Abstractive_Text_Summarization_using_seq-2-seq_RNNs.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "In-House_Abstractive_Text_Summarization(17102135).ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "voSx0eDmfSu7",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 124
        },
        "outputId": "07a8062c-f08b-45ef-a212-f59bc1e3392a"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "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": "wPgirM2YfebU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "95419fcb-0b7e-4b78-a5da-b2f13784eccb"
      },
      "source": [
        "%tensorflow_version 1.x\n",
        "import numpy as np   #Package for scientific computing and dealing with arrays\n",
        "import pandas as pd  #Package providing fast, flexible and expressive data structures\n",
        "import re            #re stands for RegularExpression providing full support for Perl-like Regular Expressions in Python\n",
        "from bs4 import BeautifulSoup   #Package for pulling data out of HTML and XML files\n",
        "from keras.preprocessing.text import Tokenizer  #For tokenizing the input sequences\n",
        "from keras.preprocessing.sequence import pad_sequences  #For Padding the seqences to same length\n",
        "from nltk.corpus import stopwords   #For removing filler words\n",
        "from tensorflow.keras.layers import Input, LSTM, Attention, Embedding, Dense, Concatenate, TimeDistributed   #Layers required to implement the model\n",
        "from tensorflow.keras.models import Model  #Helps in grouping the layers into an object with training and inference features\n",
        "from tensorflow.keras.callbacks import EarlyStopping  #Allows training the model on large no. of training epochs & stop once the performance stops improving on validation dataset\n",
        "import warnings  #shows warning message that may arise \n",
        "\n",
        "pd.set_option(\"display.max_colwidth\", 200) #Setting the data sructure display length\n",
        "warnings.filterwarnings(\"ignore\") "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "TensorFlow 1.x selected.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IrQMf8BUfedA",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 815
        },
        "outputId": "75531afa-76ea-4d28-a715-be461fef5cb3"
      },
      "source": [
        "reviewsData=pd.read_csv(\"/content/drive/My Drive/Amazon Food Reviews/Reviews.csv\",nrows=30000) #Taking 30,000 out of 500,000 reviews\n",
        "print(reviewsData.shape) #Analyzing the shape of the dataset\n",
        "reviewsData.head(n=10)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(30000, 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 dog food products and have found them all to be of good quality. The product looks more like a stew than a processed meat and it smells better. My Labr...</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 Peanuts...the peanuts were actually small sized unsalted. Not sure if this was an error or if the vendor intended to represent the product as \"Jumbo\".</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 few centuries.  It is a light, pillowy citrus gelatin with nuts - in this case Filberts. And it is cut into tiny squares and then liberally coated with ...</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 in Robitussin I believe I have found it.  I got this in addition to the Root Beer Extract I ordered (which was good) and made some cherry soda.  The fl...</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 wide assortment of yummy taffy.  Delivery was very quick.  If your a taffy lover, this is a deal.</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>6</td>\n",
              "      <td>B006K2ZZ7K</td>\n",
              "      <td>ADT0SRK1MGOEU</td>\n",
              "      <td>Twoapennything</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>1342051200</td>\n",
              "      <td>Nice Taffy</td>\n",
              "      <td>I got a wild hair for taffy and ordered this five pound bag. The taffy was all very enjoyable with many flavors: watermelon, root beer, melon, peppermint, grape, etc. My only complaint is there wa...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>7</td>\n",
              "      <td>B006K2ZZ7K</td>\n",
              "      <td>A1SP2KVKFXXRU1</td>\n",
              "      <td>David C. Sullivan</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>1340150400</td>\n",
              "      <td>Great!  Just as good as the expensive brands!</td>\n",
              "      <td>This saltwater taffy had great flavors and was very soft and chewy.  Each candy was individually wrapped well.  None of the candies were stuck together, which did happen in the expensive version, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>8</td>\n",
              "      <td>B006K2ZZ7K</td>\n",
              "      <td>A3JRGQVEQN31IQ</td>\n",
              "      <td>Pamela G. Williams</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>1336003200</td>\n",
              "      <td>Wonderful, tasty taffy</td>\n",
              "      <td>This taffy is so good.  It is very soft and chewy.  The flavors are amazing.  I would definitely recommend you buying it.  Very satisfying!!</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>9</td>\n",
              "      <td>B000E7L2R4</td>\n",
              "      <td>A1MZYO9TZK0BBI</td>\n",
              "      <td>R. James</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>1322006400</td>\n",
              "      <td>Yay Barley</td>\n",
              "      <td>Right now I'm mostly just sprouting this so my cats can eat the grass. They love it. I rotate it around with Wheatgrass and Rye too</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>10</td>\n",
              "      <td>B00171APVA</td>\n",
              "      <td>A21BT40VZCCYT4</td>\n",
              "      <td>Carol A. Reed</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>1351209600</td>\n",
              "      <td>Healthy Dog Food</td>\n",
              "      <td>This is a very healthy dog food. Good for their digestion. Also good for small puppies. My dog eats her required amount at every feeding.</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Id  ...                                                                                                                                                                                                     Text\n",
              "0   1  ...  I have bought several of the Vitality canned dog food products and have found them all to be of good quality. The product looks more like a stew than a processed meat and it smells better. My Labr...\n",
              "1   2  ...           Product arrived labeled as Jumbo Salted Peanuts...the peanuts were actually small sized unsalted. Not sure if this was an error or if the vendor intended to represent the product as \"Jumbo\".\n",
              "2   3  ...  This is a confection that has been around a few centuries.  It is a light, pillowy citrus gelatin with nuts - in this case Filberts. And it is cut into tiny squares and then liberally coated with ...\n",
              "3   4  ...  If you are looking for the secret ingredient in Robitussin I believe I have found it.  I got this in addition to the Root Beer Extract I ordered (which was good) and made some cherry soda.  The fl...\n",
              "4   5  ...                                                             Great taffy at a great price.  There was a wide assortment of yummy taffy.  Delivery was very quick.  If your a taffy lover, this is a deal.\n",
              "5   6  ...  I got a wild hair for taffy and ordered this five pound bag. The taffy was all very enjoyable with many flavors: watermelon, root beer, melon, peppermint, grape, etc. My only complaint is there wa...\n",
              "6   7  ...  This saltwater taffy had great flavors and was very soft and chewy.  Each candy was individually wrapped well.  None of the candies were stuck together, which did happen in the expensive version, ...\n",
              "7   8  ...                                                             This taffy is so good.  It is very soft and chewy.  The flavors are amazing.  I would definitely recommend you buying it.  Very satisfying!!\n",
              "8   9  ...                                                                      Right now I'm mostly just sprouting this so my cats can eat the grass. They love it. I rotate it around with Wheatgrass and Rye too\n",
              "9  10  ...                                                                This is a very healthy dog food. Good for their digestion. Also good for small puppies. My dog eats her required amount at every feeding.\n",
              "\n",
              "[10 rows x 10 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pFVcCCAWqu59",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "e1de6525-ffd0-4f2f-968f-bbec96c16ac6"
      },
      "source": [
        "DATASET_COLUMNS = [\"Id\", \"ProductId\", \"UserId\", \"ProfileName\", \"HelpfulnessNumerator\", \"HelpfulnessDenominator\", \"Score\", \"Time\", \"Summary\", \"Text\"]\n",
        "reviewsData.columns = DATASET_COLUMNS\n",
        "reviewsData.head(n=10)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>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 dog food products and have found them all to be of good quality. The product looks more like a stew than a processed meat and it smells better. My Labr...</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 Peanuts...the peanuts were actually small sized unsalted. Not sure if this was an error or if the vendor intended to represent the product as \"Jumbo\".</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 few centuries.  It is a light, pillowy citrus gelatin with nuts - in this case Filberts. And it is cut into tiny squares and then liberally coated with ...</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 in Robitussin I believe I have found it.  I got this in addition to the Root Beer Extract I ordered (which was good) and made some cherry soda.  The fl...</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 wide assortment of yummy taffy.  Delivery was very quick.  If your a taffy lover, this is a deal.</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>6</td>\n",
              "      <td>B006K2ZZ7K</td>\n",
              "      <td>ADT0SRK1MGOEU</td>\n",
              "      <td>Twoapennything</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>1342051200</td>\n",
              "      <td>Nice Taffy</td>\n",
              "      <td>I got a wild hair for taffy and ordered this five pound bag. The taffy was all very enjoyable with many flavors: watermelon, root beer, melon, peppermint, grape, etc. My only complaint is there wa...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>7</td>\n",
              "      <td>B006K2ZZ7K</td>\n",
              "      <td>A1SP2KVKFXXRU1</td>\n",
              "      <td>David C. Sullivan</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>1340150400</td>\n",
              "      <td>Great!  Just as good as the expensive brands!</td>\n",
              "      <td>This saltwater taffy had great flavors and was very soft and chewy.  Each candy was individually wrapped well.  None of the candies were stuck together, which did happen in the expensive version, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>8</td>\n",
              "      <td>B006K2ZZ7K</td>\n",
              "      <td>A3JRGQVEQN31IQ</td>\n",
              "      <td>Pamela G. Williams</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>1336003200</td>\n",
              "      <td>Wonderful, tasty taffy</td>\n",
              "      <td>This taffy is so good.  It is very soft and chewy.  The flavors are amazing.  I would definitely recommend you buying it.  Very satisfying!!</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>9</td>\n",
              "      <td>B000E7L2R4</td>\n",
              "      <td>A1MZYO9TZK0BBI</td>\n",
              "      <td>R. James</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>1322006400</td>\n",
              "      <td>Yay Barley</td>\n",
              "      <td>Right now I'm mostly just sprouting this so my cats can eat the grass. They love it. I rotate it around with Wheatgrass and Rye too</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>10</td>\n",
              "      <td>B00171APVA</td>\n",
              "      <td>A21BT40VZCCYT4</td>\n",
              "      <td>Carol A. Reed</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>1351209600</td>\n",
              "      <td>Healthy Dog Food</td>\n",
              "      <td>This is a very healthy dog food. Good for their digestion. Also good for small puppies. My dog eats her required amount at every feeding.</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Id  ...                                                                                                                                                                                                     Text\n",
              "0   1  ...  I have bought several of the Vitality canned dog food products and have found them all to be of good quality. The product looks more like a stew than a processed meat and it smells better. My Labr...\n",
              "1   2  ...           Product arrived labeled as Jumbo Salted Peanuts...the peanuts were actually small sized unsalted. Not sure if this was an error or if the vendor intended to represent the product as \"Jumbo\".\n",
              "2   3  ...  This is a confection that has been around a few centuries.  It is a light, pillowy citrus gelatin with nuts - in this case Filberts. And it is cut into tiny squares and then liberally coated with ...\n",
              "3   4  ...  If you are looking for the secret ingredient in Robitussin I believe I have found it.  I got this in addition to the Root Beer Extract I ordered (which was good) and made some cherry soda.  The fl...\n",
              "4   5  ...                                                             Great taffy at a great price.  There was a wide assortment of yummy taffy.  Delivery was very quick.  If your a taffy lover, this is a deal.\n",
              "5   6  ...  I got a wild hair for taffy and ordered this five pound bag. The taffy was all very enjoyable with many flavors: watermelon, root beer, melon, peppermint, grape, etc. My only complaint is there wa...\n",
              "6   7  ...  This saltwater taffy had great flavors and was very soft and chewy.  Each candy was individually wrapped well.  None of the candies were stuck together, which did happen in the expensive version, ...\n",
              "7   8  ...                                                             This taffy is so good.  It is very soft and chewy.  The flavors are amazing.  I would definitely recommend you buying it.  Very satisfying!!\n",
              "8   9  ...                                                                      Right now I'm mostly just sprouting this so my cats can eat the grass. They love it. I rotate it around with Wheatgrass and Rye too\n",
              "9  10  ...                                                                This is a very healthy dog food. Good for their digestion. Also good for small puppies. My dog eats her required amount at every feeding.\n",
              "\n",
              "[10 rows x 10 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zj1U-pTOo5-Z",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 461
        },
        "outputId": "c5aa2244-6915-415b-eaef-662bfec83f8e"
      },
      "source": [
        "reviewsData.drop(['Id', 'ProductId', 'UserId', 'ProfileName' , 'HelpfulnessNumerator' , 'HelpfulnessDenominator', 'Score' ,'Time'],axis = 1 ,inplace = True)\n",
        "reviewsData.head(n=10)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Summary</th>\n",
              "      <th>Text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Good Quality Dog Food</td>\n",
              "      <td>I have bought several of the Vitality canned dog food products and have found them all to be of good quality. The product looks more like a stew than a processed meat and it smells better. My Labr...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Not as Advertised</td>\n",
              "      <td>Product arrived labeled as Jumbo Salted Peanuts...the peanuts were actually small sized unsalted. Not sure if this was an error or if the vendor intended to represent the product as \"Jumbo\".</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>\"Delight\" says it all</td>\n",
              "      <td>This is a confection that has been around a few centuries.  It is a light, pillowy citrus gelatin with nuts - in this case Filberts. And it is cut into tiny squares and then liberally coated with ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Cough Medicine</td>\n",
              "      <td>If you are looking for the secret ingredient in Robitussin I believe I have found it.  I got this in addition to the Root Beer Extract I ordered (which was good) and made some cherry soda.  The fl...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Great taffy</td>\n",
              "      <td>Great taffy at a great price.  There was a wide assortment of yummy taffy.  Delivery was very quick.  If your a taffy lover, this is a deal.</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Nice Taffy</td>\n",
              "      <td>I got a wild hair for taffy and ordered this five pound bag. The taffy was all very enjoyable with many flavors: watermelon, root beer, melon, peppermint, grape, etc. My only complaint is there wa...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Great!  Just as good as the expensive brands!</td>\n",
              "      <td>This saltwater taffy had great flavors and was very soft and chewy.  Each candy was individually wrapped well.  None of the candies were stuck together, which did happen in the expensive version, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>Wonderful, tasty taffy</td>\n",
              "      <td>This taffy is so good.  It is very soft and chewy.  The flavors are amazing.  I would definitely recommend you buying it.  Very satisfying!!</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Yay Barley</td>\n",
              "      <td>Right now I'm mostly just sprouting this so my cats can eat the grass. They love it. I rotate it around with Wheatgrass and Rye too</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>Healthy Dog Food</td>\n",
              "      <td>This is a very healthy dog food. Good for their digestion. Also good for small puppies. My dog eats her required amount at every feeding.</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                         Summary                                                                                                                                                                                                     Text\n",
              "0                          Good Quality Dog Food  I have bought several of the Vitality canned dog food products and have found them all to be of good quality. The product looks more like a stew than a processed meat and it smells better. My Labr...\n",
              "1                              Not as Advertised           Product arrived labeled as Jumbo Salted Peanuts...the peanuts were actually small sized unsalted. Not sure if this was an error or if the vendor intended to represent the product as \"Jumbo\".\n",
              "2                          \"Delight\" says it all  This is a confection that has been around a few centuries.  It is a light, pillowy citrus gelatin with nuts - in this case Filberts. And it is cut into tiny squares and then liberally coated with ...\n",
              "3                                 Cough Medicine  If you are looking for the secret ingredient in Robitussin I believe I have found it.  I got this in addition to the Root Beer Extract I ordered (which was good) and made some cherry soda.  The fl...\n",
              "4                                    Great taffy                                                             Great taffy at a great price.  There was a wide assortment of yummy taffy.  Delivery was very quick.  If your a taffy lover, this is a deal.\n",
              "5                                     Nice Taffy  I got a wild hair for taffy and ordered this five pound bag. The taffy was all very enjoyable with many flavors: watermelon, root beer, melon, peppermint, grape, etc. My only complaint is there wa...\n",
              "6  Great!  Just as good as the expensive brands!  This saltwater taffy had great flavors and was very soft and chewy.  Each candy was individually wrapped well.  None of the candies were stuck together, which did happen in the expensive version, ...\n",
              "7                         Wonderful, tasty taffy                                                             This taffy is so good.  It is very soft and chewy.  The flavors are amazing.  I would definitely recommend you buying it.  Very satisfying!!\n",
              "8                                     Yay Barley                                                                      Right now I'm mostly just sprouting this so my cats can eat the grass. They love it. I rotate it around with Wheatgrass and Rye too\n",
              "9                               Healthy Dog Food                                                                This is a very healthy dog food. Good for their digestion. Also good for small puppies. My dog eats her required amount at every feeding."
