\n",
            "Total params: 1,057,505\n",
            "Trainable params: 1,057,505\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NA9lzuLNnha_"
      },
      "source": [
        "history_data = pd.DataFrame(history.history)"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DwS-mHlfvBil",
        "outputId": "42eb47d1-7786-438e-bea9-49a7dd430bd1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "history_data"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>loss</th>\n",
              "      <th>accuracy</th>\n",
              "      <th>auc</th>\n",
              "      <th>precision</th>\n",
              "      <th>recall</th>\n",
              "      <th>val_loss</th>\n",
              "      <th>val_accuracy</th>\n",
              "      <th>val_auc</th>\n",
              "      <th>val_precision</th>\n",
              "      <th>val_recall</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.690659</td>\n",
              "      <td>0.534333</td>\n",
              "      <td>0.505201</td>\n",
              "      <td>0.381262</td>\n",
              "      <td>0.380812</td>\n",
              "      <td>0.677900</td>\n",
              "      <td>0.629167</td>\n",
              "      <td>0.713398</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.676566</td>\n",
              "      <td>0.592056</td>\n",
              "      <td>0.542218</td>\n",
              "      <td>0.416520</td>\n",
              "      <td>0.209151</td>\n",
              "      <td>0.664296</td>\n",
              "      <td>0.629167</td>\n",
              "      <td>0.805493</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.667205</td>\n",
              "      <td>0.603056</td>\n",
              "      <td>0.553487</td>\n",
              "      <td>0.404244</td>\n",
              "      <td>0.115277</td>\n",
              "      <td>0.652508</td>\n",
              "      <td>0.629167</td>\n",
              "      <td>0.845346</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.658539</td>\n",
              "      <td>0.612000</td>\n",
              "      <td>0.567950</td>\n",
              "      <td>0.421488</td>\n",
              "      <td>0.082804</td>\n",
              "      <td>0.642341</td>\n",
              "      <td>0.629500</td>\n",
              "      <td>0.864246</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000899</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.649432</td>\n",
              "      <td>0.618722</td>\n",
              "      <td>0.597385</td>\n",
              "      <td>0.457282</td>\n",
              "      <td>0.069520</td>\n",
              "      <td>0.632801</td>\n",
              "      <td>0.630333</td>\n",
              "      <td>0.876934</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.003146</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>0.638324</td>\n",
              "      <td>0.626944</td>\n",
              "      <td>0.632169</td>\n",
              "      <td>0.535629</td>\n",
              "      <td>0.066568</td>\n",
              "      <td>0.622918</td>\n",
              "      <td>0.634167</td>\n",
              "      <td>0.883068</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.013483</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>0.628236</td>\n",
              "      <td>0.643778</td>\n",
              "      <td>0.669072</td>\n",
              "      <td>0.686154</td>\n",
              "      <td>0.098745</td>\n",
              "      <td>0.612213</td>\n",
              "      <td>0.650167</td>\n",
              "      <td>0.887703</td>\n",
              "      <td>0.970149</td>\n",
              "      <td>0.058427</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>0.613589</td>\n",
              "      <td>0.670722</td>\n",
              "      <td>0.718804</td>\n",
              "      <td>0.796919</td>\n",
              "      <td>0.167970</td>\n",
              "      <td>0.600333</td>\n",
              "      <td>0.686167</td>\n",
              "      <td>0.891099</td>\n",
              "      <td>0.964674</td>\n",
              "      <td>0.159551</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>0.598748</td>\n",
              "      <td>0.697611</td>\n",
              "      <td>0.767773</td>\n",
              "      <td>0.832004</td>\n",
              "      <td>0.246347</td>\n",
              "      <td>0.586700</td>\n",
              "      <td>0.726000</td>\n",
              "      <td>0.893330</td>\n",
              "      <td>0.920405</td>\n",
              "      <td>0.285843</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>0.585128</td>\n",
              "      <td>0.721056</td>\n",
              "      <td>0.797099</td>\n",
              "      <td>0.843383</td>\n",
              "      <td>0.317934</td>\n",
              "      <td>0.571071</td>\n",
              "      <td>0.755500</td>\n",
              "      <td>0.895172</td>\n",
              "      <td>0.894792</td>\n",
              "      <td>0.386067</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>0.567026</td>\n",
              "      <td>0.744778</td>\n",
              "      <td>0.828410</td>\n",
              "      <td>0.857424</td>\n",
              "      <td>0.386125</td>\n",
              "      <td>0.553866</td>\n",
              "      <td>0.777333</td>\n",
              "      <td>0.896559</td>\n",
              "      <td>0.888210</td>\n",
              "      <td>0.457079</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>0.551502</td>\n",
              "      <td>0.760500</td>\n",
              "      <td>0.846995</td>\n",
              "      <td>0.874924</td>\n",
              "      <td>0.424354</td>\n",
              "      <td>0.536030</td>\n",
              "      <td>0.789833</td>\n",
              "      <td>0.897844</td>\n",
              "      <td>0.877743</td>\n",
