LSTMblog1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "LSTMblog1.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZNdghZRU0RgS",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 122
        },
        "outputId": "22c2fda3-0218-48c6-d877-0013a51b347d"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Go to this URL in a browser: https://accounts.google.com/o/oauth2/auth?client_id=947318989803-6bn6qk8qdgf4n4g3pfee6491hc0brc4i.apps.googleusercontent.com&redirect_uri=urn%3aietf%3awg%3aoauth%3a2.0%3aoob&response_type=code&scope=email%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdocs.test%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive.photos.readonly%20https%3a%2f%2fwww.googleapis.com%2fauth%2fpeopleapi.readonly\n",
            "\n",
            "Enter your authorization code:\n",
            "··········\n",
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j59W7I6v1F0d",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "7e19135b-a792-47ec-8c8e-fc40e68783a9"
      },
      "source": [
        "%tensorflow_version 1.x"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "TensorFlow 1.x selected.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5PGWnkgE1F3l",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 306
        },
        "outputId": "421e40b0-c930-482b-881a-5922aa870590"
      },
      "source": [
        "import pandas as pd\n",
        "df = pd.read_csv('/content/drive/My Drive/Medium Blogs/Reviews.csv')\n",
        "print(df.shape)\n",
        "df.head()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(568454, 10)\n"
          ],
          "name": "stdout"
        },
        {
          "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>Id</th>\n",
              "      <th>ProductId</th>\n",
              "      <th>UserId</th>\n",
              "      <th>ProfileName</th>\n",
              "      <th>HelpfulnessNumerator</th>\n",
              "      <th>HelpfulnessDenominator</th>\n",
              "      <th>Score</th>\n",
              "      <th>Time</th>\n",
              "      <th>Summary</th>\n",
              "      <th>Text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>B001E4KFG0</td>\n",
              "      <td>A3SGXH7AUHU8GW</td>\n",
              "      <td>delmartian</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>1303862400</td>\n",
              "      <td>Good Quality Dog Food</td>\n",
              "      <td>I have bought several of the Vitality canned d...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>B00813GRG4</td>\n",
              "      <td>A1D87F6ZCVE5NK</td>\n",
              "      <td>dll pa</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1346976000</td>\n",
              "      <td>Not as Advertised</td>\n",
              "      <td>Product arrived labeled as Jumbo Salted Peanut...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>B000LQOCH0</td>\n",
              "      <td>ABXLMWJIXXAIN</td>\n",
              "      <td>Natalia Corres \"Natalia Corres\"</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>1219017600</td>\n",
              "      <td>\"Delight\" says it all</td>\n",
              "      <td>This is a confection that has been around a fe...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>B000UA0QIQ</td>\n",
              "      <td>A395BORC6FGVXV</td>\n",
              "      <td>Karl</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1307923200</td>\n",
              "      <td>Cough Medicine</td>\n",
              "      <td>If you are looking for the secret ingredient i...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>B006K2ZZ7K</td>\n",
              "      <td>A1UQRSCLF8GW1T</td>\n",
              "      <td>Michael D. Bigham \"M. Wassir\"</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>1350777600</td>\n",
              "      <td>Great taffy</td>\n",
              "      <td>Great taffy at a great price.  There was a wid...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Id  ...                                               Text\n",
              "0   1  ...  I have bought several of the Vitality canned d...\n",
              "1   2  ...  Product arrived labeled as Jumbo Salted Peanut...\n",
              "2   3  ...  This is a confection that has been around a fe...\n",
              "3   4  ...  If you are looking for the secret ingredient i...\n",
              "4   5  ...  Great taffy at a great price.  There was a wid...\n",
              "\n",
              "[5 rows x 10 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "atm1C-E_e4A0",
        "colab_type": "text"
      },
      "source": [
        "## Filtering out neutral reviews."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AzBmzzij1F6g",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 289
        },
        "outputId": "fde63761-dcc9-4839-fe1d-dc568535f4e1"
      },
      "source": [
        "filtered_data = df[df['Score'] != 3]\n",
        "filtered_data.head()"
      ],
      "execution_count": 4,
      "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>Id</th>\n",
              "      <th>ProductId</th>\n",
              "      <th>UserId</th>\n",
              "      <th>ProfileName</th>\n",
              "      <th>HelpfulnessNumerator</th>\n",
              "      <th>HelpfulnessDenominator</th>\n",
              "      <th>Score</th>\n",
              "      <th>Time</th>\n",
              "      <th>Summary</th>\n",
              "      <th>Text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>B001E4KFG0</td>\n",
              "      <td>A3SGXH7AUHU8GW</td>\n",
              "      <td>delmartian</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>1303862400</td>\n",
              "      <td>Good Quality Dog Food</td>\n",
              "      <td>I have bought several of the Vitality canned d...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>B00813GRG4</td>\n",
              "      <td>A1D87F6ZCVE5NK</td>\n",
              "      <td>dll pa</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1346976000</td>\n",
              "      <td>Not as Advertised</td>\n",
              "      <td>Product arrived labeled as Jumbo Salted Peanut...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>B000LQOCH0</td>\n",
              "      <td>ABXLMWJIXXAIN</td>\n",
              "      <td>Natalia Corres \"Natalia Corres\"</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>1219017600</td>\n",
              "      <td>\"Delight\" says it all</td>\n",