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FtdtqjNLfeh5",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Reducing the length of dataset for better training and performance\n",
        "reviewsData.drop_duplicates(subset=['Text'],inplace=True) #Dropping the rows with Duplicates values of 'Text'  \n",
        "reviewsData.dropna(axis=0,inplace=True) #Dropping the rows with Missing values "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cgW9qaJcCW_b",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 170
        },
        "outputId": "b5703189-2cf9-4f5d-bafc-7266222eed14"
      },
      "source": [
        "reviewsData.info() #Getting more info on datatypes and shape of Dataset"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 28677 entries, 0 to 29999\n",
            "Data columns (total 2 columns):\n",
            " #   Column   Non-Null Count  Dtype \n",
            "---  ------   --------------  ----- \n",
            " 0   Summary  28677 non-null  object\n",
            " 1   Text     28677 non-null  object\n",
            "dtypes: object(2)\n",
            "memory usage: 672.1+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b3bc3e1Dfeld",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Preprocessing\n",
        "\n",
        "#This the dictionary used for expanding contractions\n",
        "contraction_mapping = {\"ain't\": \"is not\", \"aren't\": \"are not\",\"can't\": \"cannot\", \"'cause\": \"because\", \"could've\": \"could have\", \"couldn't\": \"could not\",\n",
        "\n",
        "                           \"didn't\": \"did not\", \"doesn't\": \"does not\", \"don't\": \"do not\", \"hadn't\": \"had not\", \"hasn't\": \"has not\", \"haven't\": \"have not\",\n",
        "\n",
        "                           \"he'd\": \"he would\",\"he'll\": \"he will\", \"he's\": \"he is\", \"how'd\": \"how did\", \"how'd'y\": \"how do you\", \"how'll\": \"how will\", \"how's\": \"how is\",\n",
        "\n",
        "                           \"I'd\": \"I would\", \"I'd've\": \"I would have\", \"I'll\": \"I will\", \"I'll've\": \"I will have\",\"I'm\": \"I am\", \"I've\": \"I have\", \"i'd\": \"i would\",\n",
        "\n",
        "                           \"i'd've\": \"i would have\", \"i'll\": \"i will\",  \"i'll've\": \"i will have\",\"i'm\": \"i am\", \"i've\": \"i have\", \"isn't\": \"is not\", \"it'd\": \"it would\",\n",
        "\n",
        "                           \"it'd've\": \"it would have\", \"it'll\": \"it will\", \"it'll've\": \"it will have\",\"it's\": \"it is\", \"let's\": \"let us\", \"ma'am\": \"madam\",\n",
        "\n",
        "                           \"mayn't\": \"may not\", \"might've\": \"might have\",\"mightn't\": \"might not\",\"mightn't've\": \"might not have\", \"must've\": \"must have\",\n",
        "\n",
        "                           \"mustn't\": \"must not\", \"mustn't've\": \"must not have\", \"needn't\": \"need not\", \"needn't've\": \"need not have\",\"o'clock\": \"of the clock\",\n",
        "\n",
        "                           \"oughtn't\": \"ought not\", \"oughtn't've\": \"ought not have\", \"shan't\": \"shall not\", \"sha'n't\": \"shall not\", \"shan't've\": \"shall not have\",\n",
        "\n",
        "                           \"she'd\": \"she would\", \"she'd've\": \"she would have\", \"she'll\": \"she will\", \"she'll've\": \"she will have\", \"she's\": \"she is\",\n",
        "\n",
        "                           \"should've\": \"should have\", \"shouldn't\": \"should not\", \"shouldn't've\": \"should not have\", \"so've\": \"so have\",\"so's\": \"so as\",\n",
        "\n",
        "                           \"this's\": \"this is\",\"that'd\": \"that would\", \"that'd've\": \"that would have\", \"that's\": \"that is\", \"there'd\": \"there would\",\n",
        "\n",
        "                           \"there'd've\": \"there would have\", \"there's\": \"there is\", \"here's\": \"here is\",\"they'd\": \"they would\", \"they'd've\": \"they would have\",\n",
        "\n",
        "                           \"they'll\": \"they will\", \"they'll've\": \"they will have\", \"they're\": \"they are\", \"they've\": \"they have\", \"to've\": \"to have\",\n",
        "\n",
        "                           \"wasn't\": \"was not\", \"we'd\": \"we would\", \"we'd've\": \"we would have\", \"we'll\": \"we will\", \"we'll've\": \"we will have\", \"we're\": \"we are\",\n",
        "\n",
        "                           \"we've\": \"we have\", \"weren't\": \"were not\", \"what'll\": \"what will\", \"what'll've\": \"what will have\", \"what're\": \"what are\",\n",
        "\n",
        "                           \"what's\": \"what is\", \"what've\": \"what have\", \"when's\": \"when is\", \"when've\": \"when have\", \"where'd\": \"where did\", \"where's\": \"where is\",\n",
        "\n",
        "                           \"where've\": \"where have\", \"who'll\": \"who will\", \"who'll've\": \"who will have\", \"who's\": \"who is\", \"who've\": \"who have\",\n",
        "\n",
        "                           \"why's\": \"why is\", \"why've\": \"why have\", \"will've\": \"will have\", \"won't\": \"will not\", \"won't've\": \"will not have\",\n",
        "\n",
        "                           \"would've\": \"would have\", \"wouldn't\": \"would not\", \"wouldn't've\": \"would not have\", \"y'all\": \"you all\",\n",
        "\n",
        "                           \"y'all'd\": \"you all would\",\"y'all'd've\": \"you all would have\",\"y'all're\": \"you all are\",\"y'all've\": \"you all have\",\n",
        "\n",
        "                           \"you'd\": \"you would\", \"you'd've\": \"you would have\", \"you'll\": \"you will\", \"you'll've\": \"you will have\",\n",
        "\n",
        "                           \"you're\": \"you are\", \"you've\": \"you have\"}"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rDsJkVHbfes2",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "b8c89634-550e-47b3-913d-f022e03ec3c4"
      },
      "source": [
        "#Text Cleaning\n",
        "import nltk\n",
        "nltk.download('stopwords')\n",
        "\n",
        "stop_words = set(stopwords.words('english')) \n",
        "def text_cleaner(text,num):\n",
        "    newString = text.lower()  #converts all uppercase characters in the string into lowercase characters and returns it\n",
        "    newString = BeautifulSoup(newString, \"lxml\").text #parses the string into an lxml.html \n",
        "    newString = re.sub(r'\\([^)]*\\)', '', newString) #used to replace a string that matches a regular expression instead of perfect match\n",
        "    newString = re.sub('\"','', newString)           \n",
        "    newString = ' '.join([contraction_mapping[t] if t in contraction_mapping else t for t in newString.split(\" \")]) #for expanding contractions using the contraction_mapping dictionary    \n",
        "    newString = re.sub(r\"'s\\b\",\"\",newString)\n",
        "    newString = re.sub(\"[^a-zA-Z]\", \" \", newString)\n",
        "    if(num==0): \n",
        "      tokens = [w for w in newString.split() if not w in stop_words]  #converting the strings into tokens\n",
        "    else :\n",
        "      tokens = newString.split()\n",
        "    long_words=[]\n",
        "    for i in tokens:\n",
        "        if len(i)>1:                  #removing short words\n",
        "            long_words.append(i)   \n",
        "    return (\" \".join(long_words)).strip()\n",
        "\n",
        "#Calling the function\n",
        "cleaned_text = []\n",
        "for t in reviewsData['Text']:\n",
        "    cleaned_text.append(text_cleaner(t,0))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/stopwords.zip.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_iPKSs0Tfe2x",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "outputId": "0ef479b4-41d7-4c69-ec80-74aeeddff65d"
      },
      "source": [
        "reviewsData['Text'][:10] #Looking at the 'Text' column of the dataset"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    I have bought several of the Vitality canned dog food products and have found them all to be of good quality. The product looks more like a stew than a processed meat and it smells better. My Labr...\n",
              "1             Product arrived labeled as Jumbo Salted Peanuts...the peanuts were actually small sized unsalted. Not sure if this was an error or if the vendor intended to represent the product as \"Jumbo\".\n",
              "2    This is a confection that has been around a few centuries.  It is a light, pillowy citrus gelatin with nuts - in this case Filberts. And it is cut into tiny squares and then liberally coated with ...\n",
              "3    If you are looking for the secret ingredient in Robitussin I believe I have found it.  I got this in addition to the Root Beer Extract I ordered (which was good) and made some cherry soda.  The fl...\n",
              "4                                                               Great taffy at a great price.  There was a wide assortment of yummy taffy.  Delivery was very quick.  If your a taffy lover, this is a deal.\n",
              "5    I got a wild hair for taffy and ordered this five pound bag. The taffy was all very enjoyable with many flavors: watermelon, root beer, melon, peppermint, grape, etc. My only complaint is there wa...\n",
              "6    This saltwater taffy had great flavors and was very soft and chewy.  Each candy was individually wrapped well.  None of the candies were stuck together, which did happen in the expensive version, ...\n",
              "7                                                               This taffy is so good.  It is very soft and chewy.  The flavors are amazing.  I would definitely recommend you buying it.  Very satisfying!!\n",
              "8                                                                        Right now I'm mostly just sprouting this so my cats can eat the grass. They love it. I rotate it around with Wheatgrass and Rye too\n",
              "9                                                                  This is a very healthy dog food. Good for their digestion. Also good for small puppies. My dog eats her required amount at every feeding.\n",
              "Name: Text, dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XiHDDv9_Izul",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 207
        },
        "outputId": "79e7ac4d-64cc-463b-ed16-8f1c72e0fcb5"
      },
      "source": [
        "cleaned_text[:10] #Looking at the Text after removing stop words, special characters , punctuations etc."