              "      <td>0.503371</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>0.530993</td>\n",
              "      <td>0.779500</td>\n",
              "      <td>0.866365</td>\n",
              "      <td>0.894323</td>\n",
              "      <td>0.469668</td>\n",
              "      <td>0.517910</td>\n",
              "      <td>0.799333</td>\n",
              "      <td>0.899026</td>\n",
              "      <td>0.871273</td>\n",
              "      <td>0.538427</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>0.511957</td>\n",
              "      <td>0.793389</td>\n",
              "      <td>0.879244</td>\n",
              "      <td>0.901050</td>\n",
              "      <td>0.506716</td>\n",
              "      <td>0.500185</td>\n",
              "      <td>0.806000</td>\n",
              "      <td>0.900057</td>\n",
              "      <td>0.858204</td>\n",
              "      <td>0.571236</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>0.493087</td>\n",
              "      <td>0.810500</td>\n",
              "      <td>0.891680</td>\n",
              "      <td>0.902008</td>\n",
              "      <td>0.557048</td>\n",
              "      <td>0.483113</td>\n",
              "      <td>0.812667</td>\n",
              "      <td>0.901092</td>\n",
              "      <td>0.841713</td>\n",
              "      <td>0.609438</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>0.478863</td>\n",
              "      <td>0.821000</td>\n",
              "      <td>0.898296</td>\n",
              "      <td>0.896098</td>\n",
              "      <td>0.593210</td>\n",
              "      <td>0.466896</td>\n",
              "      <td>0.817000</td>\n",
              "      <td>0.902128</td>\n",
              "      <td>0.830499</td>\n",
              "      <td>0.636405</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>0.461133</td>\n",
              "      <td>0.831167</td>\n",
              "      <td>0.905182</td>\n",
              "      <td>0.896771</td>\n",
              "      <td>0.623173</td>\n",
              "      <td>0.451656</td>\n",
              "      <td>0.821333</td>\n",
              "      <td>0.903331</td>\n",
              "      <td>0.825522</td>\n",
              "      <td>0.657079</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>0.444526</td>\n",
              "      <td>0.838222</td>\n",
              "      <td>0.912700</td>\n",
              "      <td>0.902144</td>\n",
              "      <td>0.639557</td>\n",
              "      <td>0.437975</td>\n",
              "      <td>0.825000</td>\n",
              "      <td>0.904261</td>\n",
              "      <td>0.822271</td>\n",
              "      <td>0.673708</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>0.428589</td>\n",
              "      <td>0.847000</td>\n",
              "      <td>0.920207</td>\n",
              "      <td>0.902180</td>\n",
              "      <td>0.665683</td>\n",
              "      <td>0.425543</td>\n",
              "      <td>0.828000</td>\n",
              "      <td>0.905130</td>\n",
              "      <td>0.823306</td>\n",
              "      <td>0.682697</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>0.413435</td>\n",
              "      <td>0.850944</td>\n",
              "      <td>0.924076</td>\n",
              "      <td>0.901491</td>\n",
              "      <td>0.678081</td>\n",
              "      <td>0.414602</td>\n",
              "      <td>0.828500</td>\n",
              "      <td>0.905981</td>\n",
              "      <td>0.820129</td>\n",
              "      <td>0.688539</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>0.395406</td>\n",
              "      <td>0.860778</td>\n",
              "      <td>0.932317</td>\n",
              "      <td>0.906185</td>\n",
              "      <td>0.702878</td>\n",
              "      <td>0.405154</td>\n",
              "      <td>0.831000</td>\n",
              "      <td>0.906561</td>\n",
              "      <td>0.813893</td>\n",
              "      <td>0.705618</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>0.386453</td>\n",
              "      <td>0.868944</td>\n",
              "      <td>0.935491</td>\n",
              "      <td>0.914259</td>\n",
              "      <td>0.719262</td>\n",
              "      <td>0.397075</td>\n",
              "      <td>0.831333</td>\n",
              "      <td>0.907014</td>\n",
              "      <td>0.809913</td>\n",
              "      <td>0.712360</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>0.375063</td>\n",
              "      <td>0.872222</td>\n",
              "      <td>0.936952</td>\n",
              "      <td>0.911230</td>\n",
              "      <td>0.731808</td>\n",
              "      <td>0.390163</td>\n",
              "      <td>0.830667</td>\n",
              "      <td>0.907412</td>\n",
              "      <td>0.810478</td>\n",
              "      <td>0.709213</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>0.364304</td>\n",
              "      <td>0.873944</td>\n",
              "      <td>0.941261</td>\n",
              "      <td>0.911883</td>\n",
              "      <td>0.736236</td>\n",
              "      <td>0.384556</td>\n",
              "      <td>0.831167</td>\n",
              "      <td>0.907773</td>\n",
              "      <td>0.812371</td>\n",
              "      <td>0.708315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>0.352929</td>\n",
              "      <td>0.880333</td>\n",
              "      <td>0.942788</td>\n",
              "      <td>0.918493</td>\n",
              "      <td>0.748487</td>\n",
              "      <td>0.379885</td>\n",
              "      <td>0.831667</td>\n",
              "      <td>0.908083</td>\n",
              "      <td>0.808846</td>\n",
              "      <td>0.715056</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>0.338812</td>\n",