              "      <td>This is a confection that has been around a fe...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>B000UA0QIQ</td>\n",
              "      <td>A395BORC6FGVXV</td>\n",
              "      <td>Karl</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1307923200</td>\n",
              "      <td>Cough Medicine</td>\n",
              "      <td>If you are looking for the secret ingredient i...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>B006K2ZZ7K</td>\n",
              "      <td>A1UQRSCLF8GW1T</td>\n",
              "      <td>Michael D. Bigham \"M. Wassir\"</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>1350777600</td>\n",
              "      <td>Great taffy</td>\n",
              "      <td>Great taffy at a great price.  There was a wid...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Id  ...                                               Text\n",
              "0   1  ...  I have bought several of the Vitality canned d...\n",
              "1   2  ...  Product arrived labeled as Jumbo Salted Peanut...\n",
              "2   3  ...  This is a confection that has been around a fe...\n",
              "3   4  ...  If you are looking for the secret ingredient i...\n",
              "4   5  ...  Great taffy at a great price.  There was a wid...\n",
              "\n",
              "[5 rows x 10 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Wh0bCtCk1F9S",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "outputId": "dfac46bd-ccba-4254-a4c3-0e5dba027735"
      },
      "source": [
        "# Give reviews with Score>3 a positive rating(1), and reviews with a score<3 a negative rating(0).\n",
        "def partition(x):\n",
        "    if x < 3:\n",
        "        return 0\n",
        "    return 1\n",
        "\n",
        "#changing reviews with score less than 3 to be positive and vice-versa\n",
        "actualScore = filtered_data['Score']\n",
        "positiveNegative = actualScore.map(partition) \n",
        "filtered_data['Score'] = positiveNegative\n",
        "print(\"Number of data points in our data\", filtered_data.shape)\n",
        "filtered_data.head(3)"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Number of data points in our data (525814, 10)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:9: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  if __name__ == '__main__':\n"
          ],
          "name": "stderr"
        },
        {
          "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>Id</th>\n",
              "      <th>ProductId</th>\n",
              "      <th>UserId</th>\n",
              "      <th>ProfileName</th>\n",
              "      <th>HelpfulnessNumerator</th>\n",
              "      <th>HelpfulnessDenominator</th>\n",
              "      <th>Score</th>\n",
              "      <th>Time</th>\n",
              "      <th>Summary</th>\n",
              "      <th>Text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>B001E4KFG0</td>\n",
              "      <td>A3SGXH7AUHU8GW</td>\n",
              "      <td>delmartian</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1303862400</td>\n",
              "      <td>Good Quality Dog Food</td>\n",
              "      <td>I have bought several of the Vitality canned d...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>B00813GRG4</td>\n",
              "      <td>A1D87F6ZCVE5NK</td>\n",
              "      <td>dll pa</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1346976000</td>\n",
              "      <td>Not as Advertised</td>\n",
              "      <td>Product arrived labeled as Jumbo Salted Peanut...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>B000LQOCH0</td>\n",
              "      <td>ABXLMWJIXXAIN</td>\n",
              "      <td>Natalia Corres \"Natalia Corres\"</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1219017600</td>\n",
              "      <td>\"Delight\" says it all</td>\n",
              "      <td>This is a confection that has been around a fe...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Id  ...                                               Text\n",
              "0   1  ...  I have bought several of the Vitality canned d...\n",
              "1   2  ...  Product arrived labeled as Jumbo Salted Peanut...\n",
              "2   3  ...  This is a confection that has been around a fe...\n",
              "\n",
              "[3 rows x 10 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9gwx_MGsex9W",
        "colab_type": "text"
      },
      "source": [
        "## Removing duplicates from the dataset."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CD80lxF21GAR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Sorting data according to ProductId in ascending order\n",
        "sorted_data=filtered_data.sort_values('ProductId', axis=0, ascending=True, inplace=False, kind='quicksort', na_position='last')\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eORu8Bkx1GDO",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "3eb18295-1e32-42bc-bffd-d1607f22dd67"
      },
      "source": [
        "#Deduplication of entries\n",
        "final=sorted_data.drop_duplicates(subset={\"UserId\",\"ProfileName\",\"Time\",\"Text\"}, keep='first', inplace=False)\n",
        "final.shape"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(364173, 10)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YcdLQvP91GGG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "deab3a7c-cc48-41c7-deb4-b9ee57eba1eb"
      },
      "source": [
        "#Checking to see how much % of data still remains\n",
        "(final['Id'].size*1.0)/(filtered_data['Id'].size*1.0)*100"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "69.25890143662969"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aImTf_9k1GI8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "7c780334-cca7-40da-96ad-ad3e4a78d40a"
      },
      "source": [
        "final=final[final.HelpfulnessNumerator<=final.HelpfulnessDenominator]\n",
        "final['Score'].value_counts()"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1    307061\n",
              "0     57110\n",
              "Name: Score, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4wXm46MLe8T9",
        "colab_type": "text"
      },
      "source": [
        "## Preprocessing the data.\n",
        "Now that we have finished deduplication our data requires some preprocessing before we go on further with analysis and making the prediction model.\n",
        "\n",
        "Hence in the Preprocessing phase we do the following in the order below:-\n",