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['bought several vitality canned dog food products found good quality product looks like stew processed meat smells better labrador finicky appreciates product better',\n",
              " 'product arrived labeled jumbo salted peanuts peanuts actually small sized unsalted sure error vendor intended represent product jumbo',\n",
              " 'confection around centuries light pillowy citrus gelatin nuts case filberts cut tiny squares liberally coated powdered sugar tiny mouthful heaven chewy flavorful highly recommend yummy treat familiar story lewis lion witch wardrobe treat seduces edmund selling brother sisters witch',\n",
              " 'looking secret ingredient robitussin believe found got addition root beer extract ordered made cherry soda flavor medicinal',\n",
              " 'great taffy great price wide assortment yummy taffy delivery quick taffy lover deal',\n",
              " 'got wild hair taffy ordered five pound bag taffy enjoyable many flavors watermelon root beer melon peppermint grape etc complaint bit much red black licorice flavored pieces kids husband lasted two weeks would recommend brand taffy delightful treat',\n",
              " 'saltwater taffy great flavors soft chewy candy individually wrapped well none candies stuck together happen expensive version fralinger would highly recommend candy served beach themed party everyone loved',\n",
              " 'taffy good soft chewy flavors amazing would definitely recommend buying satisfying',\n",
              " 'right mostly sprouting cats eat grass love rotate around wheatgrass rye',\n",
              " 'healthy dog food good digestion also good small puppies dog eats required amount every feeding']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Vw1UFS1bfe8F",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Summary Cleaning \n",
        "cleaned_summary = []    #Using the text_cleaner function for cleaning summary too\n",
        "for t in reviewsData['Summary']:\n",
        "    cleaned_summary.append(text_cleaner(t,1))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yBOuryAKffAD",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "02da7962-9ff0-472c-a2ee-246969a45448"
      },
      "source": [
        "reviewsData['Summary'][:10]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0                            Good Quality Dog Food\n",
              "1                                Not as Advertised\n",
              "2                            \"Delight\" says it all\n",
              "3                                   Cough Medicine\n",
              "4                                      Great taffy\n",
              "5                                       Nice Taffy\n",
              "6    Great!  Just as good as the expensive brands!\n",
              "7                           Wonderful, tasty taffy\n",
              "8                                       Yay Barley\n",
              "9                                 Healthy Dog Food\n",
              "Name: Summary, dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z0SduOc8ffDO",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 187
        },
        "outputId": "01efc60d-7a05-4858-b1ef-44df1251396c"
      },
      "source": [
        "cleaned_summary[:10]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['good quality dog food',\n",
              " 'not as advertised',\n",
              " 'delight says it all',\n",
              " 'cough medicine',\n",
              " 'great taffy',\n",
              " 'nice taffy',\n",
              " 'great just as good as the expensive brands',\n",
              " 'wonderful tasty taffy',\n",
              " 'yay barley',\n",
              " 'healthy dog food']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A0Lo6geFfe6r",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "reviewsData['Cleaned_Text'] = cleaned_text  #Adding cleaned text to the dataset\n",
        "reviewsData['Cleaned_Summary'] = cleaned_summary  #Adding cleaned summary to the dataset\n",
        "#Dropping Empty Rows\n",
        "reviewsData['Cleaned_Summary'].replace('', np.nan, inplace=True)\n",
        "#Dropping rows with Missing values\n",
        "reviewsData.dropna(axis=0,inplace=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rUZO3csTvWxc",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 411
        },
        "outputId": "c0634676-7143-4924-8605-3aea7df6f8bf"
      },
      "source": [
        "#Before Cleaning\n",
        "print(\"Before Preprocessing:\\n\")\n",
        "for i in range(5):\n",
        "    print(\"Review:\",reviewsData['Text'][i])\n",
        "    print(\"Summary:\",reviewsData['Summary'][i])\n",
        "    print(\"\\n\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Before Preprocessing:\n",
            "\n",
            "Review: I have bought several of the Vitality canned dog food products and have found them all to be of good quality. The product looks more like a stew than a processed meat and it smells better. My Labrador is finicky and she appreciates this product better than  most.\n",
            "Summary: Good Quality Dog Food\n",
            "\n",
            "\n",
            "Review: Product arrived labeled as Jumbo Salted Peanuts...the peanuts were actually small sized unsalted. Not sure if this was an error or if the vendor intended to represent the product as \"Jumbo\".\n",
            "Summary: Not as Advertised\n",
            "\n",
            "\n",
            "Review: This is a confection that has been around a few centuries.  It is a light, pillowy citrus gelatin with nuts - in this case Filberts. And it is cut into tiny squares and then liberally coated with powdered sugar.  And it is a tiny mouthful of heaven.  Not too chewy, and very flavorful.  I highly recommend this yummy treat.  If you are familiar with the story of C.S. Lewis' \"The Lion, The Witch, and The Wardrobe\" - this is the treat that seduces Edmund into selling out his Brother and Sisters to the Witch.\n",
            "Summary: \"Delight\" says it all\n",
            "\n",
            "\n",
            "Review: If you are looking for the secret ingredient in Robitussin I believe I have found it.  I got this in addition to the Root Beer Extract I ordered (which was good) and made some cherry soda.  The flavor is very medicinal.\n",
            "Summary: Cough Medicine\n",
            "\n",
            "\n",
            "Review: Great taffy at a great price.  There was a wide assortment of yummy taffy.  Delivery was very quick.  If your a taffy lover, this is a deal.\n",
            "Summary: Great taffy\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qiZxHbedfe1p",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 411
        },
        "outputId": "1e8b7dd3-a5fd-425c-bd2b-811c14628b31"
      },
      "source": [
        "#Printing the Cleaned text and summary which will work as input to the model \n",
        "print(\"After Preprocessing:\\n\")\n",
        "for i in range(5):\n",
        "    print(\"Review:\",reviewsData['Cleaned_Text'][i])\n",
        "    print(\"Summary:\",reviewsData['Cleaned_Summary'][i])\n",
        "    print(\"\\n\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "After Preprocessing:\n",
            "\n",
            "Review: bought several vitality canned dog food products found good quality product looks like stew processed meat smells better labrador finicky appreciates product better\n",
            "Summary: good quality dog food\n",
            "\n",
            "\n",
            "Review: product arrived labeled jumbo salted peanuts peanuts actually small sized unsalted sure error vendor intended represent product jumbo\n",
            "Summary: not as advertised\n",
            "\n",
            "\n",
            "Review: confection around centuries light pillowy citrus gelatin nuts case filberts cut tiny squares liberally coated powdered sugar tiny mouthful heaven chewy flavorful highly recommend yummy treat familiar story lewis lion witch wardrobe treat seduces edmund selling brother sisters witch\n",
            "Summary: delight says it all\n",
            "\n",
            "\n",
            "Review: looking secret ingredient robitussin believe found got addition root beer extract ordered made cherry soda flavor medicinal\n",
            "Summary: cough medicine\n",
            "\n",
            "\n",
            "Review: great taffy great price wide assortment yummy taffy delivery quick taffy lover deal\n",
            "Summary: great taffy\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wzJRTxrSfewf",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "outputId": "df017e84-d65f-4449-8b6c-881884fb00e0"
      },
      "source": [
        "#Data Visualization\n",
        "import matplotlib.pyplot as plt\n",
        "text_word_count = []\n",
        "summary_word_count = []\n",
        "\n",
        "#Populating the lists with sentence lengths\n",
        "for i in reviewsData['Cleaned_Text']:\n",
        "      text_word_count.append(len(i.split()))\n",
        "\n",
        "for i in reviewsData['Cleaned_Summary']:\n",
        "      summary_word_count.append(len(i.split()))\n",
        "\n",
        "length_df = pd.DataFrame({'text':text_word_count, 'summary':summary_word_count})\n",
        "length_df.hist(bins = 