              "      <td>0.888111</td>\n",
              "      <td>0.948258</td>\n",
              "      <td>0.920657</td>\n",
              "      <td>0.769004</td>\n",
              "      <td>0.376251</td>\n",
              "      <td>0.833167</td>\n",
              "      <td>0.908222</td>\n",
              "      <td>0.803873</td>\n",
              "      <td>0.727640</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>0.331763</td>\n",
              "      <td>0.888167</td>\n",
              "      <td>0.949570</td>\n",
              "      <td>0.916113</td>\n",
              "      <td>0.773727</td>\n",
              "      <td>0.373435</td>\n",
              "      <td>0.833167</td>\n",
              "      <td>0.908375</td>\n",
              "      <td>0.799413</td>\n",
              "      <td>0.734382</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>0.323417</td>\n",
              "      <td>0.895167</td>\n",
              "      <td>0.953159</td>\n",
              "      <td>0.924158</td>\n",
              "      <td>0.785978</td>\n",
              "      <td>0.371334</td>\n",
              "      <td>0.834333</td>\n",
              "      <td>0.908357</td>\n",
              "      <td>0.801864</td>\n",
              "      <td>0.734831</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>0.311626</td>\n",
              "      <td>0.898167</td>\n",
              "      <td>0.955876</td>\n",
              "      <td>0.925741</td>\n",
              "      <td>0.793063</td>\n",
              "      <td>0.369863</td>\n",
              "      <td>0.835667</td>\n",
              "      <td>0.908419</td>\n",
              "      <td>0.798842</td>\n",
              "      <td>0.744270</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>0.299432</td>\n",
              "      <td>0.902278</td>\n",
              "      <td>0.959770</td>\n",
              "      <td>0.931224</td>\n",
              "      <td>0.799410</td>\n",
              "      <td>0.369071</td>\n",
              "      <td>0.835333</td>\n",
              "      <td>0.908250</td>\n",
              "      <td>0.798360</td>\n",
              "      <td>0.743820</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30</th>\n",
              "      <td>0.297893</td>\n",
              "      <td>0.903833</td>\n",
              "      <td>0.958708</td>\n",
              "      <td>0.931259</td>\n",
              "      <td>0.803838</td>\n",
              "      <td>0.368966</td>\n",
              "      <td>0.833500</td>\n",
              "      <td>0.908184</td>\n",
              "      <td>0.800196</td>\n",
              "      <td>0.734382</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31</th>\n",
              "      <td>0.291500</td>\n",
              "      <td>0.907667</td>\n",
              "      <td>0.960539</td>\n",
              "      <td>0.938132</td>\n",
              "      <td>0.807970</td>\n",
              "      <td>0.369307</td>\n",
              "      <td>0.835500</td>\n",
              "      <td>0.907925</td>\n",
              "      <td>0.797310</td>\n",
              "      <td>0.746067</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>0.280598</td>\n",
              "      <td>0.913500</td>\n",
              "      <td>0.963810</td>\n",
              "      <td>0.937311</td>\n",
              "      <td>0.825387</td>\n",
              "      <td>0.370199</td>\n",
              "      <td>0.835000</td>\n",
              "      <td>0.907664</td>\n",
              "      <td>0.792239</td>\n",
              "      <td>0.752360</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>0.274606</td>\n",
              "      <td>0.913833</td>\n",
              "      <td>0.965711</td>\n",
              "      <td>0.935043</td>\n",
              "      <td>0.828635</td>\n",
              "      <td>0.371474</td>\n",
              "      <td>0.834667</td>\n",
              "      <td>0.907325</td>\n",
              "      <td>0.792596</td>\n",
              "      <td>0.750562</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34</th>\n",
              "      <td>0.268089</td>\n",
              "      <td>0.913222</td>\n",
              "      <td>0.966477</td>\n",
              "      <td>0.936673</td>\n",
              "      <td>0.825240</td>\n",
              "      <td>0.373108</td>\n",
              "      <td>0.833000</td>\n",
              "      <td>0.907057</td>\n",
              "      <td>0.792444</td>\n",
              "      <td>0.744719</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>0.259232</td>\n",
              "      <td>0.917000</td>\n",
              "      <td>0.968989</td>\n",
              "      <td>0.938694</td>\n",
              "      <td>0.833948</td>\n",
              "      <td>0.375232</td>\n",
              "      <td>0.833167</td>\n",
              "      <td>0.906602</td>\n",
              "      <td>0.795367</td>\n",
              "      <td>0.740674</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36</th>\n",
              "      <td>0.255641</td>\n",
              "      <td>0.920444</td>\n",
              "      <td>0.969406</td>\n",
              "      <td>0.942375</td>\n",
              "      <td>0.840000</td>\n",
              "      <td>0.377658</td>\n",
              "      <td>0.833333</td>\n",
              "      <td>0.906199</td>\n",
              "      <td>0.794330</td>\n",
              "      <td>0.742921</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37</th>\n",
              "      <td>0.251580</td>\n",
              "      <td>0.922000</td>\n",
              "      <td>0.970536</td>\n",
              "      <td>0.947360</td>\n",
              "      <td>0.839410</td>\n",
              "      <td>0.380139</td>\n",
              "      <td>0.833333</td>\n",
              "      <td>0.905768</td>\n",
              "      <td>0.791528</td>\n",