        "\n",
        "1. Begin by removing the html tags\n",
        "\n",
        "2. Remove any punctuations or limited set of special characters like , or .   or # etc.\n",
        "\n",
        "3. Check if the word is made up of english letters and is not alpha-numeric\n",
        "\n",
        "4. Check to see if the length of the word is greater than 2 (as it was researched that there is no adjective in 2-letters)\n",
        "\n",
        "5. Convert the word to lowercase\n",
        "6. Remove Stopwords"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "So6ryghs1GLx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# https://stackoverflow.com/a/47091490/4084039\n",
        "import re\n",
        "from bs4 import BeautifulSoup\n",
        "\n",
        "def decontracted(phrase):\n",
        "    # specific\n",
        "    phrase = re.sub(r\"won't\", \"will not\", phrase)\n",
        "    phrase = re.sub(r\"can\\'t\", \"can not\", phrase)\n",
        "\n",
        "    # general\n",
        "    phrase = re.sub(r\"n\\'t\", \" not\", phrase)\n",
        "    phrase = re.sub(r\"\\'re\", \" are\", phrase)\n",
        "    phrase = re.sub(r\"\\'s\", \" is\", phrase)\n",
        "    phrase = re.sub(r\"\\'d\", \" would\", phrase)\n",
        "    phrase = re.sub(r\"\\'ll\", \" will\", phrase)\n",
        "    phrase = re.sub(r\"\\'t\", \" not\", phrase)\n",
        "    phrase = re.sub(r\"\\'ve\", \" have\", phrase)\n",
        "    phrase = re.sub(r\"\\'m\", \" am\", phrase)\n",
        "    return phrase\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7_mXuMBH1GOW",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# https://gist.github.com/sebleier/554280\n",
        "# we are removing the words from the stop words list: 'no', 'nor', 'not'\n",
        "# <br /><br /> ==> after the above steps, we are getting \"br br\"\n",
        "# we are including them into stop words list\n",
        "# instead of <br /> if we have <br/> these tags would have revmoved in the 1st step\n",
        "\n",
        "stopwords= set(['br', 'the', 'i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', \"you're\", \"you've\",\\\n",
        "            \"you'll\", \"you'd\", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', \\\n",
        "            'she', \"she's\", 'her', 'hers', 'herself', 'it', \"it's\", 'its', 'itself', 'they', 'them', 'their',\\\n",
        "            'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', \"that'll\", 'these', 'those', \\\n",
        "            'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', \\\n",
        "            'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', \\\n",
        "            'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after',\\\n",
        "            'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further',\\\n",
        "            'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more',\\\n",
        "            'most', 'other', 'some', 'such', 'only', 'own', 'same', 'so', 'than', 'too', 'very', \\\n",
        "            's', 't', 'can', 'will', 'just', 'don', \"don't\", 'should', \"should've\", 'now', 'd', 'll', 'm', 'o', 're', \\\n",
        "            've', 'y', 'ain', 'aren', \"aren't\", 'couldn', \"couldn't\", 'didn', \"didn't\", 'doesn', \"doesn't\", 'hadn',\\\n",
        "            \"hadn't\", 'hasn', \"hasn't\", 'haven', \"haven't\", 'isn', \"isn't\", 'ma', 'mightn', \"mightn't\", 'mustn',\\\n",
        "            \"mustn't\", 'needn', \"needn't\", 'shan', \"shan't\", 'shouldn', \"shouldn't\", 'wasn', \"wasn't\", 'weren', \"weren't\", \\\n",
        "            'won', \"won't\", 'wouldn', \"wouldn't\"])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qEr6OhaP1GRG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "7116ca7b-6908-4999-8b88-41b2cff6c357"
      },
      "source": [
        "from tqdm import tqdm\n",
        "preprocessed_reviews = []\n",
        "# tqdm is for printing the status bar\n",
        "for sentance in tqdm(final['Text'].values):\n",
        "    sentance = re.sub(r\"http\\S+\", \"\", sentance)\n",
        "    sentance = BeautifulSoup(sentance, 'lxml').get_text()\n",
        "    sentance = decontracted(sentance)\n",
        "    sentance = re.sub(\"\\S*\\d\\S*\", \"\", sentance).strip()\n",
        "    sentance = re.sub('[^A-Za-z]+', ' ', sentance)\n",
        "    # https://gist.github.com/sebleier/554280\n",
        "    sentance = ' '.join(e.lower() for e in sentance.split() if e.lower() not in stopwords)\n",
        "    preprocessed_reviews.append(sentance.strip())"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "100%|██████████| 364171/364171 [02:30<00:00, 2415.38it/s]\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Zn6cvWfGg5MT",
        "colab_type": "text"
      },
      "source": [
        "## Vectorizing the text data."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mJjcoLrxgCCH",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras.preprocessing.text import Tokenizer\n",
        "from keras.preprocessing.sequence import pad_sequences\n",
        "# Instansiating tokenizer.\n",
        "t = Tokenizer()\n",
        "# Preprocessed_reviews contains all the cleaned reviews.\n",
        "t.fit_on_texts(preprocessed_reviews)\n",
        "vocab_size = len(t.word_index) + 1\n",
        "# integer encode the reviews.\n",
        "encoded_reviews = t.texts_to_sequences(preprocessed_reviews)\n",
        "# pad reviews to a max length of 150 words\n",
        "max_length = 150\n",
        "padded_reviews = pad_sequences(encoded_reviews, maxlen=max_length, padding='post')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6NUJ1ibWgCFL",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "outputId": "ff28b054-db26-4e07-f7ae-5a48ffbfb35b"
      },
      "source": [
        "print(padded_reviews[0])"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[21845    22  1413    77   362  4884  4426 26581  1256  3259   503    79\n",
            "  7342 11910  1254 30623  2231 22781  3832    12   152  1611  1413 16776\n",
            " 23796  1467  1413  1778  2285   362    64   237 26581  3331  1926     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FNpnzOrts2jh",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "20abd550-1124-4a66-f2e3-b572a7c3b295"
      },
      "source": [
        "x_train = padded_reviews[0:int(0.9*len(padded_reviews))]\n",
        "x_test = padded_reviews[int(0.9*len(padded_reviews)):len(padded_reviews)]\n",