30)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEICAYAAACzliQjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nO3df5QdZZ3n8ffHRDASJAnBFhPGjpLBE2GQkDFxdZxINITgGvYcZEFGApsdzu4ER3dQCeqeOAIzYXYiAoPMZiQSmEjIIJisopAB+rBzdhIg/AoQkRaCSU4gQkIwIGjwu3/U06Ry+97u20n3requz+uce7rqqafqPs/tut9b97lPPY8iAjMzq4a3FF0AMzNrHQd9M7MKcdA3M6sQB30zswpx0DczqxAHfTOzCnHQNzOrEAd9MysFSZskfaIsxxmqHPRtH5KGF10GMxs4DvoDQNJFkrZK+rWkJyXNkHS9pEtzeaZL2pJb3yTpy5IelfSKpOsktUn6STrOv0oanfK2SwpJ50naLGmnpP8m6Y/T/i9J+ofcsd8n6W5JL0p6QdJySaNqnvsiSY8Cr6Ry/KCmTldJunJAXzirLEk3An8A/B9JuyV9RdI0Sf8vnc+PSJqe8v6HdB4fldaPT++B99c7TmGVKquI8KMfH8AxwGbg3Wm9HXgfcD1waS7fdGBLbn0TsBZoA8YB24EHgROAtwF3AwtzxwzgH9O2mcBrwA+Bd+b2/9OU/2jgk8DBwBHAvcC3a577YeAoYARwJPAKMCptH56Od2LRr68fQ/eRzsNPpOVxwIvAbLKL00+m9SPS9svSe2IEsAG4oN5x/Oj+8JV+/3uDLLhOkvTWiNgUEb9oct+rI+L5iNgK/F9gXUQ8FBGvAbeRfQDkXRIRr0XEnWRB+qaI2J7b/wSAiOiMiDUR8XpE/Ar4FvCnNce6KiI2R8RvImIb2QfDZ9K2WcALEbG+T6+E2f77M+D2iLg9In4fEWuAB8g+BAC+ARwG3AdsBa4ppJSDkIN+P4uITuCLZCfldkkrJL27yd2fzy3/ps76yP3Jn5qJVqQmp5eBfwbG1hxrc836MrI3HunvjU3Wwaw/vAf4TGraeUnSS8BHyb6FEhG/I/v2fCywONIlvvXOQX8ARMT3I+KjZCduAJeTXYm/PZftXS0s0t+kchwXEe8gC+KqyVP7pvkh8EeSjgU+BSwf8FJa1eXPwc3AjRExKvc4JCIWAUgaBywEvgcslnRwg+NYDQf9fibpGEknpZPwNbIr7t+TtZnPljRG0rvIvg20yqHAbmBXerN8ubcdUpPSLcD3gfsi4pcDW0Qzngfem5b/GfiPkk6WNEzS21Lnh/GSRHaVfx0wD9gGXNLgOFbDQb//HQwsAl4AniP7YfVisuaRR8h+ZLoTuLmFZfprYDKwC/gxcGuT+y0DjsNNO9Yafwt8PTXl/GdgDvBV4FdkV/5fJotZf0n2vvqfqVnnPOA8SX9SexxJX2pxHUpPbgqzRiT9AfAz4F0R8XLR5TGzA+crfatL0luAvwJWOOCbDR2++9K6kXQIWbvos2TdNc1siHDzjplZhbh5x8ysQkrdvDN27Ng44ogjOOSQQ4ouyoB55ZVXhmz9ylK39evXvxARRxRdjmaNHTs22tvbu6WX5fWsVdZyQXnLNtDl6vGcL3ociJ4eJ554Ytxzzz0xlA3l+pWlbsADUYLzudnHiSeeWLceZXk9a5W1XBHlLdtAl6unc97NO2ZmFeKgb2ZWIQ76ZmYV4qBvZlYhDvpmZhXioG9mViEO+mZmFeKgb2ZWIQ76ZmYVUuphGPqifcGPu6VtWnRqASUxGxgbtu7i3Jrz3Oe49ZWv9M3MKsRB38ysQhz0zcwqxEHfzKxCHPTNzCqkqaAvaZSkWyT9TNJGSR+WNEbSGklPpb+jU15JukpSp6RHJU3OHWduyv+UpLkDVSkzM6uv2Sv9K4GfRsT7geOBjcAC4K6ImAjcldYBTgEmpsf5wLUAksYAC4GpwIeAhV0fFGZFkLRU0nZJj9Wkfz5d4Dwu6e9y6Reni5knJZ2cS5+V0jolLcilT5C0LqXfLOmg1tTMrLFeg76kw4CPAdcBRMRvI+IlYA6wLGVbBpyWlucAN6QJXNYCoyQdCZwMrImIHRGxE1gDzOrX2pj1zfXUnIOSPk52Dh8fER8A/j6lTwLOBD6Q9vmOpGGShgHXkF3sTALOSnkBLgeuiIijgZ3AvAGvkVkvmrk5awLwK+B7ko4H1gNfANoiYlvK8xzQlpbHAZtz+29JaY3S9yHpfLJvCLS1tbF79246Ojp6LeSFx+3pltbMfkVrtn6DUdnrFhH3SmqvSf7vwKKIeD3l2Z7S5wArUvozkjrJvrECdEbE0wCSVgBzJG0ETgI+m/IsA75B+uZrVpRmgv5wYDLw+YhYJ+lK9jblABARISn6o0ARsQRYAjBlypQYOXIk06dP73W/2jsVATad3ft+Revo6GiqfoPRIK3bHwJ/Iuky4DXgSxFxP9kFytpcvvxFS+3FzFTgcOCliNhTJ383tRc79T4s20Z0v7gpw4dqmT/cy1q2IsvVTNDfAmyJiHVp/RayoP+8pCMjYltqvum6ItoKHJXbf3xK2wpMr0nv2P+imw2I4cAYYBrwx8BKSe8d6Cetvdip92F59fJVLN6w71u2DBc2Zf5wL2vZiixXr236EfEcsFnSMSlpBvAEsBro6oEzF1iVllcD56RePNOAXakZ6A5gpqTR6QfcmSnNrEy2ALem36TuA34PjKXni5l66S+S/Z41vCbdrFDNDrj2eWB56n3wNHAe2QfGSknzgGeBM1Le24HZQCfwaspLROyQdAlwf8r3zYjY0S+1MOs/PwQ+Dtwj6Q+Bg4AXyC5mvi/pW8C7yXqn3QcImChpAllQPxP4bGryvAc4HVjBvhdGZoVpKuhHxMPAlDqbZtTJG8D8BsdZCiztSwHNBoqkm8iaHMdK2kLWpXgpsDR14/wtMDed049LWkn2LXcPMD8i3kjHuYDsW+swYGlEPJ6e4iJghaRLgYdIPeDMijRkhlY266uIOKvBpj9rkP8y4LI66beTfcOtTX+avT18zErBwzCYmVWIg76ZWYU46JuZVYiDvplZhTjom5lViIO+mVmFOOibmVWIg76ZWYU46JuZVYiDvplZhTjom5lViIO+mVmFDOkB19prZtPatOjUgkpiZlYOvtI3M6sQB30zswpx0DczqxAHfassSUslbU+zZNVuu1BSSBqb1iXpKkmdkh6VNDmXd66kp9Jjbi79REkb0j5XSVJrambWmIO+Vdn1wKzaRElHATOBX+aSTyGbF3cicD5wbco7hmyaxalks2QtlDQ67XMt8Oe5/bo9l1mrOehbZUXEvcCOOpuuAL4CRC5tDnBDZNYCoyQdCZwMrImIHRGxE1gDzErb3hERa9McuzcApw1kfcyaMaS7bJr1laQ5wNaIeKSmNWYcsDm3viWl9ZS+pU56o+c9n+wbBG1tbXR0dHTL0zYCLjxuzz5p9fK12u7du0tRjnrKWrYiy+Wgb5ZIejvwVbKmnZaKiCXAEoApU6bE9OnTu+W5evkqFm/Y9y276ezu+Vqto6ODeuUtg7KWrchyuXnHbK/3AROARyRtAsYDD0p6F7AVOCqXd3xK6yl9fJ10s0I56JslEbEhIt4ZEe0R0U7WJDM5Ip4DVgPnpF4804BdEbENuAOYKWl0+gF3JnBH2vaypGmp1845wKpCKmaW01TQl7QpdT17WNIDKW2MpDWpm9qarh4L+9O1zawIkm4C/h04RtIWSfN6yH478DTQCfwT8BcAEbEDuAS4Pz2+mdJIeb6b9vkF8JOBqIdZX/SlTf/jEfFCbn0BcFdELJK0IK1fxL5d26aSdVubmuvaNoWsV8R6SatTjwezlouIs3rZ3p5bDmB+g3xLgaV10h8Ajj2wUpr1rwNp3pkDLEvLy9jbHa1PXdsO4PnNzKyPmr3SD+BOSQH879TToC21WwI8B7Sl5b52bdtHbde1Zrs21XZlq8ddt1prKNfNbLBqNuh/NCK2SnonsEbSz/IbIyLSB8IBq+26NnLkyKa6Np1bM4xyPWXo3larrF3K+sNQrpvZYNVU805EbE1/twO3kd1u/nxqtiH93Z6y97Vrm5mZtUivQV/SIZIO7Vom65L2GFkXtq4eOHPZ2x2tT13b+rU2ZmbWo2aad9qA29It6cOB70fETyXdD6xM3dyeBc5I+W8HZpN1U3sVOA+yrm2Surq2wb5d28zMrAV6DfoR8TRwfJ30F4EZddL73LXNzMxaw3fkmplViIO+mVmFOOibmVWIg76ZWYU46JuZVYiDvplZhTjom5lViIO+mVmFOOhbZUlaKmm7pMdyaf9L0s/SBEC3SRqV23ZxmhzoSUkn59JnpbTONLdEV/oESetS+s2SDmpd7czqc9C3Krue7nM6rAGOjYg/An4OXAwgaRJwJvCBtM93JA2TNAy4hmzyoEnAWSkvwOXAFRFxNLAT6GlmLrOW6MvMWYNee53hlzctOrWAklgZRMS9ktpr0u7Mra4FTk/Lc4AVEfE68IykTrLRZgE603AlSFoBzJG0ETgJ+GzKswz4BtlMcmaFqVTQN+uj/wLcnJbHkX0IdMlPAlQ7OdBU4HDgpYjYUyd/N7WTB9WbfKZtRPfJgsowSU2ZJ8spa9mKLJeDvlkdkr4G7AGWt+L5aicPqjf5zNXLV7F4w75v2TJMDFTmyXLKWrYiy+Wgb1ZD0rnAp4AZadRY6HkSoHrpL5LNDz08Xe170iArBf+Qa5YjaRbwFeDTEfFqbtNq4ExJB0uaAEwE7iObH2Ji6qlzENmPvavTh8U97P1NID/RkFlhHPStsiTdBPw7cIykLWlCoH8ADiWbC/phSf8IEBGPAyuBJ4CfAvMj4o10FX8B2SxwG4GVKS/ARcBfpR99Dweua2H1zOpy845VVkScVSe5YWCOiMuAy+qk3042Y1xt+tPs7eFjVgq+0jczqxAHfTOzCnHQNzOrEAd9M7MKcdA3M6sQB30zswpx0Dczq5Cmg34aRvYhST9K63XHCk93LN6c0tflRzFsNB65mZm1Rl+u9L9Adsdhl0Zjhc8Ddqb0K1K+huORH1jxzcysL5oK+pLGA6cC303rIhsr/JaUZRlwWlqek9ZJ22ek/G+ORx4RzwD58cjNzKwFmh2G4dtkg1AdmtZ7Git8HGl88YjYI2lXyt/TeORvqh1XvNlxp2vHGW9W0WNtl3W87/4wlOtmNlj1GvQlfQrYHhHrJU0f6ALVjis+cuTIpsadPrfOrFjNKHo88rKO990fhnLdzAarZq70PwJ8WtJs4G3AO4AraTxWeNe441skDQcOIxtbvKfxyM3MrAV6bdOPiIsjYnxEtJP9EHt3RJxN47HCV6d10va709jijcYjNzOzFjmQoZUvAlZIuhR4iL1D0l4H3JjGEN9B9kFBRDwuqWs88j2k8cgP4PnNzKyP+hT0I6ID6EjLdccKj4jXgM802L/ueORmZtYaviPXKkvSUknbJT2WSxsjaY2kp9Lf0Sldkq5KNxc+Kmlybp+5Kf9Tkubm0k+UtCHtc1XqumxWKAd9q7LryW4UzFsA3BURE4G70jrAKWS/Q00k61J8LWQfEsBCYCrZN9+FXR8UKc+f5/arfS6zlnPQt8qKiHvJfnfKy99cWHvT4Q2RWUvWe+1I4GRgTUTsiIidwBpgVtr2johYmzoy3JA7lllhPEeu2b7aImJbWn4OaEvLb950mHTdXNhT+pY66XXV3pRY76a2thHdb0Isw81vZb4Jr6xlK7JcDvpmDURESIoWPdc+NyXWu6nt6uWrWLxh37ds0TcXQrlvwitr2Yosl5t3zPb1fGqaIf3dntIb3VzYU/r4OulmhXLQN9tX/ubC2psOz0m9eKYBu1Iz0B3ATEmj0w+4M4E70raXJU1LvXbOyR3LrDBu3rHKknQTMB0YK2kLWS+cRcBKSfOAZ4EzUvbbgdlko8O+CpwHEBE7JF0C3J/yfTMiun4c/guyHkIjgJ+kh1mhHPStsiLirAabZtTJG8D8BsdZCiytk/4AcOyBlNGsv7l5x8ysQhz0zcwqxEHfzKxCHPTNzCrEQd/MrEIc9M3MKsRB38ysQtxPv0Z7zQTrmxadWlBJzMz6n6/0zcwqxEHfzKxCHPTNzCrEQd/MrEIc9M3MKsRB38ysQhz0zcwqpNegL+ltku6T9IikxyX9dUqfIGmdpE5JN0s6KKUfnNY70/b23LEuTulPSjp5oCpldqAk/Y90vj8m6ab0PvA5b4NeM1f6rwMnRcTxwAeBWWm6uMuBKyLiaGAnMC/lnwfsTOlXpHxImgScCXwAmAV8R9Kw/qyMWX+QNA74S2BKRBwLDCM7d33O26DXa9CPzO60+tb0COAk4JaUvgw4LS3PSeuk7TPSHKFzgBUR8XpEPEM27dyH+qUWZv1vODBC0nDg7cA2fM7bENDUMAzp6mQ9cDRwDfAL4KWI2JOybAHGpeVxwGaAiNgjaRdweEpfmztsfp/8c50PnA/Q1tbG7t276ejo6LWMFx63p9c89dQeu/Y4zTz3gWi2foPRYK1bRGyV9PfAL4HfAHeSnf8Dcs5D9/O+3uvWNqL152czyvx/LmvZiixXU0E/It4APihpFHAb8P6BKlBELAGWAEyZMiVGjhzJ9OnTe93v3Joxc5q16ex9j117nNrt/a2jo6Op+g1Gg7VukkaTXaVPAF4C/oWseWbA1J739V63q5evYvGGfd+yA31+NqPM/+eylq3IcvWp905EvATcA3wYGJW++gKMB7am5a3AUQBp+2HAi/n0OvuYlckngGci4lcR8TvgVuAj+Jy3IaCZ3jtHpCt8JI0APglsJAv+p6dsc4FVaXl1WidtvzsiIqWfmXo6TAAmAvf1V0XM+tEvgWmS3p7a5mcAT+Bz3oaAZpp3jgSWpXb9twArI+JHkp4AVki6FHgIuC7lvw64UVInsIOs9wIR8biklWRvnj3A/NRstF9qh0A26y8RsU7SLcCDZOfqQ2RNLz+mwHPerD/0GvQj4lHghDrpT1OnJ0JEvAZ8psGxLgMu63sxzVorIhYCC2uSfc7boOc7cs3MKsQzZ5kNYp7pzfrKV/pmZhXioG9mViEO+mZmFeKgb2ZWIQ76ZmYV4qBvZlYhDvpmZhXioG9mViEO+mZmFeKgb2ZWIQ76ZmYV4qBvZlYhDvpmZhXioG9mViEO+mZ1SBol6RZJP5O0UdKHJY2RtEbSU+nv6JRXkq6S1CnpUUmTc8eZm/I/JWlu42c0aw2Pp9+LetMyeszySrgS+GlEnC7pIODtwFeBuyJikaQFwALgIuAUsvlvJwJTgWuBqZLGkM2+NQUIYL2k1RGxs/XVMcv4St+shqTDgI+R5sCNiN9GxEvAHGBZyrYMOC0tzwFuiMxaYJSkI4GTgTURsSMF+jXArBZWxawbX+mbdTcB+BXwPUnHA+uBLwBtEbEt5XkOaEvL44DNuf23pLRG6d1IOh84H6CtrY2Ojo5uedpGwIXH7emx4PX2G2i7d+8u5HmbUdayFVkuB32z7oYDk4HPR8Q6SVeSNeW8KSJCUvTXE0bEEmAJwJQpU2L69Ond8ly9fBWLN/T8lt10dvf9BlpHRwf1ylsGZS1bkeVy845Zd1uALRGxLq3fQvYh8HxqtiH93Z62bwWOyu0/PqU1SjcrjIO+WY2IeA7YLOmYlDQDeAJYDXT1wJkLrErLq4FzUi+eacCu1Ax0BzBT0ujU02dmSjMrjJt3zOr7PLA89dx5GjiP7CJppaR5wLPAGSnv7cBsoBN4NeUlInZIugS4P+X7ZkTsaF0VzLrrNehLOgq4gexHqwCWRMSVqTvazUA7sAk4IyJ2ShJZd7fZZG+AcyPiwXSsucDX06EvjYhlmJVQRDxM1tWy1ow6eQOY3+A4S4Gl/Vs6s/3XTPPOHuDCiJgETAPmS5pE9sPWXRExEbiLvT905fssn0/WZ5lcn+WpwIeAhV03t5iZWWv0GvQjYlvXlXpE/BrYSNbtzH2WzcwGmT616UtqB04A1jFAfZZr+ys36s/aW3/lZtUeu5nj9mf/2rL2I+4PQ7luZoNV00Ff0kjgB8AXI+LlrOk+0599lmv7K48cObJuf9Zz6wyPsD9q+zU3c9z+7Atd1n7E/WEo181ssGqqy6akt5IF/OURcWtKdp9lM7NBptegn3rjXAdsjIhv5Ta5z7KZ2SDTTPPOR4DPARskPZzSvgoswn2WzcwGlV6DfkT8G6AGm91n2cxsEPEwDGZmFeKgb2ZWIQ76ZmYV4qBvZlYhDvpmZhXioG9mViEO+mZmFeKgb9aApGGSHpL0o7Q+QdI6SZ2Sbk4TrCDp4LTemba3545xcUp/UtLJxdTEbC8HfbPGvkA2lHiXy4ErIuJoYCcwL6XPA3am9CtSPtK8E2cCHyAbRvw7koa1qOxmdTnom9UhaTxwKvDdtC7gJLJJ0qH7HBJdc0vcAsxI+ecAKyLi9Yh4hmxokg+1pgZm9Tnom9X3beArwO/T+uHASxHRNeFCfj6IN+eKSNt3pfxNzSFh1kqeGN2shqRPAdsjYr2k6S16zn0mD6o3+UzbiN4n+Sli0poyT5ZT1rIVWS4HfbPuPgJ8WtJs4G3AO4Aryab+HJ6u5vPzQXTNFbFF0nDgMOBF+jCHRO3kQfUmn7l6+SoWb+j5LdufE/w0q8yT5ZS1bEWWy807ZjUi4uKIGB8R7WQ/xN4dEWcD9wCnp2y1c0h0zS1xesofKf3M1LtnAjARuK9F1TCry1f6+6G9ZkrFTYtOLagk1mIXASskXQo8RDa5EOnvjZI6gR1kHxRExOOSVgJPAHuA+RHxRuuLbbaXg75ZDyKiA+hIy09Tp/dNRLwGfKbB/pcBlw1cCc36xs07ZmYV4qBvZlYhDvpmZhXioG9mViEO+mZmFeKgb2ZWIQ76ZmYV4qBvZlYhvQZ9SUslbZf0WC5tjKQ1kp5Kf0endEm6Kk0a8aikybl95qb8T0maW++5zMxsYDVzpX892QQQeQuAuyJiInBXWgc4hWx8kYlkIwZeC9mHBLAQmEp2R+PCrg8KMzNrnV6DfkTcSzaeSF5+0ojaySRuiMxaslEJjwROBtZExI6I2AmsofsHiZmZDbD9HXunLSK2peXngLa03GjSiKYnk6gdV7zRuNO9jSverNpj789xD2Rc7LKO990fhnLdzAarAx5wLSJCUvRHYdLx9hlXfOTIkXXHnT63ZqTL/VU7/vj+HPdAxjAv63jf/WEo181ssNrf3jvPp2Yb0t/tKb3RpBFNTyZhZmYDZ3+v9LsmjVhE98kkLpC0guxH210RsU3SHcDf5H68nQlcvP/FNrN6PNeD9abXoC/pJmA6MFbSFrJeOIuAlZLmAc8CZ6TstwOzgU7gVeA8gIjYIekS4P6U75sRUfvj8KBV+0YDv9nMrJx6DfoRcVaDTTPq5A1gfoPjLAWW9ql0ZgWQdBRwA1kHhQCWRMSVqevxzUA7sAk4IyJ2ShLZHLqzyS52zo2IB9Ox5gJfT4e+NCKWYVYg35Fr1t0e4MKImARMA+ZLmoTvT7EhwEHfrEZEbOu6Uo+IXwMbyboY+/4UG/Q8R65ZDyS1AycA62jh/Sn17m9oG9H3+0hacZ9Eme/HKGvZiiyXg75ZA5JGAj8AvhgRL2dN95mBvj+l3v0NVy9fxeINfXvLHsg9JM0q8/0YZS1bkeVy845ZHZLeShbwl0fErSnZ96fYoOegb1Yj9ca5DtgYEd/Kbeq6PwW6359yThpldhrp/hTgDmCmpNHpB9yZKc2sMG7eMevuI8DngA2SHk5pX8X3p9gQ4KDfIr5TcvCIiH8D1GCz70+xQc3NO2ZmFeKgb2ZWIQ76ZmYV4qBvZlYhDvpmZhXi3jsF6erNc+Fxe96crcs9esxsoPlK38ysQnylbzaEeYIfq+UrfTOzCnHQNzOrEDfvlIiHajCzgeYrfTOzCnHQNzOrEDfvmFWMmxGrzUG/xNzdzsz6m5t3zMwqpOVX+pJmAVcCw4DvRsSiVpdhKPG3gfIr+znvc6haWhr0JQ0DrgE+CWwB7pe0OiKeaGU5hjq32ZbHYD3nfQ4NXa2+0v8Q0BkRTwNIWgHMAUr9Bhjs6l3J1fKbesAMiXO+p3MoP2hgns+pclI2vWeLnkw6HZgVEf81rX8OmBoRF+TynA+cn1aPAV4EXmhZIVtvLEO3fmWp23si4oginriZcz6l1573T9Y5XFlez1plLReUt2wDXa6G53zpeu9ExBJgSde6pAciYkqBRRpQQ7l+Q7lu/a32vK+nrK9nWcsF5S1bkeVqde+drcBRufXxKc1sqPI5b6XS6qB/PzBR0gRJBwFnAqtbXAazVvI5b6XS0uadiNgj6QLgDrLua0sj4vFeduvxK+8QMJTrN5Tr1pT9POcbKevrWdZyQXnLVli5WvpDrpmZFct35JqZVYiDvplZhZQ66EuaJelJSZ2SFhRdngMhaamk7ZIey6WNkbRG0lPp7+giy7i/JB0l6R5JT0h6XNIXUvqQqF/Rinwf9PC//YakrZIeTo/ZuX0uTmV9UtLJA1y+TZI2pDI8kNLqnnfKXJXK9qikyQNUpmNyr8vDkl6W9MWyvGZERCkfZD96/QJ4L3AQ8AgwqehyHUB9PgZMBh7Lpf0dsCAtLwAuL7qc+1m3I4HJaflQ4OfApKFSv4Jf20LfBz38b78BfKlO/kmpjAcDE1LZhw1g+TYBY2vS6p53wGzgJ4CAacC6Fv3/ngPeU5bXrMxX+m/evh4RvwW6bl8flCLiXmBHTfIcYFlaXgac1tJC9ZOI2BYRD6blXwMbgXEMkfoVrND3QQ//20bmACsi4vWIeAboJKtDKzU67+YAN0RmLTBK0pEDXJYZwC8i4tke8rT0NStz0B8HbM6tb6Hnk20waouIbWn5OaCtyML0B0ntwAnAOoZg/QpQmvdBzf8W4ILUTLI013TX6vIGcKek9WkoC2h83hXxWp4J3JRbL/w1K3PQr5TIvucN6v6zkkYCPwC+GBEv57cNhfpVWZ3/7bXA+4APAtuAxQUV7aMRMRk4BZgv6WP5jUWed+lmvE8D/5KSSvGalTnoV+H29ee7vl6mvwJKs0EAAAFWSURBVNsLLs9+k/RWsqCwPCJuTclDpn4FKvx9UO9/GxHPR8QbEfF74J/Y2xzR0vJGxNb0dztwWypHo/Ou1a/lKcCDEfF8KmMpXrMyB/0q3L6+GpiblucCqwosy36TJOA6YGNEfCu3aUjUr2CFvg8a/W9r2sL/E9DVK201cKakgyVNACYC9w1Q2Q6RdGjXMjAzlaPRebcaOCf14pkG7Mo1Aw2Es8g17ZThNQPK23sn9v7a/nOyX7O/VnR5DrAuN5F9pfsdWZvdPOBw4C7gKeBfgTFFl3M/6/ZRsq/QjwIPp8fsoVK/oh9Fvg96+N/eCGxI6auBI3P7fC2V9UnglAEs23vJer08Ajze9do0Ou/Ieu1ck8q2AZgygGU7hGxY+MNyaYW/ZhHhYRjMzKqkzM07ZmbWzxz0zcwqxEHfzKxCHPTNzCrEQd/MrEIc9M3MKsRB38ysQv4/epr8iMvkd9cAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kJQdijuYfegm",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "5a5f915e-62ef-452e-d57b-d300f1aaaa65"
      },
      "source": [
        "#Function for getting the Maximum Review length  \n",
        "count=0 \n",
        "for i in reviewsData['Cleaned_Text']:\n",
        "    if(len(i.split())<=35):\n",
        "        count=count+1\n",
        "print(count/len(reviewsData['Cleaned_Text']))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.630776211084741\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WILwzlQhgdSQ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "0f2fc601-6ddf-45f7-9586-fd7a78405fb7"
      },
      "source": [
        "#Function for getting the Maximum Summary length\n",
        "count=0\n",
        "for i in reviewsData['Cleaned_Summary']:\n",
        "    if(len(i.split())<=8):\n",
        "        count=count+1\n",
        "print(count/len(reviewsData['Cleaned_Summary']))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.9437386569872959\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3cJavH3YgdWI",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#From the above data we got an idea about maximum lengths of review and summary\n",