              "      <td>0.747416</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>38</th>\n",
              "      <td>0.245558</td>\n",
              "      <td>0.923722</td>\n",
              "      <td>0.972098</td>\n",
              "      <td>0.947779</td>\n",
              "      <td>0.843838</td>\n",
              "      <td>0.383351</td>\n",
              "      <td>0.832500</td>\n",
              "      <td>0.905353</td>\n",
              "      <td>0.784515</td>\n",
              "      <td>0.755955</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        loss  accuracy       auc  ...   val_auc  val_precision  val_recall\n",
              "0   0.690659  0.534333  0.505201  ...  0.713398       0.000000    0.000000\n",
              "1   0.676566  0.592056  0.542218  ...  0.805493       0.000000    0.000000\n",
              "2   0.667205  0.603056  0.553487  ...  0.845346       0.000000    0.000000\n",
              "3   0.658539  0.612000  0.567950  ...  0.864246       1.000000    0.000899\n",
              "4   0.649432  0.618722  0.597385  ...  0.876934       1.000000    0.003146\n",
              "5   0.638324  0.626944  0.632169  ...  0.883068       1.000000    0.013483\n",
              "6   0.628236  0.643778  0.669072  ...  0.887703       0.970149    0.058427\n",
              "7   0.613589  0.670722  0.718804  ...  0.891099       0.964674    0.159551\n",
              "8   0.598748  0.697611  0.767773  ...  0.893330       0.920405    0.285843\n",
              "9   0.585128  0.721056  0.797099  ...  0.895172       0.894792    0.386067\n",
              "10  0.567026  0.744778  0.828410  ...  0.896559       0.888210    0.457079\n",
              "11  0.551502  0.760500  0.846995  ...  0.897844       0.877743    0.503371\n",
              "12  0.530993  0.779500  0.866365  ...  0.899026       0.871273    0.538427\n",
              "13  0.511957  0.793389  0.879244  ...  0.900057       0.858204    0.571236\n",
              "14  0.493087  0.810500  0.891680  ...  0.901092       0.841713    0.609438\n",
              "15  0.478863  0.821000  0.898296  ...  0.902128       0.830499    0.636405\n",
              "16  0.461133  0.831167  0.905182  ...  0.903331       0.825522    0.657079\n",
              "17  0.444526  0.838222  0.912700  ...  0.904261       0.822271    0.673708\n",
              "18  0.428589  0.847000  0.920207  ...  0.905130       0.823306    0.682697\n",
              "19  0.413435  0.850944  0.924076  ...  0.905981       0.820129    0.688539\n",
              "20  0.395406  0.860778  0.932317  ...  0.906561       0.813893    0.705618\n",
              "21  0.386453  0.868944  0.935491  ...  0.907014       0.809913    0.712360\n",
              "22  0.375063  0.872222  0.936952  ...  0.907412       0.810478    0.709213\n",
              "23  0.364304  0.873944  0.941261  ...  0.907773       0.812371    0.708315\n",
              "24  0.352929  0.880333  0.942788  ...  0.908083       0.808846    0.715056\n",
              "25  0.338812  0.888111  0.948258  ...  0.908222       0.803873    0.727640\n",
              "26  0.331763  0.888167  0.949570  ...  0.908375       0.799413    0.734382\n",
              "27  0.323417  0.895167  0.953159  ...  0.908357       0.801864    0.734831\n",
              "28  0.311626  0.898167  0.955876  ...  0.908419       0.798842    0.744270\n",
              "29  0.299432  0.902278  0.959770  ...  0.908250       0.798360    0.743820\n",
              "30  0.297893  0.903833  0.958708  ...  0.908184       0.800196    0.734382\n",
              "31  0.291500  0.907667  0.960539  ...  0.907925       0.797310    0.746067\n",
              "32  0.280598  0.913500  0.963810  ...  0.907664       0.792239    0.752360\n",
              "33  0.274606  0.913833  0.965711  ...  0.907325       0.792596    0.750562\n",
              "34  0.268089  0.913222  0.966477  ...  0.907057       0.792444    0.744719\n",
              "35  0.259232  0.917000  0.968989  ...  0.906602       0.795367    0.740674\n",
              "36  0.255641  0.920444  0.969406  ...  0.906199       0.794330    0.742921\n",
              "37  0.251580  0.922000  0.970536  ...  0.905768       0.791528    0.747416\n",
              "38  0.245558  0.923722  0.972098  ...  0.905353       0.784515    0.755955\n",
              "\n",
              "[39 rows x 10 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pjFY-enb2E7-",
        "outputId": "0494ac93-cc14-48e3-bc8d-9c467d7747eb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "# Plot training & validation accuracy values\n",
        "plt.plot(history.history['accuracy'])\n",
        "plt.plot(history.history['val_accuracy'])\n",
        "plt.title('Model accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Test'], loc='upper left')\n",
        "plt.show()\n",
        "\n",
        "# Plot training & validation loss values\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.title('Model loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Test'], loc='upper left')\n",
        "plt.show()\n",
        "\n",
        "\n",
        "plt.plot(history.history['recall'])\n",
        "plt.plot(history.history['val_recall'])\n",
        "plt.title('Model Recall')\n",
        "plt.ylabel('Recall')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Test'], loc='upper left')\n",
        "plt.show()\n",
        "\n",
        "plt.plot(history.history['precision'])\n",
        "plt.plot(history.history['val_precision'])\n",
        "plt.title('Model Precision')\n",