        "print(x_train.shape)\n",
        "print(x_test.shape)"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(327753, 150)\n",
            "(36418, 150)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fd-Rv_NEsNer",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "d7421c7f-9dfe-4f46-cc10-ca1d51876160"
      },
      "source": [
        "import numpy as np\n",
        "y_train = np.array(list(final['Score'].iloc[0:int(0.9*len(padded_reviews))].values))\n",
        "y_test = np.array(list(final['Score'].iloc[int(0.9*len(padded_reviews)):len(padded_reviews)].values))\n",
        "print(y_train.shape, y_test.shape)"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(327753,) (36418,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f9cMcfzkh1_w",
        "colab_type": "text"
      },
      "source": [
        "## Defining the model."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "l4764vb_gCII",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 323
        },
        "outputId": "bbf1e92b-056d-4ca1-eec4-7a5782585dd8"
      },
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Input, CuDNNLSTM, Dense\n",
        "from keras.layers.embeddings import Embedding\n",
        "import keras.backend as K\n",
        "\n",
        "model = Sequential()\n",
        "model.add(Embedding(vocab_size, 32, input_length=max_length))\n",
        "# While passing output to next LSTM layer set return_sequence to Ture.\n",
        "model.add(CuDNNLSTM(128, return_sequences=True)) \n",
        "model.add(CuDNNLSTM(128))\n",
        "model.add(Dense(32,activation='relu'))\n",
        "model.add(Dense(1,activation='sigmoid'))\n",
        "model.compile(loss=['binary_crossentropy'] , optimizer='adam', metrics=['accuracy'])\n",
        "model.summary()"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_5\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_5 (Embedding)      (None, 150, 32)           3736800   \n",
            "_________________________________________________________________\n",
            "cu_dnnlstm_7 (CuDNNLSTM)     (None, 150, 128)          82944     \n",
            "_________________________________________________________________\n",
            "cu_dnnlstm_8 (CuDNNLSTM)     (None, 128)               132096    \n",
            "_________________________________________________________________\n",
            "dense_7 (Dense)              (None, 32)                4128      \n",
            "_________________________________________________________________\n",
            "dense_8 (Dense)              (None, 1)                 33        \n",
            "=================================================================\n",
            "Total params: 3,956,001\n",
            "Trainable params: 3,956,001\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rwdCaqV1tttM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras.callbacks import EarlyStopping\n",
        "es = EarlyStopping(monitor='val_loss', mode='min', verbose=5, patience=5)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Bf43uTIhgCLe",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 289
        },
        "outputId": "7717c533-c35a-4841-8e9c-1c2ec380b85f"
      },
      "source": [
        "history = model.fit(x_train, y_train, epochs=50, batch_size=1024, validation_split=0.1, callbacks=[es] )"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 294977 samples, validate on 32776 samples\n",
            "Epoch 1/50\n",
            "294977/294977 [==============================] - 22s 76us/step - loss: 0.4339 - accuracy: 0.8460 - val_loss: 0.4549 - val_accuracy: 0.8302\n",
            "Epoch 2/50\n",
            "294977/294977 [==============================] - 22s 74us/step - loss: 0.4165 - accuracy: 0.8460 - val_loss: 0.4416 - val_accuracy: 0.8302\n",
            "Epoch 3/50\n",
            "294977/294977 [==============================] - 22s 74us/step - loss: 0.4165 - accuracy: 0.8460 - val_loss: 0.4421 - val_accuracy: 0.8302\n",
            "Epoch 4/50\n",
            "294977/294977 [==============================] - 22s 74us/step - loss: 0.4172 - accuracy: 0.8460 - val_loss: 0.4428 - val_accuracy: 0.8302\n",
            "Epoch 5/50\n",
            "294977/294977 [==============================] - 22s 74us/step - loss: 0.4175 - accuracy: 0.8460 - val_loss: 0.4437 - val_accuracy: 0.8302\n",
            "Epoch 6/50\n",
            "294977/294977 [==============================] - 22s 74us/step - loss: 0.4214 - accuracy: 0.8460 - val_loss: 0.4570 - val_accuracy: 0.8302\n",
            "Epoch 7/50\n",
            "294977/294977 [==============================] - 22s 74us/step - loss: 0.4297 - accuracy: 0.8460 - val_loss: 0.4568 - val_accuracy: 0.8302\n",
            "Epoch 00007: early stopping\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NSDvR4CZ3oFS",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "27b11118-7d3a-47b0-9107-c64789f3dfe2"
      },
      "source": [
        "import matplotlib.pyplot as plt\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', 'cv'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 47,
      "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"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VkZKdN2v4ArO",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "9ac8a3a7-4023-40cc-9e68-31752e1e73ed"
      },
      "source": [
        "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', 'cv'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 48,
      "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"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V1eGD-Mh2s6w",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "prob = model.predict(x_test)\n",
        "y_pred = []\n",
        "for i in range(prob.shape[0]):\n",
        "  if(prob[i][0]>=0.5):\n",
        "    y_pred.append(1)\n",
        "  else:\n",
        "    y_pred.append(0)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u-VsnWAw2yWg",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "outputId": "d804f6f5-eb98-45be-c521-9ca1ed4e069e"
      },
      "source": [
        "from sklearn.metrics import classification_report\n",
        "print(classification_report(y_test, y_pred))"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.00      0.00      0.00      6111\n",
            "           1       0.83      1.00      0.91     30307\n",
            "\n",
            "    accuracy                           0.83     36418\n",
            "   macro avg       0.42      0.50      0.45     36418\n",