        "max_text_len = 35\n",
        "max_summary_len = 8"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KBu0OPxGgdYV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Adding START and END tags to summary for better decoding\n",
        "cleaned_text =np.array(reviewsData['Cleaned_Text'])\n",
        "cleaned_summary=np.array(reviewsData['Cleaned_Summary'])\n",
        "\n",
        "short_text=[]\n",
        "short_summary=[]\n",
        "\n",
        "for i in range(len(cleaned_text)):\n",
        "    if(len(cleaned_summary[i].split())<=max_summary_len and len(cleaned_text[i].split())<=max_text_len):\n",
        "        short_text.append(cleaned_text[i])\n",
        "        short_summary.append(cleaned_summary[i])\n",
        "        \n",
        "df=pd.DataFrame({'text':short_text,'summary':short_summary})"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UmKhWPrsgdcF",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df['summary'] = df['summary'].apply(lambda x : 'sostok '+ x + ' eostok')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-iHbqd8ogdeO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Splitting the Dataset\n",
        "from sklearn.model_selection import train_test_split\n",
        "X_train,X_test,y_train,y_test=train_test_split(np.array(df['text']),np.array(df['summary']),test_size=0.2,random_state=0,shuffle=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eHUKg3fIgdjH",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Preparing Tokenizer\n",
        "\n",
        "#Text Tokenizer\n",
        "from keras.preprocessing.text import Tokenizer \n",
        "from keras.preprocessing.sequence import pad_sequences\n",
        "\n",
        "#preparing a tokenizer for reviews on training data\n",
        "X_tokenizer = Tokenizer() \n",
        "X_tokenizer.fit_on_texts(list(X_train))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "K9YrJJHigdo4",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "fe81edaa-e2b2-43df-89ed-2db282c2407a"
      },
      "source": [
        "#Rarewords and their coverage in review\n",
        "thresh = 4  #If a word whose count is less than threshold i.e 4, then it's considered as rare word \n",
        "\n",
        "cnt = 0      #denotes no. of rare words whose count falls below threshold\n",
        "tot_cnt = 0  #denotes size of unique words in the text\n",
        "freq = 0\n",
        "tot_freq = 0\n",
        "\n",
        "for key,value in X_tokenizer.word_counts.items():\n",
        "    tot_cnt=tot_cnt+1\n",
        "    tot_freq=tot_freq+value\n",
        "    if(value<thresh):\n",
        "        cnt=cnt+1\n",
        "        freq=freq+value\n",
        "    \n",
        "print(\"% of rare words in vocabulary:\",(cnt/tot_cnt)*100)\n",
        "print(\"Total Coverage of rare words:\",(freq/tot_freq)*100)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "% of rare words in vocabulary: 65.60676192392836\n",
            "Total Coverage of rare words: 4.94040608149725\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vMviW6OJgdsO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Defining the Tokenizer with top most common words for reviews\n",
        "\n",
        "#Preparing a Tokenizer for reviews on training data\n",
        "X_tokenizer = Tokenizer(num_words=tot_cnt-cnt)   #provides top most common words\n",
        "X_tokenizer.fit_on_texts(list(X_train))\n",
        "\n",
        "#Converting text sequences into integer sequences\n",
        "X_train_seq    =   X_tokenizer.texts_to_sequences(X_train) \n",
        "X_test_seq   =   X_tokenizer.texts_to_sequences(X_test)\n",
        "\n",
        "#Padding zero upto maximum length\n",
        "X_train    =   pad_sequences(X_train_seq,  maxlen = max_text_len, padding = 'post')\n",
        "X_test   =   pad_sequences(X_test_seq, maxlen = max_text_len, padding = 'post')\n",
        "\n",
        "#Size of vocabulary (+1 for padding token)\n",
        "X_voc   =  X_tokenizer.num_words + 1"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xxwsrZLugdnP",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "a6981ef6-520d-471c-a2cc-a1493a1fe985"
      },
      "source": [
        "X_voc "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "5128"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 85
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e-2EhbNrgdhz",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Summary Tokenizer\n",
        "\n",
        "#Preparing a Tokenizer for summaries on training data\n",
        "y_tokenizer = Tokenizer()   \n",
        "y_tokenizer.fit_on_texts(list(y_train))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Jz3AyGNzhDEk",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "03ed98b0-b27e-4edf-a830-730851a94d51"
      },
      "source": [
        "#Rarewords and their coverage in summary\n",
        "\n",
        "thresh = 6  ##If a word whose count is less than threshold i.e 6, then it's considered as rare word \n",
        "\n",
        "cnt = 0\n",
        "tot_cnt = 0\n",
        "freq = 0\n",
        "tot_freq = 0\n",
        "\n",
        "for key,value in y_tokenizer.word_counts.items():\n",
        "    tot_cnt = tot_cnt+1\n",
        "    tot_freq = tot_freq+value\n",
        "    if(value<thresh):\n",
        "        cnt = cnt+1\n",
        "        freq = freq+value\n",
        "    \n",
        "print(\"% of rare words in vocabulary:\",(cnt/tot_cnt)*100)\n",
        "print(\"Total Coverage of rare words:\",(freq/tot_freq)*100)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "% of rare words in vocabulary: 80.80321285140563\n",
            "Total Coverage of rare words: 9.106280519972405\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GcHAzcb5hDGk",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Defining Tokenizer with the most common words in summary\n",
        "\n",
        "#Preparing a tokenizer for summaries on training data\n",
        "y_tokenizer = Tokenizer(num_words=tot_cnt-cnt)  #provides top most common words\n",
        "y_tokenizer.fit_on_texts(list(y_train))\n",
        "\n",
        "#Converting text sequences into integer sequences\n",
        "y_train_seq    =   y_tokenizer.texts_to_sequences(y_train) \n",
        "y_test_seq   =   y_tokenizer.texts_to_sequences(y_test) \n",
        "\n",
        "#Padding zero upto maximum length\n",
        "y_train    =   pad_sequences(y_train_seq, maxlen=max_summary_len, padding='post')\n",
        "y_test   =   pad_sequences(y_test_seq, maxlen=max_summary_len, padding='post')\n",
        "\n",
        "#size of vocabulary\n",
        "y_voc  =   y_tokenizer.num_words +1"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LGwrtZf7zwuQ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "1a0fe785-4e09-459a-8373-892f254e6268"
      },
      "source": [
        "y_voc"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "957"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 89
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GoihGf-2hDKE",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "04c87cb7-18a2-48cd-dc6b-dd83c8830362"
      },
      "source": [
        "#Checking the length of training data\n",
        "y_tokenizer.word_counts['sostok'],len(y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(14032, 14032)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 90
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "seNf04wKhDMv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Deleting rows containing START and END tokens\n",
        "#For Training set\n",
        "ind=[]\n",
        "for i in range(len(y_train)):\n",
        "    cnt=0\n",
        "    for j in y_train[i]:\n",
        "        if j!=0:\n",
        "            cnt=cnt+1\n",
        "    if(cnt==2):\n",
        "        ind.append(i)\n",
        "\n",
        "y_train=np.delete(y_train,ind, axis=0)\n",
        "X_train=np.delete(X_train,ind, axis=0)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QbQ-bOakhDSQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#For Validation set\n",
        "ind=[]\n",
        "for i in range(len(y_test)):\n",
        "    cnt=0\n",
        "    for j in y_test[i]:\n",
        "        if j!=0:\n",
        "            cnt=cnt+1\n",
        "    if(cnt==2):\n",
        "        ind.append(i)\n",
        "\n",
        "y_test=np.delete(y_test,ind, axis=0)\n",
        "X_test=np.delete(X_test,ind, axis=0)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7c6DDCDWhDWa",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Model Building\n",
        "\n",
        "#Adding Custom Attention layer \n",
        "\n",
        "import tensorflow as tf\n",
        "import os\n",
        "from tensorflow.python.keras.layers import Layer\n",
        "from tensorflow.python.keras import backend as K\n",
        "\n",
        "\n",
        "class AttentionLayer(Layer):\n",
        "    \"\"\"\n",
        "    This class implements Bahdanau attention (https://arxiv.org/pdf/1409.0473.pdf).\n",
        "    There are three sets of weights introduced W_a, U_a, and V_a\n",
        "     \"\"\"\n",
        "\n",
        "    def __init__(self, **kwargs):\n",
        "        super(AttentionLayer, self).__init__(**kwargs)\n",
        "\n",
        "    def build(self, input_shape):\n",
        "        assert isinstance(input_shape, list)\n",
        "        # Create a trainable weight variable for this layer.\n",
        "\n",
        "        self.W_a = self.add_weight(name='W_a',\n",
        "                                   shape=tf.TensorShape((input_shape[0][2], input_shape[0][2])),\n",
        "                                   initializer='uniform',\n",
        "                                   trainable=True)\n",
        "        self.U_a = self.add_weight(name='U_a',\n",
        "                                   shape=tf.TensorShape((input_shape[1][2], input_shape[0][2])),\n",
        "                                   initializer='uniform',\n",
        "                                   trainable=True)\n",
        "        self.V_a = self.add_weight(name='V_a',\n",
        "                                   shape=tf.TensorShape((input_shape[0][2], 1)),\n",
        "                                   initializer='uniform',\n",
        "                                   trainable=True)\n",
        "\n",
        "        super(AttentionLayer, self).build(input_shape)  # Be sure to call this at the end\n",
        "\n",
        "    def call(self, inputs, verbose=False):\n",
        "        \"\"\"\n",
        "        inputs: [encoder_output_sequence, decoder_output_sequence]\n",
        "        \"\"\"\n",
        "        assert type(inputs) == list\n",
        "        encoder_out_seq, decoder_out_seq = inputs\n",
        "        if verbose:\n",
        "            print('encoder_out_seq>', encoder_out_seq.shape)\n",
        "            print('decoder_out_seq>', decoder_out_seq.shape)\n",
        "\n",