        "plt.ylabel('Precision')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Test'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deZhcZZX48e+p6q7e051esnRnJRtZWAIRZJMdg6wqw4AwCjqD+BPQQYcBF0RGHdTRURxGRA0BFBFZNCwOsoVdyEJYsu9Jd5Zeku703rWc3x/3dqfS6aSruut2VVedz/Pcp+69devW6XqSOvW+5973FVXFGGNM5vIlOwBjjDHJZYnAGGMynCUCY4zJcJYIjDEmw1kiMMaYDGeJwBhjMpwlApMxRGSSiKiIZMVw7DUi8vpQxHWYGO4VkW/HcNxKETljCEIyacoSgUlJIrJFRLpEpLzX/nfdL/NJyYnsgITS4i5bROTWRL+Pql6vqv8Rw3GzVXVxot/fZA5LBCaVbQau7N4QkaOA/OSFc5ASVS3EifF2EZnf+4BYWh/GJJslApPKHgI+G7X9OeDB6ANEpFhEHhSROhHZKiLfEhGf+5xfRP5LROpFZBNwQR+v/a2I7BSRGhH5noj44w1SVd8CVgJzROQMEakWkX8XkV3A/SLiE5FbRWSjiDSIyKMiUhoVx6ki8qaINIrIdhG5xt2/UES+566Xi8jT7jF7ROS1qL9zi4ic467niMjPRGSHu/xMRHLc57pj+5qI1Lp/97Xx/r0m/VgiMKns78AIEZnpfkFfAfyu1zG/AIqBI4DTcRJH95fbvwAXAnOBecBlvV67EAgBU91jzgP+OZ4AxXEKMBt41909BigFJgLXATcCl7rxVQJ7gXvc108E/ur+HRXAscCKPt7qa0C1e8xo4BtAX+PDfBP4qHueY4ATgG9FPT8G5/OqAr4A3CMiI+P5m036sURgUl13q+BcYDVQ0/1EVHK4TVWbVXUL8BPgn9xDLgd+pqrbVXUP8J9Rrx0NfAL4qqq2qmot8N/u+WJVD+wBfgPcqqovuvsjwHdUtVNV24HrgW+qarWqdgJ3AJe53UafAV5Q1T+oalBVG1S1r0QQBMYCE93jXtO+Bwq7CrhTVWtVtQ74btTn0X2eO91zPAu0ADPi+JtNGrL+S5PqHgJeBSbTq1sIKAeyga1R+7bi/NoF59f39l7PdZvovnaniHTv8/U6vj/lqhrqY3+dqnb0eq8nRSQStS+M88t+PLAxhvf6MU4C+Zsb732qelcfx1Vy8OdRGbXd0CvmNqAwhvc3acxaBCalqepWnKLxJ4Anej1dj/MLd2LUvgnsbzXsxPmijX6u23agE+fLvMRdRqjq7ESE3Wt7O3B+1PuUqGquqta4z03p94ROi+drqnoEcDFws4ic3cehOzj489gxsD/DZApLBGY4+AJwlqq2Ru9U1TDwKPB9ESly+9tvZn8d4VHgJhEZ5/aD3xr12p3A34CfiMgIt6A7RURO9yD+e90YJwKISIWIXOI+93vgHBG5XESyRKRMRI7tfQIRuVBEporTHGjCaVFEeh8H/AH4lvse5cDtHFxXMeYAlghMylPVjaq69BBP3wi0ApuA14GHgQXuc78GngPeA5ZzcIvis0AAWIVTwH0Mpx8+0X4OLMLp1mnGKYKfCKCq23BaO1/DqTeswCny9jYNeAGnT/8t4H9V9eU+jvsesBR4H/gA5+/+XiL/GJN+xCamMcaYzGYtAmOMyXCWCIwxJsNZIjDGmAxnicAYYzLcsLuhrLy8XCdNmpTsMIwxZlhZtmxZvapW9PXcsEsEkyZNYunSQ11JaIwxpi8isvVQz1nXkDHGZDhLBMYYk+EsERhjTIYbdjWCvgSDQaqrq+no6Oj/4GEuNzeXcePGkZ2dnexQjDFpIi0SQXV1NUVFRUyaNImoIYXTjqrS0NBAdXU1kydPTnY4xpg04VnXkIgscKfD+/AQz4uI3C0iG0TkfRE5bqDv1dHRQVlZWVonAQARoaysLCNaPsaYoeNljWAhcNBk3lHOxxlRcRrOdH6/HMybpXsS6JYpf6cxZuh41jWkqq+KyKTDHHIJ8KA73d7fRaRERMa648QPP+EgtDXAYEZz9fkhvxx8VsM3xgydZNYIqjhwWsBqd99BiUBErsNpNTBhwoTeTyddQ0MDZ595OoSD7KprwO/3UVHqzAf+zjMPEQgcurC79L1VPPjY09z9H7c4O9r2wMhJkJ07BJEbY8wwKRar6n3AfQDz5s1LuQkUysrKWPHGC9C8kzt+9WcKi4r4+te/3vN8KBQiK6vvj3pe5VzmnX+Vs9HRBHu3Qv1aKB4P+aVDEb4xJsMlsw+ihgPnkx3H/rlmhx+NAAJuH/4111zD9ddfz4knnsgtt9zCO++8w0knncTcuXM5+eSTWbt2LQCLFy/mwgsvBOCOu/6bz9/2U8647DqOmHk0d991B0TCSfqDjDGZIpktgkXADSLyCM60fU2JqA9896mVrNqxb9DBRZtVOYLvXNTPnOYaATkwr1ZXV/Pmm2/i9/vZt28fr732GllZWbzwwgt84xvf4PHHHz/oNGvWrePll16heecGZhx3Kl+6+pNkj54G2fmJ/JOMMaaHZ4lARP4AnAGUi0g18B0gG0BV7wWexZmrdQPQBlzrVSxDoo9E8A//8A/4/X4Ampqa+NznPsf69esREYLBYJ+nueCCC8jJzSVn8hxGjR7N7to6xvkUiqucQrJdNWSMSTAvrxq6sp/nFfhyot+331/uXtHIQV/SBQUFPevf/va3OfPMM3nyySfZsmULZ5xxRp+nycnJ6Vn3Z2UTKp4IOT5oqnZqCIEiCLbBzvdg5GTIHeHJn2OMyRzDolg8LPTRIojW1NREVVUVAAsXLoz9vP5sKJ0IrbXQUgudzdBaD09c7jyfX+ZcZTRyMow9Bj76Jec1xhgTI7tgPVH6SQS33HILt912G3PnziUUCsV3bhEoHA1jjnKWojFw+YNwzndh5kWQUwTVS+D5b8OimyASGeQfY4zJJKKDuQEqCebNm6e9J6ZZvXo1M2fOTFJErrp1zhd2+TTP3+qQf+/iH8LiH8DJN8J53/M8DmPM8CEiy1R1Xl/PWddQomgEfEnukjn9Fmirhzd/AQUVcMpXkhuPMSYmqsqe1i46QxHCEUUVwqruuhJWJRKBMcW5lBYEEv7+lggSpZ+uoSEhAvN/6Ax18fztTv1g7tXJjckYQ1coQkNrJzsa26neu3+paWynem8bNXvb6Qz136X7vUvncPVHJyY8PksEiZIKiQCccYouvRfa9zr1grxSOPITyY7KGM+oKrv3dbKpvgVVyPIJWX4fAb+PLL+Q7ReyfD6ys3wUBrIoyPGT5Y///6qq0hGM0NoVorUzREtniNbO8P7tjhANrV3UNXdS39K9dFHf0klj28GXi5cVBBg3Mo8jxxRxzszRjC3OJS/bj88n+EXw+cAngk8Ev895nF3pzVWClggSRSOpM1hcVgAufwgevBgeuxaufgImnZLsqIwZFFWlvqWL9bubWbe7mbW7W1i/u5m1u5tp7ojvAoy8bD9FuVkU5mZRlOM8FgSyCIYjtAfDtAcjtHeFnPWuCB3BMG1dISIxlFQLc7KoKMqhvDDAtFGFnDyljPLCHMoLc6gsyWXcyDwqS/LID6TO12/qRDLcpUqLoFtOIXzmT3D/fPjDlXDtM84VR8YkkapS29zJtj1tbG1oY1tDq7O+p40dje2Ewvu/aXt/5wZDEZo793/hl+RnM310EZccW8mM0UVMqSgky+8jGI4QDEcIhdVZjyihcISuUITWrjAtHSFaOoM0d4Rodn/Jt3SGaGhpI9vvIy/bT0leNmNH5JIX8JOb7Sc/4Cc320dBThaFOU7SKMjxU5CTtX9fThZlBQFys/1D9GkmjiWCRFAFNLUSAUBBmdMaWPBx+N2n4fPPQanNbGYSR1Wp3tvOhzVNrNnVTEun8yu6oyvs/rIO094VpiMYpqUzRE1jOx3B/X3hPoGxxXlMLMvnY9MqyMk+8P+QIAccO7GsgOmji5g+ppCKwhybnyNBLBEkQEN9HWefewX4s9lVW4/f76eiogKAd955h0Dg8FX+xYsXEwgEOPnkkxMfXMl4JxncPx8euhQuWwBVxyf+fUzK2tPqdKdsaWilqT1IS6fzq7i7n7u501lv7wpTVhigsjiPsSW5jC3OZWxxHpUluYwpzqMg4Gf7nnY+qGnig5omPqxp4sMdTT393yKQn+3v+RWdF7Vekh+gamQeZ84YxcSyfMaX5jOxrICqkjwCWSn2AyoDWSJIgLLSkax4/hEoHscdP76HwsLCA4ah7s/ixYspLCz0JhEAjDoSrnoc/ngV/OYcOOGLcNY3nRvRTFKFwhHW17bwQU0Tm+payQ/4GZmfTUl+gJH5AUrysxlZEKA0P0BeYH+XQziihCIR91EJh5X2YJjN9a2s393MhroW1u9uYUNtCw2tXQe9b2FPd4afwtxsinKyKMnLpr6lkzW7mqlv6TxojqWA30dX2Pk1n+0Xpo8uYv7sMcypKmZOVTFHjikalt0ixhJBYqjb1I3qGlq2bBk333wzLS0tlJeXs3DhQsaOHcvdd9/NvffeS1ZWFrNmzeKuu+7i3nvvxe/387vf/Y5f/OIXnHbaaYmPcdzx8OV34MU74e17YfVTcMF/wYzzE/9epk/BcIT1u1v40P1F/UFNE6t37uu5bDDLJ4QOU43M9jvdICH3OvPDGZGbxbTRRZw7azRTRxUydVQhUyoKGVkQIN+9MuVwukIRdu/rYGdTBzub2tnZ1EFDSyeTygs4qqqYGWOKyMmyL/10kX6J4K+3wq4PEnvOMUfB+Xcd+vleiUBVufHGG/nLX/5CRUUFf/zjH/nmN7/JggULuOuuu9i8eTM5OTk0NjZSUlLC9ddfH3crYkByRzhf/kdfDk99Bf5wBcy6BM7/kTNsRYpQVYLdhb5whK5whKBbRBxdlDOgS//609oZYmdTB5Gob9i+vmz9PnEvT3QuSfT7nMsT/T5Bgdp9HdQ0drCzsZ0dje3OepOzvqOxo+cXdUHAz+yqYq7+6ESOcn9RH1FeQCiiNLZ30dgWZE9rF41tXextC7K3rYt97SF84iQMv8/nxiA9MQWy/Ewqy2fq6MH3nweyfIwvdbpwTPpLv0SQDL0SQWdnJx9++CHnnnsuAOFwmLFjxwJw9NFHc9VVV3HppZdy6aWXJiVcxp8A170Cb94Nr/wINi6Gc74Dx187JJfAhiPKtj1trN/dzPraFvdywBa272mjMxTp+bLsS7ZfGF+az+SyAiaVFzDZXSaVFzB2RC4+n/Qkku7uk1DY6T7pCIap3tvO9j1tbItatu9p67P7ZLB8AqNHOH3tc6qK+ficMcwaO4I5VcVMLivo81d5wCeMKsplVJFNVWqGTvolgsP9cvdKHy2C2bNn89Zbbx106DPPPMOrr77KU089xfe//30++CDBrZdYZQXgY1+H2Z+Ep78Kz9wMyx+EyR+DihlQcSSUT0/IMNeqyhsbGnh8eTVrdjWzsa6Frqi7KKtK8pg6qpATJpeSF/CT7fcR8AvZfp+zZDnb4Qhs39vG5rpWtjS08sbG+gOuQPG7SSCWa739PqGyJJcJpfmcN3s040vzqSzOI7tXayP6R3X3bf+hcMTpl48cuK4Ko0bkUFmSx9jiXEaPyD3ofMakovRLBMnQKxHk5ORQV1fHW2+9xUknnUQwGGTdunXMnDmT7du3c+aZZ3LqqafyyCOP0NLSQlFREfv2JXZWtZiVTYHPLoIVD8Pff+nUD8JRv46LKt3EMANGzYSJpzqviaHbIRJRXli9m3sWb+S97Y2UFgQ4elwxp00rZ+qoQqaPLmLqqEIKcwb2zzASUXbt62BLfSubG1rZ0djecxdm992lB3ad+KgqyWdCaT5jS+xL2phulggSoVci8Pl8PPbYY9x00000NTURCoX46le/yvTp07n66qtpampCVbnpppsoKSnhoosu4rLLLuMvf/mLd8XiwxGBuVc5SzgEjVuhbg3UrXWXNbD8IQi2OsePqHJaDt1L8bgDThcKR3jmg53878sbWbu7mQml+fzgk0fx6eOrElpg9PmEyhLnLs2Tp5Yn7LzGZBobhjoR2hqgcRuMmgVZOf0fP0hJ+XsjEdizCba8Cpvdpa3Bea70CJj8MYLjT+b5+jJ+ujzChj1Bpo0q5MtnTuXCo8d6UuA1xsTOhqH2Wh+Xj6Ydnw/KpzrLvM87iaFuNWx6ha4Ni9EVj5GzbCGfAD6Oj47y8eSPmYnUTYf3Z0C5271kU2sak3IsESRCJiSC3nw+1uoEFu48lSfXTSYYvJrLxjdzzbROjszaQUH9OqhfDxtf2l9