            "weighted avg       0.69      0.83      0.76     36418\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/metrics/_classification.py:1272: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eVIWwy0s1bcj",
        "colab_type": "text"
      },
      "source": [
        "## Architecture using bidirectional LSTMS"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o-qoEHfEgCOi",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 323
        },
        "outputId": "f3c1b3dd-f24a-4a90-af28-76a54b62768c"
      },
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Input, CuDNNLSTM, Dense, Bidirectional\n",
        "from keras.layers.embeddings import Embedding\n",
        "import keras.backend as K\n",
        "\n",
        "model = Sequential()\n",
        "model.add(Embedding(vocab_size, 32, input_length=max_length))\n",
        "# While passing output to next LSTM layer set return_sequence to Ture.\n",
        "model.add(Bidirectional(CuDNNLSTM(128, return_sequences=True))) \n",
        "model.add(Bidirectional(CuDNNLSTM(128)))\n",
        "model.add(Dense(32,activation='relu'))\n",
        "model.add(Dense(1,activation='sigmoid'))\n",
        "model.compile(loss=['binary_crossentropy'] , optimizer='adam', metrics=['accuracy'])\n",
        "model.summary()"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_6\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_6 (Embedding)      (None, 150, 32)           3736800   \n",
            "_________________________________________________________________\n",
            "bidirectional_3 (Bidirection (None, 150, 256)          165888    \n",
            "_________________________________________________________________\n",
            "bidirectional_4 (Bidirection (None, 256)               395264    \n",
            "_________________________________________________________________\n",
            "dense_9 (Dense)              (None, 32)                8224      \n",
            "_________________________________________________________________\n",
            "dense_10 (Dense)             (None, 1)                 33        \n",
            "=================================================================\n",
            "Total params: 4,306,209\n",
            "Trainable params: 4,306,209\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nZ0dDnxEgCRv",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 323
        },
        "outputId": "e1a87b6c-4efa-43da-a480-589b9b2d0139"
      },
      "source": [
        "history = model.fit(x_train, y_train, epochs=50, batch_size=1024, validation_split=0.1, callbacks=[es] )"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 294977 samples, validate on 32776 samples\n",
            "Epoch 1/50\n",
            "294977/294977 [==============================] - 52s 176us/step - loss: 0.2414 - accuracy: 0.9063 - val_loss: 0.2022 - val_accuracy: 0.9188\n",
            "Epoch 2/50\n",
            "294977/294977 [==============================] - 51s 173us/step - loss: 0.1581 - accuracy: 0.9399 - val_loss: 0.1838 - val_accuracy: 0.9254\n",
            "Epoch 3/50\n",
            "294977/294977 [==============================] - 51s 173us/step - loss: 0.1331 - accuracy: 0.9504 - val_loss: 0.1752 - val_accuracy: 0.9309\n",
            "Epoch 4/50\n",
            "294977/294977 [==============================] - 51s 173us/step - loss: 0.1163 - accuracy: 0.9573 - val_loss: 0.1990 - val_accuracy: 0.9292\n",
            "Epoch 5/50\n",
            "294977/294977 [==============================] - 51s 173us/step - loss: 0.1010 - accuracy: 0.9638 - val_loss: 0.1909 - val_accuracy: 0.9253\n",
            "Epoch 6/50\n",
            "294977/294977 [==============================] - 51s 173us/step - loss: 0.0898 - accuracy: 0.9679 - val_loss: 0.2376 - val_accuracy: 0.9212\n",
            "Epoch 7/50\n",
            "294977/294977 [==============================] - 51s 173us/step - loss: 0.0780 - accuracy: 0.9726 - val_loss: 0.2979 - val_accuracy: 0.9118\n",
            "Epoch 8/50\n",
            "294977/294977 [==============================] - 51s 174us/step - loss: 0.0687 - accuracy: 0.9760 - val_loss: 0.2536 - val_accuracy: 0.9206\n",
            "Epoch 00008: early stopping\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "krEYxUdd5t_0",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "0d6b0731-54b3-4159-c16c-93ff99d20ee7"
      },
      "source": [
        "import matplotlib.pyplot as plt\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', 'cv'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nO3deXyU5bXA8d/JAmEJBBL2EEgA2UQWI/suVMR9wZ2qreJttbW1tnXp1da2V29r21urVgFRsCoKKlLrAiqr7PuOQEhIwhYICVtCljn3j/cFhxjCAJm8ycz5fj7zYeZdZk6GZM48z/M+5xFVxRhjjCkrwusAjDHGVE+WIIwxxpTLEoQxxphyWYIwxhhTLksQxhhjymUJwhhjTLksQRgDiMgbIvKHAI9NF5ERwY7JGK9ZgjDGGFMuSxDGhBARifI6BhM6LEGYGsPt2vmliKwTkWMi8pqINBORT0XkiIh8ISKN/I6/VkQ2ikieiMwVkc5++3qKyCr3vHeBmDKvdbWIrHHPXSQilwQY41UislpEDotIpoj8tsz+ge7z5bn773G31xGRv4hIhojki8hCd9tQEckq530Y4d7/rYhMF5F/ichh4B4R6S0ii93X2CMiL4pILb/zu4rIbBHJFZF9IvKEiDQXkeMiEu93XC8RyRGR6EB+dhN6LEGYmuYmYCRwEXAN8CnwBNAE5/f5pwAichHwDvAzd98nwL9FpJb7YTkDeBNoDExznxf33J7AJOABIB54FZgpIrUDiO8Y8H0gDrgK+JGIXO8+bxs33n+4MfUA1rjnPQ9cCvR3Y/oV4AvwPbkOmO6+5ltAKfBzIAHoB1wO/NiNIRb4AvgMaAm0B75U1b3AXOAWv+cdC0xV1eIA4zAhxhKEqWn+oar7VDUbWAAsVdXVqloIfAj0dI+7FfiPqs52P+CeB+rgfAD3BaKB/1PVYlWdDiz3e41xwKuqulRVS1V1MnDCPa9CqjpXVderqk9V1+EkqSHu7juAL1T1Hfd1D6rqGhGJAH4APKyq2e5rLlLVEwG+J4tVdYb7mgWqulJVl6hqiaqm4yS4kzFcDexV1b+oaqGqHlHVpe6+ycBdACISCdyOk0RNmLIEYWqafX73C8p5XN+93xLIOLlDVX1AJtDK3Zetp1eqzPC73wb4hdtFkycieUBr97wKiUgfEZnjds3kA/+F800e9zl2lHNaAk4XV3n7ApFZJoaLRORjEdnrdjv9TwAxAHwEdBGRZJxWWr6qLjvPmEwIsARhQtVunA96AEREcD4cs4E9QCt320lJfvczgT+qapzfra6qvhPA674NzARaq2pD4BXg5OtkAu3KOecAUHiGfceAun4/RyRO95S/siWZ/wlsATqoagOcLjj/GFLKC9xthb2H04oYi7Uewp4lCBOq3gOuEpHL3UHWX+B0Ey0CFgMlwE9FJFpEbgR6+507AfgvtzUgIlLPHXyODeB1Y4FcVS0Ukd443UonvQWMEJFbRCRKROJFpIfbupkE/FVEWopIpIj0c8c8vgFi3NePBn4DnG0sJBY4DBwVkU7Aj/z2fQy0EJGfiUhtEYkVkT5++6cA9wDXYgki7FmCMCFJVbfifBP+B8439GuAa1S1SFWLgBtxPghzccYrPvA7dwVwP/AicAjY7h4biB8Dz4jIEeApnER18nl3AaNxklUuzgB1d3f3o8B6nLGQXOB/gQhVzXefcyJO6+cYcNpVTeV4FCcxHcFJdu/6xXAEp/voGmAvsA0Y5rf/a5zB8VWq6t/tZsKQ2IJBxhh/IvIV8LaqTvQ6FuMtSxDGmFNE5DJgNs4YyhGv4zHesi4mYwwAIjIZZ47Ezyw5GLAWhDHGmDOwFoQxxphyhUxhr4SEBG3btq3XYRhjTI2ycuXKA6padm4NEEIJom3btqxYscLrMIwxpkYRkTNezmxdTMYYY8plCcIYY0y5LEEYY4wpV8iMQZSnuLiYrKwsCgsLvQ4l6GJiYkhMTCQ62tZ2McZUjpBOEFlZWcTGxtK2bVtOL9wZWlSVgwcPkpWVRXJystfhGGNCREh3MRUWFhIfHx/SyQFARIiPjw+LlpIxpuqEdIIAQj45nBQuP6cxpuqEdBeTMcaEqpJSH1v3HWHVrjwiRbijT9LZTzpHliCCLC8vj7fffpsf//jH