        "        def energy_step(inputs, states):\n",
        "            \"\"\" Step function for computing energy for a single decoder state \"\"\"\n",
        "\n",
        "            assert_msg = \"States must be a list. However states {} is of type {}\".format(states, type(states))\n",
        "            assert isinstance(states, list) or isinstance(states, tuple), assert_msg\n",
        "\n",
        "            \"\"\" Some parameters required for shaping tensors\"\"\"\n",
        "            en_seq_len, en_hidden = encoder_out_seq.shape[1], encoder_out_seq.shape[2]\n",
        "            de_hidden = inputs.shape[-1]\n",
        "\n",
        "            \"\"\" Computing S.Wa where S=[s0, s1, ..., si]\"\"\"\n",
        "            # <= batch_size*en_seq_len, latent_dim\n",
        "            reshaped_enc_outputs = K.reshape(encoder_out_seq, (-1, en_hidden))\n",
        "            # <= batch_size*en_seq_len, latent_dim\n",
        "            W_a_dot_s = K.reshape(K.dot(reshaped_enc_outputs, self.W_a), (-1, en_seq_len, en_hidden))\n",
        "            if verbose:\n",
        "                print('wa.s>',W_a_dot_s.shape)\n",
        "\n",
        "            \"\"\" Computing hj.Ua \"\"\"\n",
        "            U_a_dot_h = K.expand_dims(K.dot(inputs, self.U_a), 1)  # <= batch_size, 1, latent_dim\n",
        "            if verbose:\n",
        "                print('Ua.h>',U_a_dot_h.shape)\n",
        "\n",
        "            \"\"\" tanh(S.Wa + hj.Ua) \"\"\"\n",
        "            # <= batch_size*en_seq_len, latent_dim\n",
        "            reshaped_Ws_plus_Uh = K.tanh(K.reshape(W_a_dot_s + U_a_dot_h, (-1, en_hidden)))\n",
        "            if verbose:\n",
        "                print('Ws+Uh>', reshaped_Ws_plus_Uh.shape)\n",
        "\n",
        "            \"\"\" softmax(va.tanh(S.Wa + hj.Ua)) \"\"\"\n",
        "            # <= batch_size, en_seq_len\n",
        "            e_i = K.reshape(K.dot(reshaped_Ws_plus_Uh, self.V_a), (-1, en_seq_len))\n",
        "            # <= batch_size, en_seq_len\n",
        "            e_i = K.softmax(e_i)\n",
        "\n",
        "            if verbose:\n",
        "                print('ei>', e_i.shape)\n",
        "\n",
        "            return e_i, [e_i]\n",
        "\n",
        "        def context_step(inputs, states):\n",
        "            \"\"\" Step function for computing ci using ei \"\"\"\n",
        "            # <= batch_size, hidden_size\n",
        "            c_i = K.sum(encoder_out_seq * K.expand_dims(inputs, -1), axis=1)\n",
        "            if verbose:\n",
        "                print('ci>', c_i.shape)\n",
        "            return c_i, [c_i]\n",
        "\n",
        "        def create_inital_state(inputs, hidden_size):\n",
        "            # We are not using initial states, but need to pass something to K.rnn funciton\n",
        "            fake_state = K.zeros_like(inputs)  # <= (batch_size, enc_seq_len, latent_dim\n",
        "            fake_state = K.sum(fake_state, axis=[1, 2])  # <= (batch_size)\n",
        "            fake_state = K.expand_dims(fake_state)  # <= (batch_size, 1)\n",
        "            fake_state = K.tile(fake_state, [1, hidden_size])  # <= (batch_size, latent_dim\n",
        "            return fake_state\n",
        "\n",
        "        fake_state_c = create_inital_state(encoder_out_seq, encoder_out_seq.shape[-1])\n",
        "        fake_state_e = create_inital_state(encoder_out_seq, encoder_out_seq.shape[1])  # <= (batch_size, enc_seq_len, latent_dim\n",
        "\n",
        "        \"\"\" Computing energy outputs \"\"\"\n",
        "        # e_outputs => (batch_size, de_seq_len, en_seq_len)\n",
        "        last_out, e_outputs, _ = K.rnn(\n",
        "            energy_step, decoder_out_seq, [fake_state_e],\n",
        "        )\n",
        "\n",
        "        \"\"\" Computing context vectors \"\"\"\n",
        "        last_out, c_outputs, _ = K.rnn(\n",
        "            context_step, e_outputs, [fake_state_c],\n",
        "        )\n",
        "\n",
        "        return c_outputs, e_outputs\n",
        "\n",
        "    def compute_output_shape(self, input_shape):\n",
        "        \"\"\" Outputs produced by the layer \"\"\"\n",
        "        return [\n",
        "            tf.TensorShape((input_shape[1][0], input_shape[1][1], input_shape[1][2])),\n",
        "            tf.TensorShape((input_shape[1][0], input_shape[1][1], input_shape[0][1]))\n",
        "        ]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QBy5PM7MhDQP",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 649
        },
        "outputId": "03f9d700-c055-4d3b-d1b9-7d7b1ae33cda"
      },
      "source": [
        "from keras import backend as K \n",
        "K.clear_session()  #Resets all state generated by Keras\n",
        "\n",
        "latent_dim = 256\n",
        "embedding_dim = 256\n",
        "\n",
        "# Encoder\n",
        "encoder_inputs = Input(shape=(max_text_len,))\n",
        "\n",
        "#embedding layer\n",
        "enc_emb =  Embedding(X_voc, embedding_dim,trainable=True)(encoder_inputs)\n",
        "\n",
        "#encoder lstm 1\n",
        "encoder_lstm1 = LSTM(latent_dim,return_sequences=True,return_state=True,dropout=0.4,recurrent_dropout=0.4)\n",
        "encoder_output1, state_h1, state_c1 = encoder_lstm1(enc_emb)\n",
        "\n",
        "#encoder lstm 2\n",
        "encoder_lstm2 = LSTM(latent_dim,return_sequences=True,return_state=True,dropout=0.4,recurrent_dropout=0.4)\n",
        "encoder_output2, state_h2, state_c2 = encoder_lstm2(encoder_output1)\n",
        "\n",
        "#encoder lstm 3\n",
        "encoder_lstm3= LSTM(latent_dim, return_state=True, return_sequences=True,dropout=0.4,recurrent_dropout=0.4)\n",
        "encoder_outputs, state_h, state_c= encoder_lstm3(encoder_output2)\n",
        "\n",
        "#Setting up the Decoder using 'encoder_states' as initial state\n",
        "decoder_inputs = Input(shape=(None,))\n",
        "\n",
        "#Embedding layer\n",
        "dec_emb_layer = Embedding(y_voc, embedding_dim,trainable=True)\n",
        "dec_emb = dec_emb_layer(decoder_inputs)\n",
        "\n",
        "decoder_lstm = LSTM(latent_dim, return_sequences=True, return_state=True,dropout=0.4,recurrent_dropout=0.2)\n",
        "decoder_outputs,decoder_fwd_state, decoder_back_state = decoder_lstm(dec_emb,initial_state=[state_h, state_c])\n",
        "\n",
        "#Attention layer\n",
        "attn_layer = AttentionLayer(name='attention_layer')\n",
        "attn_out, attn_states = attn_layer([encoder_outputs, decoder_outputs])\n",
        "\n",
        "#Concating Attention input and Decoder LSTM output\n",
        "decoder_concat_input = Concatenate(axis=-1, name='concat_layer')([decoder_outputs, attn_out])\n",
        "\n",
        "#Dense layer\n",
        "decoder_dense =  TimeDistributed(Dense(y_voc, activation='softmax'))\n",
        "decoder_outputs = decoder_dense(decoder_concat_input)\n",
        "\n",
        "#Defining the model \n",
        "model = Model([encoder_inputs, decoder_inputs], decoder_outputs)\n",
        "\n",
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:Entity <bound method AttentionLayer.call of <__main__.AttentionLayer object at 0x7fdb2ed21208>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Num'\n",
            "WARNING: Entity <bound method AttentionLayer.call of <__main__.AttentionLayer object at 0x7fdb2ed21208>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Num'\n",
            "Model: \"model\"\n",
            "__________________________________________________________________________________________________\n",
            "Layer (type)                    Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            "input_1 (InputLayer)            [(None, 35)]         0                                            \n",
            "__________________________________________________________________________________________________\n",
            "embedding (Embedding)           (None, 35, 256)      1312768     input_1[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "lstm (LSTM)                     [(None, 35, 256), (N 525312      embedding[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "input_2 (InputLayer)            [(None, None)]       0                                            \n",
            "__________________________________________________________________________________________________\n",
            "lstm_1 (LSTM)                   [(None, 35, 256), (N 525312      lstm[0][0]                       \n",
            "__________________________________________________________________________________________________\n",
            "embedding_1 (Embedding)         (None, None, 256)    244992      input_2[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "lstm_2 (LSTM)                   [(None, 35, 256), (N 525312      lstm_1[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "lstm_3 (LSTM)                   [(None, None, 256),  525312      embedding_1[0][0]                \n",
            "                                                                 lstm_2[0][1]                     \n",
            "                                                                 lstm_2[0][2]                     \n",
            "__________________________________________________________________________________________________\n",
            "attention_layer (AttentionLayer ((None, None, 256),  131328      lstm_2[0][0]                     \n",
            "                                                                 lstm_3[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "concat_layer (Concatenate)      (None, None, 512)    0           lstm_3[0][0]                     \n",
            "                                                                 attention_layer[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "time_distributed (TimeDistribut (None, None, 957)    490941      concat_layer[0][0]               \n",
            "==================================================================================================\n",
            "Total params: 4,281,277\n",
            "Trainable params: 4,281,277\n",
            "Non-trainable params: 0\n",
            "__________________________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b5idXyLEgdQl",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 976
        },
        "outputId": "d4873af8-63b7-4339-fe6a-27e2b7e7e0e1"
      },
      "source": [
        "#Visualize the Model\n",
        "from tensorflow.keras.utils import plot_model\n",
        "\n",
        "plot_model(model, to_file='model_plot.png', show_shapes=True, show_layer_names=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 95
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FqgFcv1sht7R",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Adding Metrics\n",
        "model.compile(optimizer='rmsprop' , loss='sparse_categorical_crossentropy' , metrics=['accuracy'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DXsoihrLhuBP",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Adding Callback\n",
        "es = EarlyStopping(monitor='val_loss', mode='min', verbose=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tPICqxDehuGX",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 595
        },
        "outputId": "97517801-977a-405c-9280-9b728b26e6f6"
      },
      "source": [
        "#Training the Model\n",
        "%tensorflow_version 1.x\n",
        "history = model.fit([X_train,y_train[:,:-1]], y_train.reshape(y_train.shape[0],y_train.shape[1], 1)[:,1:] ,epochs=50,callbacks=[es],batch_size= 64, validation_data=([X_test,y_test[:,:-1]], y_test.reshape(y_test.shape[0],y_test.shape[1], 1)[:,1:]))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 13414 samples, validate on 3358 samples\n",
            "Epoch 1/50\n",
            "13414/13414 [==============================] - 365s 27ms/sample - loss: 2.6544 - acc: 0.5845 - val_loss: 2.3949 - val_acc: 0.6132\n",
            "Epoch 2/50\n",
            "13414/13414 [==============================] - 362s 27ms/sample - loss: 2.3650 - acc: 0.6115 - val_loss: 2.2825 - val_acc: 0.6200\n",
            "Epoch 3/50\n",
            "13414/13414 [==============================] - 366s 27ms/sample - loss: 2.2386 - acc: 0.6207 - val_loss: 2.1829 - val_acc: 0.6261\n",
            "Epoch 4/50\n",
            "13414/13414 [==============================] - 367s 27ms/sample - loss: 2.1318 - acc: 0.6294 - val_loss: 2.1235 - val_acc: 0.6343\n",
            "Epoch 5/50\n",
            "13414/13414 [==============================] - 359s 27ms/sample - loss: 2.0499 - acc: 0.6372 - val_loss: 2.0656 - val_acc: 0.6386\n",
            "Epoch 6/50\n",
            "13414/13414 [==============================] - 361s 27ms/sample - loss: 1.9898 - acc: 0.6422 - val_loss: 2.0401 - val_acc: 0.6419\n",
            "Epoch 7/50\n",
            "13414/13414 [==============================] - 357s 27ms/sample - loss: 1.9399 - acc: 0.6459 - val_loss: 2.0233 - val_acc: 0.6423\n",
            "Epoch 8/50\n",
            "13414/13414 [==============================] - 356s 27ms/sample - loss: 1.8962 - acc: 0.6492 - val_loss: 1.9968 - val_acc: 0.6460\n",
            "Epoch 9/50\n",
            "13414/13414 [==============================] - 354s 26ms/sample - loss: 1.8574 - acc: 0.6519 - val_loss: 1.9909 - val_acc: 0.6460\n",
            "Epoch 10/50\n",
            "13414/13414 [==============================] - 356s 27ms/sample - loss: 1.8219 - acc: 0.6558 - val_loss: 1.9903 - val_acc: 0.6463\n",
            "Epoch 11/50\n",
            "13414/13414 [==============================] - 356s 27ms/sample - loss: 1.7846 - acc: 0.6584 - val_loss: 1.9797 - val_acc: 0.6488\n",
            "Epoch 12/50\n",
            "13414/13414 [==============================] - 356s 27ms/sample - loss: 1.7540 - acc: 0.6614 - val_loss: 1.9786 - val_acc: 0.6501\n",
            "Epoch 13/50\n",
            "13414/13414 [==============================] - 356s 27ms/sample - loss: 1.7227 - acc: 0.6651 - val_loss: 1.9762 - val_acc: 0.6480\n",
            "Epoch 14/50\n",
            "13414/13414 [==============================] - 355s 26ms/sample - loss: 1.6929 - acc: 0.6677 - val_loss: 1.9730 - val_acc: 0.6486\n",
            "Epoch 15/50\n",
            "13414/13414 [==============================] - 350s 26ms/sample - loss: 1.6618 - acc: 0.6695 - val_loss: 1.9694 - val_acc: 0.6478\n",
            "Epoch 16/50\n",
            "13414/13414 [==============================] - 345s 26ms/sample - loss: 1.6314 - acc: 0.6727 - val_loss: 1.9727 - val_acc: 0.6456\n",
            "Epoch 00016: early stopping\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ohK9YLnPhuLH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "outputId": "c94206f6-bbf7-4168-ef44-625fdd042782"
      },
      "source": [
        "#Visualizing Accuracy \n",
        "from matplotlib import pyplot\n",
        "pyplot.plot(history.history['acc'], label='train') \n",
        "pyplot.plot(history.history['val_acc'], label='test') \n",
        "pyplot.legend() \n",
        "pyplot.show()"
      ],
      "execution_count": null,
      "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": "UQcf72bnhuRj",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "outputId": "d6c60752-0e45-45d3-fb82-680941f470aa"
      },
      "source": [
        "#Visualizing Loss \n",
        "pyplot.plot(history.history['loss'], label='train') \n",
        "pyplot.plot(history.history['val_loss'], label='test') \n",
        "pyplot.legend() \n",
        "pyplot.show()"
      ],
      "execution_count": null,
      "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": "7wtfEh16huVw",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Building Dictionary for Source Vocabulary\n",
        "reverse_target_word_index=y_tokenizer.index_word \n",
        "reverse_source_word_index=X_tokenizer.index_word \n",
        "target_word_index=y_tokenizer.word_index"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7mDay3BohueE",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 71
        },
        "outputId": "0478b736-55f8-4661-bf5a-ebf3a9b5de72"
      },
      "source": [
        "#Inference/Validation Phase\n",
        "#Encoding the input sequence to get the feature vector\n",
        "encoder_model = Model(inputs=encoder_inputs,outputs=[encoder_outputs, state_h, state_c])\n",
        "\n",
        "#Decoder setup\n",
        "#These tensors will hold the states of the previous time step\n",
        "decoder_state_input_h = Input(shape=(latent_dim,))\n",
        "decoder_state_input_c = Input(shape=(latent_dim,))\n",
        "decoder_hidden_state_input = Input(shape=(max_text_len,latent_dim))\n",
        "\n",
        "#Getting the embeddings of the decoder sequence\n",
        "dec_emb2= dec_emb_layer(decoder_inputs) \n",
        "\n",
        "#Setting the initial states to the states from the previous time step for better prediction\n",
        "decoder_outputs2, state_h2, state_c2 = decoder_lstm(dec_emb2, initial_state=[decoder_state_input_h, decoder_state_input_c])\n",
        "\n",
        "#Attention inference\n",
        "attn_out_inf, attn_states_inf = attn_layer([decoder_hidden_state_input, decoder_outputs2])\n",
        "decoder_inf_concat = Concatenate(axis=-1, name='concat')([decoder_outputs2, attn_out_inf])\n",
        "\n",
        "#Adding Dense softmax layer to generate proability distribution over the target vocabulary\n",
        "decoder_outputs2 = decoder_dense(decoder_inf_concat) \n",
        "\n",
        "#Final Decoder model\n",
        "decoder_model = Model(\n",
        "    [decoder_inputs] + [decoder_hidden_state_input,decoder_state_input_h, decoder_state_input_c],\n",
        "    [decoder_outputs2] + [state_h2, state_c2])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:Entity <bound method AttentionLayer.call of <__main__.AttentionLayer object at 0x7fdb2ed21208>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Num'\n",
            "WARNING: Entity <bound method AttentionLayer.call of <__main__.AttentionLayer object at 0x7fdb2ed21208>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Num'\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vzLFjn-Zhuhd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Function defining the implementation of inference process\n",
        "def decode_sequence(input_seq):\n",
        "    #Encoding the input as state vectors\n",
        "    e_out, e_h, e_c = encoder_model.predict(input_seq)\n",
        "    \n",
        "    #Generating empty target sequence of length 1\n",
        "    target_seq = np.zeros((1,1))\n",
        "    \n",
        "    #Populating the first word of target sequence with the start word\n",
        "    target_seq[0, 0] = target_word_index['sostok']\n",
        "\n",
        "    stop_condition = False\n",
        "    decoded_sentence = ''\n",
        "    while not stop_condition:\n",
        "      \n",
        "        output_tokens, h, c = decoder_model.predict([target_seq] + [e_out, e_h, e_c])\n",
        "\n",
        "        #Sampling a token\n",
        "        sampled_token_index = np.argmax(output_tokens[0, -1, :])\n",
        "        sampled_token = reverse_target_word_index[sampled_token_index]\n",
        "        \n",
        "        if(sampled_token!='eostok'):\n",
        "            decoded_sentence += ' '+sampled_token\n",
        "\n",
        "        #Exit condition: either hit max length or find stop word\n",
        "        if (sampled_token == 'eostok'  or len(decoded_sentence.split()) >= (max_summary_len-1)):\n",
        "            stop_condition = True\n",
        "\n",
        "        #Updating the target sequence (of length 1)\n",
        "        target_seq = np.zeros((1,1))\n",
        "        target_seq[0, 0] = sampled_token_index\n",
        "\n",
        "        #Updating internal states\n",
        "        e_h, e_c = h, c\n",
        "\n",
        "    return decoded_sentence"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "q-pLsERkhubk",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Functions to convert an integer sequence to a word sequence for summary as well as reviews \n",
        "def seq2summary(input_seq):\n",
        "    newString=''\n",
        "    for i in input_seq:\n",
        "        if((i!=0 and i!=target_word_index['sostok']) and i!=target_word_index['eostok']):\n",
        "            newString=newString+reverse_target_word_index[i]+' '\n",
        "    return newString\n",
        "\n",
        "def seq2text(input_seq):\n",
        "    newString=''\n",
        "    for i in input_seq:\n",
        "        if(i!=0):\n",
        "            newString=newString+reverse_source_word_index[i]+' '\n",
        "    return newString"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UWHUjVHghuZ7",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "e8886fc9-6ebd-4bf7-dc43-679d8b782e0f"
      },
      "source": [
        "#Summaries generated by the model\n",
        "\n",
        "for i in range(0,20):\n",
        "    print(\"Review:\",seq2text(X_train[i]))\n",
        "    print(\"Original summary:\",seq2summary(y_train[i]))\n",
        "    print(\"Predicted summary:\",decode_sequence(X_train[i].reshape(1,max_text_len)))\n",
        "    print(\"\\n\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Review: noodles box broken sauce salted good flavor threw would recommend purchasing product \n",
            "Original summary: not good \n",
            "Predicted summary:  not as good as\n",
            "\n",
            "\n",
            "Review: nothing wrong product far dogs concerned always want one day try limit amount good pocket book \n",
            "Original summary: the dogs love it \n",
            "Predicted summary:  dog treats\n",
            "\n",
            "\n",