zEB9MPMWZVe3IC6G4Krl/gzEGSKNEoKrJG3dkCBNBsrvywhHlpTW13P/GZt7c2EBOlo9Pzq3imlMmceSYPn7t99Qc1sKO5bD6afjrLc5SNQ9mXewkhtIjhv6PMcYAaVIj2Lx5M0VFRZSVlSUnGeyrgZY6qDzW07dRVRoaGmhubmby5KEdPK6hpZMn363hwbe2sm1PG5XFufzTSZO44iPjGRnvjEn162H1Ili1CHaucPaNPspJCsd+5qDiszFm8A5XI0iLRBAMBqmurqajoyM5QbXvha7WIfkCy83NZdy4cWRnez8tZnNHkOdW7uap93bw+oZ6whHlI5NGcu0pkzlv1ujEFIAbtznDXax+Crb93bmCafp8pw4x5Szw2TAGxiRC2ieCpPvLl2Hjy3DzqmRHMmgdwTAvrall0YodvLS2lq5QhHEj87jomEouObay7+6fRNm7FZY/4NzY1loHJRPg+Gtg7j9B4Sjv3teYDGBXDXmtqw2y85IdxYA1tQV5Y2M9L6zazd9W7aalM0RFUQ6fOWECFx9bydzxJUPT5TZyIpx9O5x+K6x5GpYucAbJe/k/nTrC8dc4xWa//bM1JpHsf1QiBNshe/gMzhWJKB/UNPHKujpeWVfHu9v2ElEozsvmwqPHctExlXz0iDL8/YxQ6ZmsAMz5lLPUrYNlC2HF72HlE5BTDEd8DKacDVPPdloNxphBsUSQCMHWlE8EnaEwz7y/k1fW1fHa+nr2tHYhAkdXFXPDmVM5fUYFx4wrSb0bvyqmw/wfwNnfhnXPwcYXYcNLTk0BoGyakxCmnAWTToVAQXLjNWYYskSQCMH2lJ7kZW9rF9c9tJQlW/ZSXhjgjBkVnD69gtOmVVAa7xU/yZKdB7MvdRZVqF8HG150EsOyB5wxksQPIyqdVkLxeOexZPz+7eJxQ3LntzHDjSWCRAi2Q+HoZEfRpy31rVy7cAk1je38/Ipjuejoyn4nJUl5IvsHwjvp/0GwA7a9CVvfcq5CatwGW9+ADx7df49Ht5xiZy7n/DLIL3ceu7dzS0DDEOpyboILd0atd0EkDHkl7mv7WLJt6GgzPFkiSISu1pQsFi/buod/eXAZqsrD/3wi8yaVJjskb2TnOl1DU846cH84CPt2OImhaTs0VUNrPbTVO+MkNVXDzvec7fBh5iPw54A/4Nww2Nl06OMCRVA62Rm+u3walE111sumQiC1uw5NZrNEkAgpWCx+5v2d/OujK6gszmXhtScwqTwD+8792c6VSCMnHv44VehqgfZG5zX+QNSSfeCQ2+EQdDQ6iaT30lILDRugegl8+DgQdWl28Xhn+O6iSigaDYVjDn7sK1moQiTktEYiQecKta4W58dHz+JuhzogbyQUVDiX2xZUOK2cIZhsyAxvniYCEZkP/BzwA79R1bt6PT8BeAAocY+5VVWf9TImT6RQIlBVfvXqJu766xrmTRzJfZ+dN3zqAMki4tR4Yqnz+LOgoNxZDifY7ozWWr8O6jc4j3s2Qv0r0LLb+XLvLbvAiaXniz/EAclkIMTvxjsKCiucIcRLJh5YOxlRaTfuZTjPEoGI+IF7gHOBamCJiCxS1ei7rr4FPKqqvxSRWcCzwCSvYvJMMDW6hkLhCLcvWsnDb2/jomMq+fFlR5Obbf/BkyI7D0bPdpbeIhFo3wPNu6Bll/PYvMsd1lucL2WfH3xZzhe5L2v/diAfAoXO1VGBggPX/TnOXe6tdQcvLXVOAtr1IbTWHhiPL8tJBsUTnIEAR1Q6CWNE1Hp+2f6WhaqT6DqaoHOf89ixz2mZ5BRCXqlbNyl14kvUPSiqiTuXOYCXLYITgA2quglARB4BLgGiE4EC3beqFgM7PIzHG+Gg88styX3AbV0hvvS75byyro7/d8YUvn7ejOFfFE5XPl9Uq2JOYs89Ymz/xwTbnfpId+2kcRs0uo/b3oJ9O51uqGj+gNPVFOpwvvj7atH0xZftJIS8Uqfbyp/tJDXxRz36nEfxObEFo7u93K6wYJvz3oFC5zx9Lfml+1s8Iyc6yWioEoeqE6P4nM9qmLWwvEwEVcD2qO1q4MRex9wB/E1EbgQKgHP6OpGIXAdcBzBhQordQNTlztqV5K6hBa9v5pV1dfznp47iyhNS7DMyqSU7zylml0/r+/lIxGlF7Ktxiu37dsC+aqcGkp3vzCmRWww57mP3EiiAzmZo2+O0btr3OOs9j43Ol3kk7FydFQk7V3V1b2vEiS27wKltjKh0vviz851zZ+U6/9/a9zgtn/a9ULtq/3rv5JSd715C7C7F45zWj0b2L5HIgdvgJg85+BGc+lB74/4Y2qJiiU6e4ncSQlZ3rSnHWc8Z4Saukv0JLLdkfyIrm+qMxOv3fiyxaMkuFl8JLFTVn4jIScBDIjJH9cBr/lT1PuA+cMYaSkKchxZsdx6TmAhUlSeW13Di5FJLAmbwfD6ngF00GqqOS3Y0sVF1Wir7avZfQty4zRkCvXGbU8Bv3zv498nKc1s47pd4xYz927klgB54+XE4CKFOZzvU4XShdTQ6LbH2vU5S0fCB7+EPOFebjZrpzHo4apazXjzes8K/l4mgBhgftT3O3RftC8B8AFV9S0RygXKgVydmCgu2OY9JTAQrtjeyqb6V60+fkrQYjEkqEfdXdknfdRlwWhORsNsl5Yta/M7ro7uRVJ2FqEdI/C91VacV1b7XuYy5fgPUroTa1c5ovB/8af+xgUI4/4cw9+rExoC3iWAJME1EJuMkgCuAz/Q6ZhtwNrBQRGYCuUCdhzElXk8iSF6x+Ml3a8jJ8nH+UWOSFoMxKS+e4Ud6JwaviLhdbSOcukbV8Qc+39EEtWucLrDaVU7XkQc8SwSqGhKRG4DncC4NXaCqK0XkTmCpqi4Cvgb8WkT+FSflXqPDbVzs7q6hJBWLu0IRFr23g/Nmj6Eod2j7FY0xHssthgknOouHPK0RuPcEPNtr3+1R66uAU7yMwXNJ7hp6eW0tjW1BPnWczf9rjBkYu+VwsLqS2zX05PIaygtzOG1qPzc4GWPMIVgiGKyeFsHQD+HQ2NbFi2t2c8mxlak3fLQxZtiwb4/BSmKx+Kn3dxIMq3ULGWMGxRLBYCXxPoInl1czY3QRs8Z6OI+wMSbtWSIYrO4WwRBfNbS5vpXl2xr51HFVQzOfsDEmbVkiGKyuNkCc29+H0JPv1uATuHSudQsZYwbHEsFgBducbqEh/FUeiShPLK/mlKnljB5hs2IZYwbHEsFgBduHvFC8dOteqve2W5HYGJMQlggGq7tFMISefLea/ICfj8+2ISWMMYNniWCwgm1DWijuCIZ5+v2dzJ8zhvxAsgePNcakA0sEg9XVNqRdQy+s3k1zR4hPHzduyN7TGJPeLBEM1hDPV/zk8hrGjMjlo0eUDdl7GmPSmyWCwRrCGkF9SyeL19Vx6dwq/DYNpTEmQSwRDFZw6LqGnnpvB+GIDSlhjEksSwSDFWyLb8KLQXhieQ1zqkYwfXTRkLyfMSYzWCIYrCG6j2D97mY+qGniU3OtSGyMSSxLBIPVNTQ1gr9+uAsRuOiYSs/fyxiTWSwRDIbqkBWLX19fz+zKEVQU5Xj+XsaYzGKJYDBCHYB63jXU0hli+ba9nDq1wtP3McZkJksEgzFEcxG8s7mBUEQ5bZpNR2mMSTxLBIMxRHMRvLa+npwsH8dPHOnp+xhjMpMlgsHombje20Tw+vp6TphcSm6239P3McZkJksEgxH0PhHsaupgfW2LdQsZYzxjiWAwemoE3hWLX99QD2CFYmOMZywRDEaw1Xn0sEXw+vo6ygsDHDnG7iY2xnjDEsFgdLcIPCoWqyqvb2jglKnl+GyQOWOMRywRDIbHxeI1u5qpb+nk1KlWHzDGeMcSwWD0FIu9qRG8vt6tD1ih2BjjIUsEg+HxDWWvbahnSkUBY4uHbgY0Y0zmsUQwGB4WizuCYd7Z3MBp0+xqIWOMtzxNBCIyX0TWisgGEbn1EMdcLiKrRGSliDzsZTwJF2wHXxZkBRJ+6uVb99IRjFh9wBjjuSyvTiwifuAe4FygGlgiIotUdVXUMdOA24BTVHWviIzyKh5PeDhf8Wsb6snyCR+dYnMTG2O85WWL4ARgg6puUtUu4BHgkl7H/Atwj6ruBVDVWg/jSbyuVk8LxXMnlFCY41muNsYYwNtEUAVsj9qudvdFmw5MF5E3ROTvIjK/rxOJyHUislREltbV1XkU7gB41CLY29rFhzua7G5iY8yQiCkRiMgpIvK8iKwTkU0isllENiXg/bOAacAZwJXAr0WkpPdBqnqfqs5T1XkVFSn05ejRpDRvbmxA1S4bNcYMjVj7HX4L/CuwDAjH+JoaYHzU9jh3X7Rq4G1VDQKbRWQdTmJYEuN7JFewzZOuodc31FGUm8Ux44oTfm5jjOkt1q6hJlX9q6rWqmpD99LPa5YA00RksogEgCuARb2O+TNOawARKcfpKkpES2NoBNsTPryEqvLa+npOOqKMLL9d3WuM8V6s3zQvi8iPReQkETmuezncC1Q1BNwAPAesBh5V1ZUicqeIXOwe9hzQICKrgJeBf4shwaSOrtaEdw1tbWijem+7dQsZY4ZMrF1DJ7qP86L2KXDW4V6kqs8Cz/bad3vUugI3u8vw40Gx+LWeYactERhjhkZMiUBVz/Q6kGHJg0Tw+vo6qkrymFxekNDzGmPMocR61VCxiPy0+xJOEfmJiFglM5jY+whC4QhvboxyjHwAABLySURBVGzg1KnliNiw08aYoRFrjWAB0Axc7i77gPu9CmrYSHCx+P2aJpo7QlYfMMYMqVhrBFNU9dNR298VkRVeBDRsRCIQ6kho19Dr6+sRgVOsPmCMGUKxtgjaReTU7g0ROQVo9yakYcKDuQheX1/P7MoRlBYkfhA7Y4w5lFhbBF8CHnDrAgLsAa7xKqhhIcFzEbR0hli+bS//fNoRCTmfMcbEKtarhlYAx4jICHd7n6dRDQcJnovg7xsbCEWU06w+YIwZYodNBCJytar+TkRu7rUfAFX9qYexpbYETlyvqvz6tU2UF+Zw/MSRgz6fMcbEo78WQffF7EVeBzLsBBM3cf2bGxt4e/MevnPRLHKz/YM+nzHGxOOwiUBVf+U+fndowhlGuhJTLFZVfvzcWiqLc/nMiRMSEJgxxsQn1hvKfiQiI0QkW0ReFJE6Ebna6+BSWk+xeHB3AL+0ppYV2xu58exp5GRZa8AYM/RivXz0PLdAfCGwBZgK/JtXQQ0LCbh8NBJRfvK3dUwsy+ey48clKDBjjIlPrImguwvpAuBPqtrkUTzDRwISwV8/3MWqnfv46jnTyLYhp40xSRLrfQRPi8ganJvIviQiFUCHd2ENA92JIDCwrqFwRPnp82uZNqqQi4/pPYOnMcYMnZh+hqrqrcDJwDx3NrFWDp6IPrMMslj853dr2FjXys3nTsfvswHmjDHJ0999BGep6ksi8qmofdGHPOFVYClvEHcWB8MRfvbiOmZXjuDjs8ckODBjjIlPf11DpwMvARf18ZyS0YmgDfw54Iv/Sp8/La1m+5527r9mDj5rDRhjkqy/+wi+4z5eOzThDCMDnLi+IxjmFy+t57gJJZwxo8KDwIwxJj6x3kfwAxEpidoeKSLf8y6sYSDYNqBC8cNvb2NnUwdf//gMm3zGGJMSYr1m8XxVbezeUNW9wCe8CWmYCLbH3SJo6wrxv4s3cPKUMk6eYoPLGWNSQ6yJwC8iOd0bIpIH5Bzm+PTXFX/X0MI3t1Df0sXXzpvhUVDGGBO/WO8j+D3wooh0T095LfCANyENE8G2uIaX2NcR5FevbOKsI0fZCKPGmJQS63wEPxSR94Bz3F3/oarPeRfWMBBsg0BhTIc2dwT58u+X09Qe5OZzp3scmDHGxCfWFgHAaiCkqi+ISL6IFKlqs1eBpbxgOxSM6vew3fs6uPb+Jazd3cyPLjuaOVXFQxCcMcbELqZEICL/AlwHlAJTgCrgXuBs70JLccG2fielWb+7mWvuX0JjWxcLrvkIp0+3y0WNMakn1mLxl4FTgH0Aqroe6P/ncDrrp1j89qYGPv3LN+kKR/jjF0+yJGCMSVmxdg11qmpX93XvIpKFc2dx5gq2H7JY/PT7O7j5j+8xvjSPhdeewPjSxMxrbIwxXoi1RfCKiHwDyBORc4E/AU95F9YwcIg7i3/z2iZuePhdjhlfzONfOtmSgDEm5cXaIvh34J+BD4AvAs8Cv/EqqJQXDkIkeMCAc+GI8v1nVrPgjc184qgx/PTyY23+YWPMsNBvIhARP7BSVY8Efu19SMmhqjS2BRlZEOj/4J65CPJpag/yp6XbefCtrWzb08bnT5nMty6YaYPJGWOGjX4TgaqGRWStiExQ1W1DEdRQa+0MccPDy3l5bR1HVBRw7szRnDNrNMdNGNn3XAHuXAR/XrmH2559kfZgmHkTR/KNTxzJ/Dljhzh6Y4wZnFi7hkYCK0XkHZxJaQBQ1YsP9yIRmQ/8HPADv1HVuw5x3KeBx4CPqOrSGGNKiN37Ovj8wiWs2dXM50+ZzPraZha8sZlfvbqJkfnZnHnkKM6ZOZqPTa8gN8vHC6tr+b9X3+RnwJtb27jomLF89qRJdn+AMWbYijURfDveE7tdSvcA5wLVwBIRWaSqq3odVwR8BXg73vcYrLW7mrn2/ndobA/ym8/N48wZzhWxzR1BXl1Xzwurd/Pi6lqeWF5DwO+jOD+buuZOPjbCmbL59k99hMK5xwx12MYYk1D9zVCWC1wPTMUpFP9WVUMxnvsEYIOqbnLP9QjO9Jareh33H8APgX+LI+5Be2NDPdc/tIy8gJ9Hv3jSAb/oi3KzueDosVxw9FhC4QjLtu7lhdW72b6nnUvnVnFO0Ra4HwoLRwxlyMYY44n+WgQPAEHgNeB8YBbOr/dYVAHbo7argROjDxCR44DxqvqMiBwyEYjIdTh3NjNhwoQY3/7QHltWza2Pv8+UikIWXPsRqkoOfWNYlt/HiUeUceIRZft3blrjPPZzZ7ExxgwH/SWCWap6FICI/BZ4J1FvLCI+4KfANf0dq6r3AfcBzJs3b8A3sqkqP3thPT9/cT2nTi3nf68+jhG52fGfqGe+4oFNXG+MMamkv0QQ7F5R1VCcM2rVAOOjtse5+7oVAXOAxe55xwCLRORiLwrGXaEItz3xAY8vr+ay48fxg08eRSAr1vvpep/MrZcPYOJ6Y4xJNf0lgmNEZJ+7Ljh3Fu9z11VVD9dJvgSYJiKTcRLAFcBnup9U1SagZ5ouEVkMfN2rq4Z+/uI6Hl9ezb+eM52bzp46uGkie1oElgiMMcNff5PXD/jWWLcFcQPwHM7lowtUdaWI3AksVdVFAz33QHzx9CkcVVWcmOv8u28os0RgjEkD8cxHEDdVfRZnOIrofbcf4tgzvIxlRG524m726kkEViMwxgx/A+wkz3BWLDbGpBFLBAPR1ep0Cw2mzmCMMSnCEsFABNutPmCMSRuWCAbCEoExJo1YIhiIYKvVB4wxacMSwUAE2214CWNM2rBEMBBdbdY1ZIxJG5YIBuIQ8xUbY8xwZIlgIKxYbIxJI5YIBiLYaonAGJM2LBEMhBWLjTFpxBLBQFjXkDEmjVgiiJeqO8SEFYuNMenBEkG8Qp2AWovAGJM2LBHEy+YiMMakGUsE8bK5CIwxacYSQby65yIIFCQ3DmOMSRBLBPHqmbjeWgTGmPRgiSBeNnG9MSbNWCKIlxWLjTFpxhJBvKxYbIxJM5YI4mXFYmNMmrFEEC9rERhj0owlgnh1WSIwxqQXSwTx6mkRWNeQMSY9WCKIV7ANxA/+7GRHYowxCWGJIF7BdqdQLJLsSIwxJiEsEcTL5is2xqQZSwTx6rJEYIxJL5YI4hVss0KxMSateJoIRGS+iKwVkQ0icmsfz98sIqtE5H0ReVFEJnoZT0IE261FYIxJK54lAhHxA/cA5wOzgCtFZFavw94F5qnq0cBjwI+8iidhrEZgjEkzXrYITgA2qOomVe0CHgEuiT5AVV9WVffCfP4OjPMwnsQIttnwEsaYtOJlIqgCtkdtV7v7DuULwF/7ekJErhORpSKytK6uLoEhDoAVi40xaSYlisUicjUwD/hxX8+r6n2qOk9V51VUVAxtcL0F261YbIxJK1kenrsGGB+1Pc7ddwAROQf4JnC6qnZ6GE9iWI3AGJNmvGwRLAGmichkEQkAVwCLog8QkbnAr4CLVbXWw1gSxxKBMSbNeJYIVDUE3AA8B6wGHlXVlSJyp4hc7B72Y6AQ+JOIrBCRRYc4XWqIRCDUYcViY0xa8bJrCFV9Fni2177bo9bP8fL9Ey7UPV+xtQiMMekjJYrFw0aXzVdsjEk/lgjiYRPXG2PSkCWCeNg0lcaYNGSJIB7dicCKxcaYNGKJIB5BKxYbY9KPJYJ4WLHYGJOGLBHEw4rFxpg0ZIkgHtY1ZIxJQ5YI4hFsdR6tRWCMSSOWCOLR3SIIWCIwxqQPSwTx6C4WZ1nXkDEmfVgiiEewDfwB8Hs6RJMxxgwpSwTxCLZbfcAYk3YsEcQj2GqJwBiTdiwRxCPYboViY0zasUQQj2C73UNgjEk7lgji0WVdQ8aY9GOJIB5WLDbGpCFLBPEItlkiMMakHUsE8Qi2WY3AGJN2LBHEw64aMsakIUsE8eiyriFjTPqxRBAPqxEYY9KQJYJYhYMQCVoiMMakHUsEseqZncyKxcaY9GKJIFY2F4ExJk1ZIohVl81OZoxJT5YIYmXzFRtj0pQlglj1JIKC5MZhjDEJZokgVj0T11uLwBiTXiwRxMqKxcaYNOVpIhCR+SKyVkQ2iMitfTyfIyJ/dJ9/W0QmeRnPoPRcPmqJwBiTXjxLBCLiB+4BzgdmAVeKyKxeh30B2KuqU4H/Bn7oVTyD1mX3ERhj0lOWh+c+AdigqpsAROQR4BJgVdQxlwB3uOuPAf8jIqKqmvBolj8Eb/3PwF/fvtd5tBaBMSbNeJkIqoDtUdvVwImHOkZVQyLSBJQB9dEHich1wHUAEyZMGFg0+aVQMWNgr+1WPB7yywZ3DmOMSTFeJoKEUdX7gPsA5s2bN7DWwpEXOIsxxpgDeFksrgHGR22Pc/f1eYyIZAHFQIOHMRljjOnFy0SwBJgmIpNFJABcASzqdcwi4HPu+mXAS57UB4wxxhySZ11Dbp//DcBzgB9YoKorReROYKmqLgJ+CzwkIhuAPTjJwhhjzBDytEagqs8Cz/bad3vUegfwD17GYIwx5vDszmJjjMlwlgiMMSbDWSIwxpgMZ4nAGGMynAy3qzVFpA7YOsCXl9PrruUUY/ENjsU3eKkeo8U3cBNVtaKvJ4ZdIhgMEVmqqvOSHcehWHyDY/ENXqrHaPF5w7qGjDEmw1kiMMaYDJdpieC+ZAfQD4tvcCy+wUv1GC0+D2RUjcAYY8zBMq1FYIwxphdLBMYYk+EyJhGIyHwRWSsiG0Tk1mTH05uIbBGRD0RkhYgsTYF4FohIrYh8GLWvVESeF5H17uPIFIvvDhGpcT/DFSLyiSTGN15EXhaRVSKyUkS+4u5Pic/wMPGlxGcoIrki8o6IvOfG9113/2QRedv9f/xHd4j7VIpvoYhsjvr8jk1GfPHKiBqBiPiBdcC5OFNmLgGuVNVVh33hEBKRLcA8VU2Jm1FE5GNAC/Cgqs5x9/0I2KOqd7nJdKSq/nsKxXcH0KKq/5WMmKKJyFhgrKouF5EiYBlwKXANKfAZHia+y0mBz1BEBChQ1RYRyQZeB74C3Aw8oaqPiMi9wHuq+ssUiu964GlVfWyoYxqMTGkRnABsUNVNqtoFPAJckuSYUpqqvoozR0S0S4AH3PUHcL44kuIQ8aUMVd2pqsvd9WZgNc4c3SnxGR4mvpSgjhZ3M9tdFDgL6P6STebnd6j4hqVMSQRVwPao7WpS6B+9S4G/icgyEbku2cEcwmhV3emu7wJGJzOYQ7hBRN53u46S1nUVTUQmAXOBt0nBz7BXfJAin6GI+EVkBVALPA9sBBpVNeQektT/x73jU9Xuz+/77uf33yKSk6z44pEpiWA4OFVVjwPOB77sdn2kLHdK0VT7BfRLYApwLLAT+ElywwERKQQeB76qqvuin0uFz7CP+FLmM1TVsKoeizPf+QnAkcmKpS+94xOROcBtOHF+BCgFktJ1Gq9MSQQ1wPio7XHuvpShqjXuYy3wJM4//FSz2+1b7u5jrk1yPAdQ1d3uf84I8GuS/Bm6fcePA79X1Sfc3SnzGfYVX6p9hm5MjcDLwElAiYh0z6yYEv+Po+Kb73a5qap2AveTAp9fLDIlESwBprlXHARw5kZelOSYeohIgVuwQ0QKgPOADw//qqRYBHzOXf8c8JckxnKQ7i9Y1ydJ4mfoFhN/C6xW1Z9GPZUSn+Gh4kuVz1BEKkSkxF3Pw7nQYzXOF+5l7mHJ/Pz6im9NVJIXnPpFKv4/PkhGXDUE4F4G9zPADyxQ1e8nOaQeInIETisAnHmkH052fCLyB+AMnGF1dwPfAf4MPApMwBkK/HJVTUrB9hDxnYHTpaHAFuCLUf3xQx3fqcBrwAdAxN39DZx++KR/hoeJ70pS4DMUkaNxisF+nB+sj6rqne7/lUdwul3eBa52f32nSnwvARWAACuA66OKyikrYxKBMcaYvmVK15AxxphDsERgjDEZzhKBMcZkOEsExhiT4SwRGGNMhrNEYEwvIhKOGj1yhSRwtFoRmSRRI6Yakwqy+j/EmIzT7g4dYExGsBaBMTESZ86IH4kzb8Q7IjLV3T9JRF5yBxp7UUQmuPtHi8iT7pj174nIye6p/CLya3cc+7+5d6YakzSWCIw5WF6vrqF/jHquSVWPAv4H5051gF8AD6jq0cDvgbvd/XcDr6jqMcBxwEp3/zTgHlWdDTQCn/b47zHmsOzOYmN6EZEWVS3sY/8W4CxV3eQO2LZLVctEpB5nkpegu3+nqpaLSB0wLnoIBHfI5+dVdZq7/e9Atqp+z/u/zJi+WYvAmPjoIdbjET02Thir1Zkks0RgTHz+MerxLXf9TZwRbQGuwhnMDeBF4EvQM4lJ8VAFaUw87JeIMQfLc2ee6vZ/qtp9CelIEXkf51f9le6+G4H7ReTfgDrgWnf/V4D7ROQLOL/8v4Qz2YsxKcVqBMbEyK0RzFPV+mTHYkwiWdeQMcZkOGsRGGNMhrMWgTHGZDhLBMYYk+EsERhjTIazRGCMMRnOEoExxmS4/w80oRIBOwRifwAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JANhEnjNH0vg"
      },
      "source": [
        "results = model_DNN.evaluate(X_test_tfidf, y_test, verbose=False)"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-Cda4vvBJO8M",
        "outputId": "36b38ec5-e696-4279-aa8e-24d36868b23a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 116
        }
      },
      "source": [
        "print(\"Testing\")\n",
        "print(\"LOSS : {:.4f}\\nBINARY_ACC : {:.4f}\\nAUC : {:.4f}\\nPRECISION : {:.4f}\\nRECALL : {:.4f}\".format(*results))"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Testing\n",
            "LOSS : 0.3834\n",
            "BINARY_ACC : 0.8325\n",