53Te6NGjefvtt4mLiwtSZMaYmiTnyAlW7zrE6sw8VmUcYl1WPgXFpQD0TIqzBFET5eXl8fLLL38nQZSUlBAVdea3/5NPPgl2aMaYaqqoxMemPYedhLArj1W7DpF1qACAqAiha8sG3HpZa3omxdErqRGJjeoEJQ5LEEH22GOPsWPHDnr06EF0dDQxMTE0atSILVu28M0333D99deTmZlJYWEhDz/8MOPGjQO+LR1y9OhRrrzySgYOHMiiRYto1aoVH330EXXqBOcXwhhT9fbkFziJIMNpIazPzqeoxAdAi4Yx9EyK4+5+benVJo6uLRsSEx1ZJXGFTYL43b83smn34Up9zi4tG/D0NV0rPOa5555jw4YNrFmzhrlz53LVVVexYcOGU5ejTpo0icaNG1NQUMBll13GTTfdRHx8/GnPsW3bNt555x0mTJjALbfcwvvvv89dd91VqT+LMaZqFBaXsiE7/1TLYPWuPPYedq5ArBUVwSWtGnJ3vzb0TGpEz6Q4WjT07stg2CSI6qJ3796nzVV44YUX+PDDDwHIzMxk27Zt30kQycnJ9OjRA4BLL72U9PT0KovXGHP+VJWsQwWnEsHqXYfYtOcwxaXOOjytG9ehd3JjeiXF0TOpEZ1bNKBWVPW5uDRsEsTZvulXlXr16p26P3fuXL744gsWL15M3bp1GTp0aLlzGWrXrn3qfmRkJAUFBVUSqzHm3Bw7UcK6rPxTCWFN5iEOHC0CoE50JN1bN+S+QSn0bO0khCaxtc/yjN4KmwThldjYWI4cKX/1xvz8fBo1akTdunXZsmULS5YsqeLojDHnS1VJO3DsVMtg1a48tu49jM9dpDMloR6DL2pCL7erqGOzWKIiq0/rIBCWIIIsPj6eAQMGcPHFF1OnTh2aNWt2at+oUaN45ZVX6Ny5Mx07dqRv374eRmqMqcjhwmLWZuaxKiOP1ZlOCyG/oBiA2NpR9EiKY+Sw9vRMakSP1nE0qlfL44gvXMisSZ2amqplFwzavHkznTt39iiiqhduP68xwVLqU7bvP+q2DJxksD3nKKogAh2a1j/VMuiZ1Ij2TeoTEVEzqxmIyEpVTS1vn7UgjDEGOFJYzNytOczatI+5W/dzpLAEgLi60fRsHcc13VvSK6kRl7RuSIOY8KiabAnCGBO29h0uZPamfczatI/FOw5QXKrE16vFlRc3p09yPD2T4khOqBe2tc4sQRhjwoaqsiPnKJ9v3MfsTftYk5kHQJv4utw7IJnvdWlGz6RGRNbQ7qLKZgnCGBPSfD5ldWYeszbtZfbGfaQdOAZA98SG/PKKjozs0owOTeuHbSuhIpYgjDEhp7C4lMU7DjpJYdN+Dhw9QVSE0K9dPPcOaMuILs08naFcUwQ1QYjIKODvQCQwUVWfK7O/DTAJaALkAnepapaIDAP+5ndoJ+A2VZ0RzHiNMTVX/vFi5mzdz2x3kPlYUSn1akUytFNTvtelGUM7NqVhnfAYXK4sQUsQIhIJvASMBLKA5SIyU1U3+R32PDBFVSeLyHDgWWCsqs4BerjP0xjYDswKVqzGmJppd14BX2zex6yN+1iSdpASn9IktjbX9WzFyC7N6N8untpRVVPYLhQFswXRG9iuqmkAIjIVuA7wTxBdgEfc+3OA8loINwOfqurxIMZqjKkBVJVv9h1l1sa9zNq0j/XZ+QC0a1KP+wal8L2uzeiRGFdj5yRUN8FMEK2ATL/HWUCfMsesBW7E6Ya6AYgVkXhVPeh3zG3AX8t7AREZB4wDSEqq/MUyKsuUKVN4/vnnERFSUlJYtWoVO3fuJCIigmPHjtGpUyfS0tKIjrbmrzFllfqUVbsOnUoKGQed74o9k+L49ahOjOzSjPZN63scZWjyepD6UeBFEbkHmA9kA6Und4pIC6Ab8Hl5J6vqeGA8ODOpK3ylTx+DvesrJehTmneDK5+r8JCNGzfyhz/8gUWLFpGQkEBubi733nsv8+bNY9iwYXz88cdcccUVlhyM8VNYXMrCbQeYtWkvX27ez8FjRdSKjKB/+3geGNyOEZ2b0rRBjNdhhrxgJohsoLXf40R32ymquhunBYGI1AduUtU8v0NuAT5U1eIgxhlUX331FWPGjCEhIQGAxo0bc+utt/Luu+8ybNgwpk6des7LkRoTig4dK+KrLfuZtWkv8785QEFxKbExUQzr2JTvdW3GkIuaEBsmM5iri2AmiOVABxFJxkkMtwF3+B8gIglArqr6gMdxrmjyd7u7/cKd5Zt+Vbr22mt54oknyM3NZeXKlQwfPtzrkIzxRNah485M5o37WJaeS6lPad4ghpsvTeR7XZvRJzm+Wq2PEG6CliBUtUREHsLpHooEJqnqRhF5BlihqjOBocCzIqI4XUwPnjxfRNritEDmBSvGqjB8+HBuuOEGHnnkEeLj48nNzaVx48ZcdtllPPzww1x99dVERtpVFiY8qCqb9xxh1qa9zNq4j017nFUeL2pWnx8NacfILs3o1qqhDTJXE0Edg1DVT4BPymx7yu/+dGD6Gc5NxxnortG6du3Kk08+yZAhQ4iMjKRnz5688cYb3HrrrYwZM4a5c+d6HaIxQVVYXMqK9EOnuo+yDhUgApcmNeKJ0Z0Y2aU5yQn1zv5Epsp5PUgdFu6++27uvvvu07bdfPPNhEqpdWP8lfqUTbsPs3D7ARZuz2F5+iGKSnzUiopgUPsEfjK8PcM7Nav2q6kZSxDGmEqw6+BxFm4/wNfbD/D1jgPkHXeuK+nUPJaxfdswsH0CvZMbU6+2feTUJPa/ZYw5Z4eOFbFox0EWbs9h4fYDZOY666Q3bxDDiM7NGNg+gf7t42kaa5ei1mQhnyBUNSyqNFp3lQmmk+MIC7bn8PX2A2zcfRhVZ6nNvu3iuW9gCgM7JJASxmsnhKKQThAxMTEcPHiQ+Pj4kP6lVVUOHjxITIx9WzOV4+Q4wsmEcHIcITpS6JnUiJ+PuIiBHRK4pFVDoiLtMtRQFdIJIjExkaysLHJycrwOJehiYmJITEz0OgxTg+06ePxUQli04+B3xxE6JNC7rY0jhJOQ/p+Ojo4mOTnZ6zCMqZZyjxWxaIczsOw/jtCioTOOMKhDAv3a2ThCOAvpBGGM+VZhcSnL03NPXW1Udhzh/kEpDGhv4wjmW5YgjAlRpT5l4+78Uwmh7DjCIyMuYoCNI5gKWIIwJkSoKrtyv52PYOMI5kLZb4kxNZj/OMKCbQfIOvTtOMLIzs0YaOMI5gJYgjCmhjl2ooQZa7J5d3km67KcFdVia0fRr1084wbbOIKpPJYgjKkhdh44xpuLM5i2MpMjhSV0adGAX4y0cQQTPJYgjKnGfD5l3jc5TF6cztytOURFCKO7teDu/m3oldTIWgkmqCxBGFMN5RcUM21FJm8uySDj4HGaxNbmZyM6cEfvJFtq01QZSxDGVCNb9h5m8qIMZqzOpqC4lNQ2jfjF9zoyqmtzW1nNVDlLEMZ4rKTUx6xN+5i8KJ2lO3OpHRXBdT1a8v1+bbm4VUOvwzNhzBKEMR45cPQEU5ft4q2lu9iTX0hiozo8fmUnbkltTaN6tbwOzxhLEMZUtTWZeUxZlM7H6/ZQVOpjUIcEnrnuYoZ3akqkrcVsqhFLEMZUgRMlpfxn3R4mL0pnbVY+9WpFcnvv1ozt15b2Tet7HZ4x5QpqghCRUcDfgUhgoqo+V2Z/G2AS0ATIBe5S1Sx3XxIwEWgNKDBaVdODGa8xlW13XgFvLc1g6rJMDh4rIqVJPX53bVdu7NWK2Jhor8MzpkJBSxAiEgm8BIwEsoDlIjJTVTf5HfY8MEVVJ4vIcOBZYKy7bwrwR1WdLSL1AV+wYjWmMqkqS9JymbI4nVmb9uFT5fJOzbinf1sGtA/txatMaAlmC6I3sF1V0wBEZCpwHeCfILoAj7j35wAz3GO7AFGqOhtAVY8GMU5jKsXxohI+XJ3NlEUZbN13hLi60dw3KJm7+rShdeO6XodnzDkLZoJoBWT6Pc4C+pQ5Zi1wI0431A1ArIjEAxcBeSLyAZAMfAE8pqql/ieLyDhgHEBSUlIwfgZjzir9wDHeXJLBeyu+LYHxp5su4doeLYmJjvQ6PGPOm9eD1I8CL4rIPcB8IBsoxYlrENAT2AW8C9wDvOZ/sqqOB8YDpKamalUFbYzPp8zblsPkRd+WwLiyWwvusRIYJoQEM0Fk4wwwn5TobjtFVXfjtCBwxxluUtU8EckC1vh1T80A+lImQRhT1awEhgknwUwQy4EOIpKMkxhuA+7wP0BEEoBcVfUBj+Nc0XTy3DgRaaKqOcBwYEUQYzWmQlv2HmbK4gw+XGUlMEz4CFqCUNUSEXkI+BznMtdJqrpRRJ4BVqjqTGAo8KyIKE4X04PuuaUi8ijwpTht9ZXAhGDFakx5Skp9zN60j8mL01mSZiUwTPgR1dDouk9NTdUVK6yRYS5c2RIYreLqMLZfG261EhgmBInISlVNLW+f14PUxlQb67Pyef3rnadKYAxsbyUwTHizBGHCms+nfLVlPxMWpLF0Z65fCYw2tG8a63V4xnjKEoQJS4XFpXy4OpsJC9JIyzlGy4YxPDm6M7f2bk0DK4FhDGAJwoSZ3GNFvLk4gymL0zl4rIiLWzXg77f1YHS3FkTbms7GnMYShAkLaTlHeW3hTqavzOJEiY/hnZpy36Bk+qVYbSRjzsQShAlZqsry9ENMWJDGF5v3ER0RwY29WnHfoGQbXzAmAJYgTMgpKfXx2ca9TFiwk7WZeTSqG81PhrVnbL+2NImt7XV4xtQYliBMyDh6ooT3lmcy6eudZB0qoG18XX5//cXc3CuROrWsaJ4x58oShKnx9uYX8saidN5emsHhwhJS2zTiv6/uwojOzWz+gjEXwBKEqbE27znMhAVp/Hvtbkp9yqiLm3PfoBR6JTXyOjRjQoIlCFOjqCoLth1gwoI0Fmw7QN1akdzZpw0/GJBMUrwtymNMZbIEYWqEohIfM9fuZuKCNLbsPULT2Nr88oqO3Nknibi6Vh/JmGCwBGGqtfzjxby1LIM3vk5n/5ETdGwWy/NjunNN9xbUjrKBZ2OCyRKEqZYyc4/z2sKdvLcik+NFpQzqkMCfx3RncIcEm9hmTBWxBGGqldW7DjFxwU4+3bCHyAjhmu4tuW9gCl1aNvA6NGPCjiUI47lSn/LF5n1MXJDG8vRDxMZEMW5wO+7p35bmDW0ZT2O8YgnCeKagqJTpq7KYtHAnOw8cI7FRHZ66ugu3XNaa+rXtV9MYr9lfoalyOUdO8ObidN5cksGh48V0T2zIi3f0ZFTX5kRZRVVjqg1LEKbKbN9/hIkLdvLB6myKS32M6NyM+welcFnbRjbwbEw1FNQEISKjgL8DkcBEVX2uzP42wCSgCZAL3KWqWe6+UmC9e+guVb02mLGa4FBVlqTlMmFBGl9t2U/tqAjGXJrIDwcmk9KkvtfhGWMqELQEISKRwEvASCALWC4iM1V1k99hzwNTVHWyiAwHngXGuvsKVLVHsOIzwbchO58nP1zP2qx84uvV4ucjLuKuvknE17eKqsbUBMFsQfQGtqtqGoCITAWuA/wTRBfgEff+HGBGEOMxVaS41MfLc3bwj6+20bheLZ69sRs39GxFTLRNbDOmJglmgmgFZPo9zgL6lDlmLXAjTjfUDUCsiMSr6kEgRkRWACXAc6r6neQhIuOAcQBJSUmV/xOYc7Zt3xF+MW0t67Lyub5HS357bVcrhWFMDeX1IPWjwIsicg8wH8gGSt19bVQ1W0RSgK9EZL2q7vA/WVXHA+MBUlNTterCNmX5fMqkr3fyp8+3Uq9WJC/f2YvR3Vp4HZYx5gIEM0FkA639Hie6205R1d04LQhEpD5wk6rmufuy3X/TRGQu0BM4LUGY6mHXweM8On0ty3bmMqJzM569sZut3GZMCAhmglgOdBCRZJzEcBtwh/8BIpIA5KqqD3gc54omRKQRcFxVT7jHDAD+FMRYzXlQVd5etos//mczkSI8P6Y7N/VqZZesGhMigpYgVLVERB4CPse5zHWSqm4UkWeAFao6ExgKPCsiitPF9KB7emfgVRHxARE4YxCbvvMixjN78wv59fvrmPdNDgPax/Onm7vTKq6O12EZYyqRqJ69615EPgBeAz51v+1XO6mpqbpixQqvwwh5qspHa3bz1EcbKCr18cToztzVpw0RtrSnMTWSiKxU1dTy9gXagngZuBd4QUSmAa+r6tbKCtDUDAePnuA3Mzbw6Ya99EqK4y+39CA5oZ7XYRljgiSgBKGqXwBfiEhD4Hb3fiYwAfiXqhYHMUZTDczauJcnPlzP4YISfj2qE+MGpxBprQZjQlrAYxAiEg/chTPTeTXwFjAQuBtnLMGEoPyCYp759ybeX5VFlxYN+Nd93enU3NZmMCYcBJQgRORDoCPwJnCNqu5xd73rTmYzIWjhtgP8cvpa9h85wU+Gt+cnwztQK8qqrRoTLgJtQbygqnPK23GmwQ1Tcx0vKuG5T7cwZXEG7ZrU4/0f9adH6zivwzLGVLFAE0QXEVl9chKbO0/hdlV9OXihGS+szMjlF++tJf3gcX4wIJlfjepoNZSMCVOB9hfcfzI5AKjqIeD+4IRkvHCipJRnP93MmFcWU+JT3rm/L09d08WSgzFhLNAWRKSIiLqTJtxS3laBLURsyM7nF++tZeu+I9zeuzVPXtXFlvw0xgScID7DGZB+1X38gLvN1GAlpT5enruDF750ynK/fs9lDOvU1OuwjDHVRKAJ4tc4SeFH7uPZwMSgRGSqxPb9R/jFe2tZm5XPtd1b8sx1VpbbGHO6QCfK+YB/ujdTg50sy/3nz7dSt1YkL93Ri6susbLcxpjvCnQeRAec5UC7ADEnt6tqSpDiMkGQmXucR6etZenOXEZ0bsr/3NiNprExZz/RGBOWAu1ieh14GvgbMAynLpPNmKohVJWpyzP5w8ebEBH+dPMljLk00cpyG2MqFGiCqKOqX7pXMmUAvxWRlcBTQYzNVIJ9h52y3HO35tC/XTx/uvkSEhvV9TosY0wNEGiCOCEiEcA2d42HbKB+8MIyF0pVmbl2N099tJETJaX87tqujO1rZbmNMYELNEE8DNQFfgr8Hqeb6e5gBWUuTO6xIn4zYz2frN9Lz6Q4/jKmOylNLJ8bY87NWROEOynuVlV9FDiKM/5gqqkvNu3jsQ/Wk19QxK9GdWTcoBSiIm24yBhz7s6aIFS1VEQGVkUw5vwdLnTKck9fmUXnFg1484e96dzCynIbY85foF1Mq0VkJjANOHZyo6p+EJSozDn5evsBfjltLXsPF/LQsPb89HIry22MuXCBJogY4CAw3G+bAhUmCBEZBfwdiAQmqupzZfa3ASYBTYBc4C5VzfLb3wDYBMxQ1YcCjDVsFBSV8tynm5m8OIOUBKcsd8+kRl6HZYwJEYHOpD7ncQd37OIlYCSQBSwXkZmqusnvsOeBKao6WUSG40zGG+u3//fA/HN97XCwMuMQj05by84Dx7h3QFt+dUUn6tSyyqvGmMoT6Ezq13FaDKdR1R9UcFpvYLuqprnPMRW4DqdFcFIX4BH3/hxght9rXgo0wykKaIsSuU6UlPJ/X2zj1Xk7aNGwDm/f34f+7RK8DssYE4IC7WL62O9+DHADsPss57QCMv0eZwF9yhyzFrgRpxvqBiDWXfv6EPAXnDWwR5zpBURkHDAOICkp6aw/RCh4+qONTF2eya2prfnN1Z2JjYn2OiRjTIgKtIvpff/HIvIOsLASXv9R4EURuQenKykbKAV+DHyiqlkVlYNQ1fHAeIDU1NTvtHBCzd78Qt5flcXYvm34/fUXex2OMSbEne+qMB2Asy0ckA209nuc6G47RVV347QgEJH6wE2qmici/YBBIvJjnBnbtUTkqKo+dp7xhoTXF+2k1KeMG2w1Eo0xwRfoGMQRTh+D2IuzRkRFlgMdRCQZJzHcBtxR5nkTgFy3nPjjOFc0oap3+h1zD5Aa7snhSGExby/ZxZXdWtC6sdVSMsYEX6BdTLHn+sSqWuLWbfoc5zLXSaq6UUSeAVao6kxgKPCsiChOF9OD5/o64WLqskyOnChh3CBrPRhjqoa4y0xXfJDIDcBXqprvPo4DhqrqjIrPrDqpqam6YsUKr8MIiuJSH4P/NIekxnV594F+XodjjAkhIrJSVcu9UjTQ6bZPn0wOAKqah7M+hKkC/167mz35hTwwxFoPxpiqE2iCKO+48x3gNudAVRk/P40OTesz9KKzXRdgjDGVJ9AEsUJE/ioi7dzbX4GVwQzMOBZsO8CWvUe4f3CKreVgjKlSgSaInwBFwLvAVKAQG1CuEuPnp9E0tjbX9WjpdSjGmDAT6FVMx4CwvszUCxuy81m4/QC/GtWR2lFWZ8kYU7UCakGIyGz3yqWTjxuJyOfBC8sATFiQRr1akdzZp43XoRhjwlCgXUwJ7pVLAKjqIc4+k9pcgOy8Aj5et4fbeifRsI7VWzLGVL1AE4RPRE5VwxORtpRT3dVUnkkLdwLwg4HJHkdijAlXgV6q+iSwUETmAQIMwq2iaipffkExU5ft4upLWtAqro7X4RhjwlSgg9SfiUgqTlJYjbNuQ0EwAwtnby3N4FhRqRXlM8Z4KtBiffcBD+NUZF0D9AUWc/oSpKYSnCgp5fWv0xnYPoGuLRt6HY4xJowFOgbxMHAZkKGqw4CeQF7Fp5jz8dGa3eQcOWGtB2OM5wJNEIWqWgggIrVVdQvQMXhhhSefT5kwP43OLRowqIMtI2qM8Vagg9RZ7jyIGcBsETkEZAQvrPA095v9bNt/lL/d2p2KVtIzxpiqEOgg9Q3u3d+KyBygIfBZ0KIKU6/OS6NFwxiuvsTKahhjvHfOFVlVdV4wAgl3azPzWLozlydHdyY6MtCeP2OMCR77JKomxi9II7Z2FLf1bn32g40xpgpYgqgGdh08zqfr93BH3yRiY6yshjGmeg4QiH8AABp5SURBVLAEUQ28tjCNyAjh3v5WVsMYU33YqnAeO3SsiPdWZHFt91Y0bxjjdTiVRxV2r4ZvPoPoOtCgFTRo6dxiW0J0CP2sxoSooCYIERkF/B2IBCaq6nNl9rcBJgFNgFzgLlXNcrd/iNPCiQb+oaqvBDNWr7y5JIOC4hAqq5G7E9ZPh3XvwsFtOKW7yqnrWDfeTRh+iePU/VYQ2wJq16/q6I0xfoKWIEQkEngJGAlkActFZKaqbvI77HlgiqpOFpHhwLPAWGAP0E9VT4hIfWCDe+7uYMXrhcLiUiYvSmdoxyZ0bB7rdTjn73gubPwA1r0HmUudbW0GQP+HoMt1EBENR/bA4Ww4vNvvX/d+1nI4fvC7zxvT8AwJxO9+7QZgc0aMCYpgtiB6A9tVNQ1ARKYC1wH+CaIL8Ih7fw7ORDxUtcjvmNqE6FjJB6uyOXisqGa2HooLYOunTlLYPht8JdCkE1z+NHS7GeKSTj++dgdI6FDx8x3Zc3ri8L+/dz0c3ffd82rVrziBNGgFdRpZEjHmPAQzQbQCMv0eZwF9yhyzFrgRpxvqBiBWROJV9aCItAb+A7QHflle60FExuGWHU9KSiq7u1rz+ZSJC9Lo1qoh/VLivQ4nML5SSF/gJIVNM6HoiNMV1PdH0O0WaN7t/D+Io+tA4xTndiYlRXB0b/kJ5PBuSJvrJBn1nX5eVEwFCcS9XzcBIkLye4gx583rQepHgRdF5B5gPpANlAKoaiZwiYi0BGaIyHRVPe0rpKqOB8YDpKam1qgFjGZv3kfagWO8cHvP6l1WQ9X59r7uXdjwvvMBXCvW6Tq6ZAy0HQQRVbRedlQtp2VStnXir7QEju0vk0T8ksmuxXB4D/iKTz8vIhoatID4DtB2ICQPhhY9INLrPxFjvBPM3/5swH/WV6K77RS3VXAjgDvWcJP/0qYnjxGRDTiLFE0PYrxVavz8NBIb1WH0xc29DqV8ebtg/TSntZCzBSKioMP3oNv/QMcrnW/81VFk1LctA1LLP8bng+MHvtsKyc+GfRvgy985x9WKhTb9IXmQkwibd6u6ZGhMNRDMBLEc6CAiyTiJ4TbgDv8DRCQByFVVH/A4zhVNiEgicFBVC0SkETAQ+FsQY61SKzNyWZlxiKev6UJUdSqrUXAINs5wEkPG18621n3hqr9AlxugXg3pCjubiAio39S5tez53f1HcyBjIeycDzsXwLbPne0xDaHNQCdhJA+GJp2tW8qEtKAlCFUtEZGHgM9xLnOdpKobReQZYIWqzgSGAs+KiOJ0MT3ont4Z+Iu7XYDnVXV9sGKtauPnp9GwTjS3pFaDshrFhc4H4Lr3YNssKC1yulmG/cYZbG4chpP36jeBrjc4N3C6pNIXws55zhjM1v842+vGO91Rbd2EkXCRDYabkCKqNarr/oxSU1N1xYoVXodxVmk5R7n8r/N4cGh7Hr3CoyU1fD6nhbD+Pdj4EZzIh3pNnYRwyS1O37t90J1ZXqaTKHYucP7Nd6/FqN/MTRZul1TjFHsfTbUnIitVtdz+WBuBq2ITF+4kOiKCu/u3rfoX37fRaSmsnw6HsyC6HnS+xkkKyUNsQDZQca2hxx3OTRUO7fw2WexcABvcobIGrZyWxcmkUdHgujHVkH0iVKEDR08wfWUWN/ZqRZPY2lXzovnZzgfWuvecAViJhPYjYOTvnMHmWvWqJo5QJfLt5bmX3u0kjIPbne6onQtg22xY+45zbFwbt3Ux2Pm3ga37Yao3SxBVaMriDIpKfNw3KMgT4wrznXkK6951+s5RSLwMrvyz069ev0lwXz+ciTgTAhM6wGX3OQlj/2a3dTEfNn8Mq//lHBvf/vQuqfpNvY3dmDIsQVSRgqJS3lyczojOzWjfNAg1hkqKnBnN696FrZ9B6QnnW+3Qx6DbGIhvV/mvac5OBJp1cW59HnAmG+7b8G2X1Ib3YeXrzrFNOn+bLNoOhLqNvY3dhD1LEFVk2spMDh0v5oEhldh68Pmc2kfr3oVNM5zLVOsmwKX3wCW3QqteNkha3UREQovuzq3/Q87Evj1rId29pHb1W7BsPCDQ7OJvL6lt09+5zNaYKmQJogqU+pSJC3bSMymO1DaNLvwJc7Y6SWH9NGdCW1Qd6Hy1kxRShkKkLTpUY0RGQeKlzm3gz6G0GLJXOd1R6fNhxSRY8jJIhJNU2g5yLihI6mvVbk3QWYKoAp9t2Muu3OM8fmWn8y+rUVIEm2fC0lcha5nzgZEyzJmv0Okq+7AIFZHRkNTHuQ35pTNPJXuF07rYOR+W/BMWveDMbG/ZC1KGOL8HiZc5pUiMqUQ2DyLIVJXrX/qa/IJivvzFUCIjzjFBHM2BlW/A8olOobrGKZD6Q2dcIbZZUGI21VjRcadb8eSgd/Yq0FKIruuUWG83zGlFNu1i3YsmIDYPwkPLduayNiuf319/8bklh92rYel45xLV0iJodzlc+w/nElUr7xC+atV1kkC7Yc7jwnznSrW0uc7t8yec7fWaftu6SBkKDVt5E6+p0SxBBNn4+Wk0rleLMZcmnv3g0mLY/G+nGylziTORrdfd0HscNLko+MGamiemodPF2Okq53F+FqTN+zZhrJ/mbI/v4CSKdsOcK6RswNsEwBJEEG3bd4Qvt+znZyM6EBNdQRXQYwfcbqTX4MhuaNQWrngWet5pf8jm3DRMdH5vet7pzsHY9G2yWPM2LJ/gjF+1utRJGDZ+YSpgCSKIJixIo3ZUBN/v17b8A/asdbqR1k9z5i2kDIOr/wYdRlpZaXPhRKBZV+fW70HnQoes5d8mjAV/hfl//nb8ImWo08Kw8QvjsgQRJPsPFzJj9W5uuSyRxvX8vp2VlsCWj51upF2LnD/Onnc53UhNO3kXsAl9UbWg7QDnNvzJ745fzHrSOe7U+MVQd/wigO5RE5IsQQTJG4vSKfb5uG+gOzHu2EFYNdm5GulwtlOX53t/dJJDnThvgzXh6VzHL1KGOhP3rNszbFiCCIKjJ0r415IMRnVtTtuSNPjoVeePraTQmeQ0+nm46ArrRjLVy3fGLzZD2hwbvwhjNg8iCCbN38bSz/7F80mLid271Jnp3P02pxupWRevwzPm3JUdv8heefr8i5ShNn5RQ1U0D8ISRGU6nkvpysnkfPkSzcmBhknQ+36nG8kKr5lQUnb84sA3znYbv6hxbKJcsO3b6Aw6r3uPyJIC0kq7sHfwb+lx+e3WjWRC03fGL7K/TRZnGr9IGQK1Yz0I1pwva0GcL18pbP0Ulr7ilD2IikEvuZUHt1/GVm3N7J8PIeJcy2oYEwrKjl+kfw3Fx6BOIxj5e+hxp1UDqEYqakEE9X9JREaJyFYR2S4ij5Wzv42IfCki60Rkrogkutt7iMhiEdno7rs1mHGek4JD8PUL8EIPePdOOJQOI34Hj2xmUZen+GR/Y+4flGLJwYSvk2tg9HsQ7pwGv06Huz+GJp1g5kPwxmjYt8nrKEPHgW3O1WdBELQWhIhEAt8AI4EsYDlwu6pu8jtmGvCxqk4WkeHAvao6VkQuAlRVt4lIS2Al0FlV8870ekFvQezf7HYjvQvFx6HNQGcBmI6jT63l/P1Jy9i0+zALfz2s4pnTxoQjnw/Wvg2z/htOHIZ+D8GQX9myt+erIA/m/QmWvQqN28GDS8/rAgGvxiB6A9tVNc0NYipwHeD/1aEL8Ih7fw4wA0BVvzl5gKruFpH9QBPgjAkiKHyl8M3nTjfSznkQFeNUUe3zADTvdtqhm/ccZv43Ofzyio6WHIwpT0SEc8HGRVfCF0/B1/8HGz6A0X+GjqO8jq7m8JU6pXnm/BGO50KvsTD8v4Ny9VgwE0QrINPvcRbQp8wxa4Ebgb8DNwCxIhKvqgdPHiAivYFawI6yLyAi44BxAElJSZUXeUGes27wsvGQlwENWsHlTzuF8+rFl3vKhAVp1K0VyZ19KjEOY0JRvXi47iVnLOLjR+CdW6HT1XDl/9pVT2eTNg8+exz2b3QuLx71rLOQVJB4fRXTo8CLInIPMB/IBkpP7hSRFsCbwN2q6it7sqqOB8aD08V0wdHkbHW6kdZOdQbVkvrDyGecX97IM79Ve/ILmLlmN3f1bUNcXZs0ZExA2vSHB+bDkpdg7v/Ci71h2OPQ579sVcSycnfCrN84ZXrikmDMZOhyXdDnnAQzQWQDrf0eJ7rbTlHV3TgtCESkPnDTyXEGEWkA/Ad4UlWXBC1Knw+2zXK6kdLmQGRttxtpXMCZ+fWv01HghwOTgxamMSEpqpaz1GrXG+HTXzkfgmunOkUrW/f2OjrvnTgCC/4Ci1+CiGinK6nfQxAdUyUvH8wEsRzoICLJOInhNuAO/wNEJAHIdVsHjwOT3O21gA+BKao6PYgxOl1I79wGsS2cN//Se6BeQsCnHy4s5u2luxjdrQWtG9cNXpzGhLJGbeD2qbDlP06ieG2k87d4+dPhOcn05ID+l8/A0X3Q/XbnvWjQokrDCFqCUNUSEXkI+ByIBCap6kYReQZYoaozgaHAsyKiOF1MD7qn3wIMBuLd7ieAe1R1TaUH2jgZ7vkYWvc5r2btO0t3cfRECQ8MTqn00IwJKyLQ+WpnUt3cZ531tzd/DFf8ES65NXxKeGQshs8egz1rnFpXt70DiZd6EopNlLsARSU+Bv9pDilN6vH2/X2r9LWNCXl718PHP3dqQLUdBFf9NbRXVszLhNlPwcYPILYljPyd090d5MTo2US5UPfvtbvZe7iQ+631YEzla94NfjALrv4/2LsO/tkfvvw9FBd4HVnlKjoGc/4HXkyFrZ/AkF/DT1bAJbd43mry+iqmGktVmbAgjY7NYhl6UROvwzEmNEVEQOq9zpWEs34DC56HDdNh9F+gwwivo7swqk7NqtlPO0sNX3yTU5UhrvXZz60i1oI4T/O+yWHL3iPcPzgFCZe+UWO8Ur8J3Pgq3P1v52qet26C9+6Gw7u9juz8ZK+E174HH9wP9ZvCvZ/BzZOqVXIASxDnbfz8NJo3iOHa7i29DsWY8JE8GH70NQz7DXzzmTN3YskrzlK+NcHhPfDhf8GE4U4dt+tegvvnQJt+XkdWLksQ52FDdj6Ldhzk3gFtqRVlb6ExVSqqNgz5Jfx4sTNX4rNfw8Thzrfy6qq4AOb/Gf5xKWx435n78dNVTumRalzZtvpGVo2Nn59G/dpR3G5lNYzxTuMUuOt9GPMGHNkHEy6H//zCKZVTXajCxhlOS+erPzir7j24DEb8tkasjWGD1Oco69Bx/rN+Dz8Y0JYGMVYOwBhPiUDXG6Dd5U7xumXjYdNMp0bRxTd5exXQnnXOfIaMr6HZxXD9v50ushrEWhDn6LWFOxHg3gFWVsOYaiOmgVPs7/6voGEreP+H8Ob1cPA7NT6D72gOzPwpvDoYcrY4ZUMemF/jkgNYgjgn+ceLeXd5Jtd2b0nLuDpeh2OMKatlT7jvSxj9PGSvgpf7wdznoLgw+K9dUuQsJvaPXrDmLej7Y/jJKkj9QY1deti6mM7Bv5ZmcLyo1CbGGVOdRURC7/uh8zXw+RNO2Y5178FVf3HGACqbqrP88KwnITcNOlzhlAdJ6FD5r1XFrAURoBMlpbyxKJ1BHRLo3KKB1+EYY84mtrkzt+CuD0B9TpfT9B86A9qVZd8m53mn3u7Mz7jrfbjzvZBIDmAJImAzVmeTc+QEDwxu53Uoxphz0f5y+PESGPIYbJ4JL14GyyY4K7Odr+O58J9H4ZWBsHs1jPpfZ35G+xo+u7sMSxAB8PmU8fPT6NKiAQPal7+inDGmGouOcRYj+tFiaNkDPnkUJo6A3edYILq02JmY90JPWPGaM77w0zXQNzQXObIEEYCvtuxnR84xHhhiZTWMqdES2sP3P4IbJ0J+JkwYBp8+BoWHz37uti/gnwOciXkte8B/fQ1XPR/S61VYggjA+AVptIqrw+huVbtYhzEmCETgkjHw0AqnBbD0FXiptzOhrbzlDw5sg7duceo/+Yqd9RnGzoBmXao+9ipmCeIs1mTmsWxnLj8YmEx0pL1dxoSMOnHOlU33femsIjntbnjrZmf9Z3BmZH/2BLzcF3YthpG/d8YyOo32vAx3VbHLXM9i/PwdxMZEcetl1avKojGmkiReCvfPheUTnHIYL/d1lvjcPNMZjO411lmOuH5TryOtcpYgKpBx8BifbdjLA0PaUb+2vVXGhKzIKOj7I+hynVMeY+Xr0GaAU7KjRXevo/OMfepVYOKCnURFRHBv/7Zeh2KMqQoNWsItU5yWQ51GYdOVdCZB7VQXkVEislVEtovIY+XsbyMiX4rIOhGZKyKJfvs+E5E8Efk4mDGeSe6xIqatzOT6ni1p2iDGixCMMV6p2zjskwMEMUGISCTwEnAl0AW4XUTKDvs/D0xR1UuAZ4Bn/fb9GRgbrPjO5s3FGRQW+7h/kJXVMMaEp2C2IHoD21U1TVWLgKnAdWWO6QJ85d6f479fVb8EjgQxvjMqLC5lyuJ0hndqSodm1b9muzHGBEMwE0QrINPvcZa7zd9a4Eb3/g1ArIgEPFVZRMaJyAoRWZGTk3NBwfqbvjKLg8eKGGdF+YwxYczrC/sfBYaIyGpgCJANBFwgRVXHq2qqqqY2adKkUgIq9SkTF6TRPbEhfZJDd4akMcacTTCvYsoG/CcPJLrbTlHV3bgtCBGpD9ykqp6uFzh70z7SDx7npTt6WVkNY0xYC2YLYjnQQUSSRaQWcBsw0/8AEUkQkZMxPA5MCmI8ARk/fwetG9dh1MXNvQ7FGGM8FbQEoaolwEPA58Bm4D1V3Sgiz4jIte5hQ4GtIvIN0Az448nzRWQBMA24XESyROSKYMV60or0XFbtyuO+gSlERljrwRgT3oI6UU5VPwE+KbPtKb/704HpZzh3UDBjK8+r89OIqxvNmNTEsx9sjDEhzutB6mpjR85Rvti8j+/3bUPdWjbB3BhjLEG4Ji7YSa3ICL5vZTWMMQawBAFAzpETvL8qi5suTSShfm2vwzHGmGrBEgQwZXE6xaU+7huY7HUoxhhTbYR9gjheVMKbSzIY2bkZKU3qex2OMcZUG2E/GnuksIQB7RL4wcC2XodijDHVStgniGYNYnjpzl5eh2GMMdVO2HcxGWOMKZ8lCGOMMeWyBGGMMaZcliCMMcaUyxKEMcaYclmCMMYYUy5LEMYYY8plCcIYY0y5RFW9jqFSiEgOkHEBT5EAHKikcIKtJsUKNSvemhQr1Kx4a1KsULPivZBY26hqk/J2hEyCuFAiskJVU72OIxA1KVaoWfHWpFihZsVbk2KFmhVvsGK1LiZjjDHlsgRhjDGmXJYgvjXe6wDOQU2KFWpWvDUpVqhZ8dakWKFmxRuUWG0MwhhjTLmsBWGMMaZcliCMMcaUK+wThIiMEpGtIrJdRB7zOp6KiMgkEdkvIhu8juVsRKS1iMwRkU0islFEHvY6poqISIyILBORtW68v/M6prMRkUgRWS0iH3sdy9mISLqIrBeRNSKywut4KiIicSIyXUS2iMhmEenndUxnIiId3ff05O2wiPys0p4/nMcgRCQS+AYYCWQBy4HbVXWTp4GdgYgMBo4CU1T1Yq/jqYiItABaqOoqEYkFVgLXV+P3VoB6qnpURKKBhcDDqrrE49DOSEQeAVKBBqp6tdfxVERE0oFUVa32E89EZDKwQFUnikgtoK6q5nkd19m4n2fZQB9VvZBJw6eEewuiN7BdVdNUtQiYClzncUxnpKrzgVyv4wiEqu5R1VXu/SPAZqCVt1GdmTqOug+j3Vu1/fYkIonAVcBEr2MJJSLSEBgMvAagqkU1ITm4Lgd2VFZyAEsQrYBMv8dZVOMPsZpKRNoCPYGl3kZSMbfLZg2wH5itqtU53v8DfgX4vA4kQArMEpGVIjLO62AqkAzkAK+73XcTRaSe10EF6Dbgncp8wnBPECbIRKQ+8D7wM1U97HU8FVHVUlXtASQCvUWkWnbjicjVwH5VXel1LOdgoKr2Aq4EHnS7S6ujKKAX8E9V7QkcA6r12CSA2xV2LTCtMp833BNENtDa73Giu81UArcv/33gLVX9wOt4AuV2KcwBRnkdyxkMAK51+/WnAsNF5F/ehlQxVc12/90PfIjTvVsdZQFZfq3H6TgJo7q7Elilqvsq80nDPUEsBzqISLKbgW8DZnocU0hwB31fAzar6l+9judsRKSJiMS59+vgXLiwxduoyqeqj6tqoqq2xfmd/UpV7/I4rDMSkXruhQq43TXfA6rllXiquhfIFJGO7qbLgWp5YUUZt1PJ3UvgNKfClqqWiMhDwOdAJDBJVTd6HNYZicg7wFAgQUSygKdV9TVvozqjAcBYYL3brw/whKp+4mFMFWkBTHavBIkA3lPVan/5aA3RDPjQ+c5AFPC2qn7mbUgV+gnwlvulMQ241+N4KuQm3ZHAA5X+3OF8masxxpgzC/cuJmOMMWdgCcIYY0y5LEEYY4wplyUIY4wx5bIEYYwxplyWIIypBkRkaE2oymrCiyUIY4wx5bIEYcw5EJG73HUj1ojIq26Bv6Mi8jd3HYkvRaSJe2wPEVkiIutE5EMRaeRuby8iX7hrT6wSkXbu09f3W4fgLXc2ujGesQRhTIBEpDNwKzDALepXCtwJ1ANWqGpXYB7wtHvKFODXqnoJsN5v+1vAS6raHegP7HG39wR+BnQBUnBmoxvjmbAutWHMObocuBRY7n65r4NTGtwHvOse8y/gA3ddgThVnedunwxMc2sStVLVDwFUtRDAfb5lqprlPl4DtMVZuMgYT1iCMCZwAkxW1cdP2yjy32WOO9/6NSf87pdif5/GY9bFZEzgvgRuFpGmACLSWETa4Pwd3ewecwewUFXzgUMiMsjdPhaY566ulyUi17vPUVtE6lbpT2FMgOwbijEBUtVNIvIbnJXRIoBi4EGcRWV6u/v244xTANwNvOImAP+qoGOBV0XkGfc5xlThj2FMwKyaqzEXSESOqmp9r+MwprJZF5MxxphyWQvCGGNMuawFYYwxplyWIIwxxpTLEoQxxphyWYIwxhhTLksQxhhjyvX/K31rdywZjGYAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "C87DzTRG5uG3",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "f9c35be3-3673-4c6c-b0b8-a9e687916cd7"
      },
      "source": [
        "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', 'cv'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 56,
      "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"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EOgGcsuIgCUj",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "prob = model.predict(x_test)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WiVJIXpD2UvB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "y_pred = []\n",
        "for i in range(prob.shape[0]):\n",
        "  if(prob[i][0]>=0.5):\n",
        "    y_pred.append(1)\n",
        "  else:\n",
        "    y_pred.append(0)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JG9MPeMB2FrY",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 170
        },
        "outputId": "c3a7bd58-4e31-4fb1-c74b-bb3b90bc0863"
      },
      "source": [
        "from sklearn.metrics import classification_report\n",
        "print(classification_report(y_test, y_pred))"
      ],
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.75      0.76      0.76      6111\n",
            "           1       0.95      0.95      0.95     30307\n",
            "\n",
            "    accuracy                           0.92     36418\n",
            "   macro avg       0.85      0.86      0.85     36418\n",
            "weighted avg       0.92      0.92      0.92     36418\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HNvevds4gCi4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}