            "Review: love gluten free granola plain cranberry flavors great breakfast yogurt snack right bag definitely recommend product clearly says bag everything contain makes easier reading ingredients every package delicious \n",
            "Original summary: delicious love them \n",
            "Predicted summary:  great gluten free snack\n",
            "\n",
            "\n",
            "Review: gave product one zero option like generic tomato soup bits fish lobster soup shell salty fishy nasty inedible cans dented complete waste money beyond disappointing made \n",
            "Original summary: disgusting nasty \n",
            "Predicted summary:  not bad\n",
            "\n",
            "\n",
            "Review: taste okay away hidden fact truly decaffeinated coffee place find bottom box want first thing morning \n",
            "Original summary: just okay decaf \n",
            "Predicted summary:  not bad\n",
            "\n",
            "\n",
            "Review: got decaf keurig assortment pack enjoyed flavors weak post review \n",
            "Original summary: no flavor very weak \n",
            "Predicted summary:  weak\n",
            "\n",
            "\n",
            "Review: bought juice really like mangoes great refreshing smooth beverage everybody family enjoy buy \n",
            "Original summary: nice sweet mango \n",
            "Predicted summary:  sweet\n",
            "\n",
            "\n",
            "Review: oregon chai tea brews fast tastes better commercial chai tea oh much better starbuck much economical truly yummy treat \n",
            "Original summary: heavenly tea \n",
            "Predicted summary:  great tea\n",
            "\n",
            "\n",
            "Review: thing buy lot stores much say love loves bought numerous times \n",
            "Original summary: great product great price \n",
            "Predicted summary:  great product\n",
            "\n",
            "\n",
            "Review: good brand good tea like blend alot purchase also makes good iced tea refreshing \n",
            "Original summary: tea \n",
            "Predicted summary:  tea\n",
            "\n",
            "\n",
            "Review: tasty satisfying meat protien sticks perfect meal snack addition protien diet pepper flavor adds something missing regular flavor variety snacks recommended \n",
            "Original summary: meat of \n",
            "Predicted summary:  delicious\n",
            "\n",
            "\n",
            "Review: first time gave two little dogs hard treats real nicely next day noticed really excited opened package front keep mind treats really hard know large breeds smaller pups definitely trouble \n",
            "Original summary: even dog has \n",
            "Predicted summary:  my dogs love these\n",
            "\n",
            "\n",
            "Review: chips great delicious healthier regular deli store chips first purchased supermarket found delicious found amazon cheaper ordered variety flavor flavors liking con addicting warned purchased amazon \n",
            "Original summary: delicious \n",
            "Predicted summary:  great chips\n",
            "\n",
            "\n",
            "Review: actually order wrote seller indeed roasted raw raw ones cost bit shipping pounds get raw pounds shipping still fairly reasonable \n",
            "Original summary: these are roasted \n",
            "Predicted summary:  great product\n",
            "\n",
            "\n",
            "Review: drink mostly german coffees smooth strong flavor received coffee gift found pleasing favorites bitter acidic taste strong enough please palate organic free trade coffee one delicious \n",
            "Original summary: no bitter taste \n",
            "Predicted summary:  bitter but not bitter\n",
            "\n",
            "\n",
            "Review: buy pet store web rip sorry \n",
            "Original summary: the best \n",
            "Predicted summary:  great product\n",
            "\n",
            "\n",
            "Review: save lots money using instead flavored coffee creamer grocery store love raspberry italian sodas whipped cream great fresh berry topping chocolate least favorite chocolate snob made decadent raspberry mochas great value small bottle grocery store \n",
            "Original summary: love \n",
            "Predicted summary:  delicious\n",
            "\n",
            "\n",
            "Review: really like suckers taste good one broken whole box happy product gave stars big thought would size versus small size \n",
            "Original summary: great taste \n",
            "Predicted summary:  not as good as\n",
            "\n",
            "\n",
            "Review: always happy newman products coffee exception found priced love fact organic \n",
            "Original summary: coffee \n",
            "Predicted summary:  great coffee\n",
            "\n",
            "\n",
            "Review: product great dogs sensitive eating rawhide great choking getting sick \n",
            "Original summary: love that they are not \n",
            "Predicted summary:  dog treats\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MNpmmTN5huUi",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "outputId": "c7127d23-59f3-4778-db66-666727f238f7"
      },
      "source": [
        "#BLEU Score of Training set\n",
        "#n-gram individual BLEU\n",
        "from nltk.translate.bleu_score import sentence_bleu\n",
        "for i in range(0,1000):\n",
        "  reference = seq2summary(y_train[i])\n",
        "  candidate = decode_sequence(X_train[i].reshape(1, max_text_len))\n",
        "\n",
        "print('Individual 1-gram: %f' % sentence_bleu(reference, candidate, weights=(1, 0, 0, 0)))\n",
        "print('Individual 2-gram: %f' % sentence_bleu(reference, candidate, weights=(0, 1, 0, 0)))\n",
        "print('Individual 3-gram: %f' % sentence_bleu(reference, candidate, weights=(0, 0, 1, 0)))\n",
        "print('Individual 4-gram: %f' % sentence_bleu(reference, candidate, weights=(0, 0, 0, 1)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Individual 1-gram: 0.600000\n",
            "Individual 2-gram: 1.000000\n",
            "Individual 3-gram: 1.000000\n",
            "Individual 4-gram: 1.000000\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cc9hAjePhuQQ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "db76d0c7-03b0-4551-fe4c-95305d296403"
      },
      "source": [
        "#4-gram cumulative BLEU\n",
        "from nltk.translate.bleu_score import sentence_bleu\n",
        "for i in range(0,1000):\n",
        "  reference = seq2summary(y_train[i])\n",
        "  candidate = decode_sequence(X_train[i].reshape(1, max_text_len))\n",
        "\n",
        "score = sentence_bleu(reference, candidate, weights=(0.25, 0.25, 0.25, 0.25))\n",
        "print(score)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.8801117367933934\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JGn7Dm_ehuPO",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "outputId": "4d3e03c4-494a-462b-8c19-b32ae816c533"
      },
      "source": [
        "#cumulative BLEU scores\n",
        "from nltk.translate.bleu_score import sentence_bleu\n",
        "for i in range(0,1000):\n",
        "  reference = seq2summary(y_train[i])\n",
        "  candidate = decode_sequence(X_train[i].reshape(1, max_text_len))\n",
        "\n",
        "print('Cumulative 1-gram: %f' % sentence_bleu(reference, candidate, weights=(1, 0, 0, 0)))\n",
        "print('Cumulative 2-gram: %f' % sentence_bleu(reference, candidate, weights=(0.5, 0.5, 0, 0)))\n",
        "print('Cumulative 3-gram: %f' % sentence_bleu(reference, candidate, weights=(0.33, 0.33, 0.33, 0)))\n",
        "print('Cumulative 4-gram: %f' % sentence_bleu(reference, candidate, weights=(0.25, 0.25, 0.25, 0.25)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cumulative 1-gram: 0.600000\n",
            "Cumulative 2-gram: 0.774597\n",
            "Cumulative 3-gram: 0.844870\n",
            "Cumulative 4-gram: 0.880112\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BdHImnC2ZNWx",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "outputId": "77bdd5ac-0f62-46f6-ebf8-8113a034ad44"
      },
      "source": [
        "#BLEU Score of Test/Validation set\n",
        "#n-gram individual BLEU\n",
        "from nltk.translate.bleu_score import sentence_bleu\n",
        "for i in range(0,1000):\n",
        "  reference = seq2summary(y_test[i])\n",
        "  candidate = decode_sequence(X_test[i].reshape(1, max_text_len))\n",
        "print(\"Test/Validation Set :\")\n",
        "print('Individual 1-gram: %f' % sentence_bleu(reference, candidate, weights=(1, 0, 0, 0)))\n",
        "print('Individual 2-gram: %f' % sentence_bleu(reference, candidate, weights=(0, 1, 0, 0)))\n",
        "print('Individual 3-gram: %f' % sentence_bleu(reference, candidate, weights=(0, 0, 1, 0)))\n",
        "print('Individual 4-gram: %f' % sentence_bleu(reference, candidate, weights=(0, 0, 0, 1)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Test/Validation Set :\n",
            "Individual 1-gram: 0.428571\n",
            "Individual 2-gram: 1.000000\n",
            "Individual 3-gram: 1.000000\n",
            "Individual 4-gram: 1.000000\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fWTKxT4pcj-p",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "552abfeb-91a5-47a3-f872-2d700d31b569"
      },
      "source": [
        "#4-gram cumulative BLEU\n",
        "from nltk.translate.bleu_score import sentence_bleu\n",
        "for i in range(0,1000):\n",
        "  reference = seq2summary(y_test[i])\n",
        "  candidate = decode_sequence(X_test[i].reshape(1, max_text_len))\n",
        "\n",
        "score = sentence_bleu(reference, candidate, weights=(0.25, 0.25, 0.25, 0.25))\n",
        "print(score)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.8091067115702212\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WzsNOQenckIN",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "outputId": "c4caab14-1eb4-4f02-db21-5eacc16d2251"
      },
      "source": [
        "#cumulative BLEU scores\n",
        "from nltk.translate.bleu_score import sentence_bleu\n",
        "for i in range(0,1000):\n",
        "  reference = seq2summary(y_test[i])\n",
        "  candidate = decode_sequence(X_test[i].reshape(1, max_text_len))\n",
        "\n",
        "print('Cumulative 1-gram: %f' % sentence_bleu(reference, candidate, weights=(1, 0, 0, 0)))\n",
        "print('Cumulative 2-gram: %f' % sentence_bleu(reference, candidate, weights=(0.5, 0.5, 0, 0)))\n",
        "print('Cumulative 3-gram: %f' % sentence_bleu(reference, candidate, weights=(0.33, 0.33, 0.33, 0)))\n",
        "print('Cumulative 4-gram: %f' % sentence_bleu(reference, candidate, weights=(0.25, 0.25, 0.25, 0.25)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cumulative 1-gram: 0.428571\n",
            "Cumulative 2-gram: 0.654654\n",
            "Cumulative 3-gram: 0.756080\n",
            "Cumulative 4-gram: 0.809107\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}