            "AUC : 0.9054\n",
            "PRECISION : 0.7845\n",
            "RECALL : 0.7560\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "y09Vj_dLlweV"
      },
      "source": [
        "#Grid Search"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MsZLoW1g1-qW"
      },
      "source": [
        "X = vectorizer_x.fit_transform(df_X).toarray()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GTVVMxSSfOkb"
      },
      "source": [
        "#######################################################\n",
        "## Neurons\n",
        "## Dropout"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0oGKyyj1fTxi"
      },
      "source": [
        "# define the grid search parameters\n",
        "hidden = [1,2]\n",
        "neurons = [16,32,64]\n",
        "drops = [0.85]\n",
        "param_grid = dict(dropout=drops, neurons=neurons, nLayers=hidden)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HCsQRUwfgrE5"
      },
      "source": [
        "model = KerasClassifier(build_fn=Build_Model_DNN_Text, epochs=20, batch_size=2048, verbose=0, shape = X.shape[1], optimizer='Adam')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FKnyTsIlgoi2",
        "outputId": "8dddd673-2eaf-49b3-b326-280f8aa24e65",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 683
        }
      },
      "source": [
        "grid = GridSearchCV(estimator=model, param_grid=param_grid, cv=3, n_jobs=1, verbose=2)\n",
        "grid_result = grid.fit(X, df_Y)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fitting 3 folds for each of 6 candidates, totalling 18 fits\n",
            "[CV] dropout=0.85, nLayers=1, neurons=16 .............................\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "[CV] .............. dropout=0.85, nLayers=1, neurons=16, total=  25.5s\n",
            "[CV] dropout=0.85, nLayers=1, neurons=16 .............................\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:   25.5s remaining:    0.0s\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "[CV] .............. dropout=0.85, nLayers=1, neurons=16, total=  25.1s\n",
            "[CV] dropout=0.85, nLayers=1, neurons=16 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=1, neurons=16, total=  25.2s\n",
            "[CV] dropout=0.85, nLayers=1, neurons=32 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=1, neurons=32, total=  25.2s\n",
            "[CV] dropout=0.85, nLayers=1, neurons=32 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=1, neurons=32, total=  25.6s\n",
            "[CV] dropout=0.85, nLayers=1, neurons=32 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=1, neurons=32, total=  25.3s\n",
            "[CV] dropout=0.85, nLayers=1, neurons=64 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=1, neurons=64, total=  25.2s\n",
            "[CV] dropout=0.85, nLayers=1, neurons=64 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=1, neurons=64, total=  25.1s\n",
            "[CV] dropout=0.85, nLayers=1, neurons=64 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=1, neurons=64, total=  25.1s\n",
            "[CV] dropout=0.85, nLayers=2, neurons=16 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=2, neurons=16, total=  25.3s\n",
            "[CV] dropout=0.85, nLayers=2, neurons=16 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=2, neurons=16, total=  25.2s\n",
            "[CV] dropout=0.85, nLayers=2, neurons=16 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=2, neurons=16, total=  25.2s\n",
            "[CV] dropout=0.85, nLayers=2, neurons=32 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=2, neurons=32, total=  25.4s\n",
            "[CV] dropout=0.85, nLayers=2, neurons=32 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=2, neurons=32, total=  24.9s\n",
            "[CV] dropout=0.85, nLayers=2, neurons=32 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=2, neurons=32, total=  25.0s\n",
            "[CV] dropout=0.85, nLayers=2, neurons=64 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=2, neurons=64, total=  24.9s\n",
            "[CV] dropout=0.85, nLayers=2, neurons=64 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=2, neurons=64, total=  24.8s\n",
            "[CV] dropout=0.85, nLayers=2, neurons=64 .............................\n",
            "[CV] .............. dropout=0.85, nLayers=2, neurons=64, total=  24.8s\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Done  18 out of  18 | elapsed:  7.5min finished\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4kAOibtiglTW",
        "outputId": "e29b164b-1440-47c5-8ee4-02d06eb3d792",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 133
        }
      },
      "source": [
        "# summarize results\n",
        "print(\"Best: %f using %s\" % (grid_result.best_score_, grid_result.best_params_))\n",
        "means = grid_result.cv_results_['mean_test_score']\n",
        "stds = grid_result.cv_results_['std_test_score']\n",
        "params = grid_result.cv_results_['params']\n",
        "for mean, stdev, param in zip(means, stds, params):\n",
        "    print(\"%f (%f) with: %r\" % (mean, stdev, param))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Best: 0.836375 using {'dropout': 0.85, 'nLayers': 1, 'neurons': 16}\n",
            "0.836375 (0.002477) with: {'dropout': 0.85, 'nLayers': 1, 'neurons': 16}\n",
            "0.832000 (0.001987) with: {'dropout': 0.85, 'nLayers': 1, 'neurons': 32}\n",
            "0.829500 (0.002624) with: {'dropout': 0.85, 'nLayers': 1, 'neurons': 64}\n",
            "0.835083 (0.002399) with: {'dropout': 0.85, 'nLayers': 2, 'neurons': 16}\n",
            "0.833792 (0.002536) with: {'dropout': 0.85, 'nLayers': 2, 'neurons': 32}\n",
            "0.830292 (0.002359) with: {'dropout': 0.85, 'nLayers': 2, 'neurons': 64}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NiTSOtJZaZ6X"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}