Notebooks/Bi_di_GRU_Automatic_Story_Generation.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 5,
  "metadata": {
    "colab": {
      "name": "Bi_di_GRU_Automatic_Story_Generation.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.8"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3878ac84"
      },
      "source": [
        "# **CSE4022: Natural Language Proccessing J Component (Slot:B2)**\n",
        "## **Project Title:** Automatic Story Generation\n",
        "### **Team Members:**\n",
        "1.   Talatala Rahul Reddy  (19BCE1778)\n",
        "2.   Pulimi Bhargava Reddy (19BCE1342)\n",
        "\n",
        "#### In this notebook we have trained the Bi-directional GRU Model on multiple short stories"
      ],
      "id": "3878ac84"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lobZMpVoeGL5"
      },
      "source": [
        "### Importing Libraries & Downloading Dependecies"
      ],
      "id": "lobZMpVoeGL5"
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0c6e038e",
        "outputId": "15cfe834-a6b6-4157-ced1-e4cf6ba37826"
      },
      "source": [
        "#%tensorflow_version 2.x  #selects the tensorflow 2.0+ version\n",
        "import tensorflow as tf\n",
        "import string\n",
        "import requests\n",
        "import nltk\n",
        "import bs4\n",
        "import re\n",
        "import nltk\n",
        "nltk.download('punkt')\n",
        "import numpy as np\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.utils import to_categorical\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense, Dropout, GRU, Embedding, Bidirectional, LSTM\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences"
      ],
      "id": "0c6e038e",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Package punkt is already up-to-date!\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Wre8x-2DeQd1"
      },
      "source": [
        "### Scraping the Short Stories Website http://www.classicshorts.com/ to train the above models"
      ],
      "id": "Wre8x-2DeQd1"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SWWjOq2pm7RM"
      },
      "source": [
        "#return bs4 object\n",
        "def getPage(url,headers):\n",
        "    res = requests.get(url, headers=headers)\n",
        "    try:\n",
        "        res.raise_for_status()\n",
        "    except:\n",
        "        pass\n",
        "    pageHtml=bs4.BeautifulSoup(res.text,'html.parser')\n",
        "    return pageHtml\n",
        "\n",
        "#check if previous scrape already contained the given listing\n",
        "def getStoryText(link):\n",
        "    regex = re.compile(r'[\\n\\r\\t]')\n",
        "    headers = {'User-Agent': 'Mozilla/5.0'}\n",
        "    pageHtml=getPage(link,headers)\n",
        "    elems = pageHtml.find_all(\"div\", class_=\"StoryPara\")\n",
        "    totalText = \"\"\n",
        "    for e in elems:\n",
        "        totalText += regex.sub(\" \", e.text.strip())\n",
        "    return totalText\n",
        "\n",
        "# Returns an array of dcitionaries - turnip listings\n",
        "def getListings():\n",
        "    no_of_stories = 0\n",
        "    badTitles = {'tlrm','fiddler','frog','ItalianMaster','luck'}\n",
        "    #f = open(\"short_stories.txt\", \"a\", encoding='utf-8')\n",
        "    url = \"http://www.classicshorts.com/bib.html\"\n",
        "    headers = {'User-Agent': 'Mozilla/5.0'}\n",
        "    storyTexts = []\n",
        "    rawText = \"\"\n",
        "    pageHtml=getPage(url,headers)\n",
        "    elems = pageHtml.find_all(\"div\", class_=\"biolisting\")\n",
        "    for e in elems:\n",
        "        currentUrl = \"http://www.classicshorts.com/stories/\"+e.attrs['onclick'][11:-2]+\".html\"\n",
        "        if e.attrs['onclick'][11:-2] not in badTitles:\n",
        "            pog = getStoryText(currentUrl)\n",
        "            storyTexts.append(pog)\n",
        "            no_of_stories = no_of_stories + 1\n",
        "            #print(\"No of Stories:\"+str(no_of_stories))\n",
        "            rawText += pog\n",
        "        if no_of_stories == 100:\n",
        "            break\n",
        "        \n",
        "    #print(\"Len of story texts:\"+str(len(storyTexts)))\n",
        "    #print(\"No of stories:\"+str(no_of_stories))\n",
        "    #print(storyTexts)\n",
        "    #f.write(rawText)\n",
        "    #f.close()\n",
        "    return rawText"
      ],
      "id": "SWWjOq2pm7RM",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "69c3f921",
        "outputId": "84af48ec-3ecf-43d5-f937-b43dd7e55ea9"
      },
      "source": [
        "#fetching the short stories by scraping\n",
        "stories = getListings()\n",
        "stories = stories[81:]\n",
        "len(stories)"
      ],
      "id": "69c3f921",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1983654"
            ]
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Hbp8pUELerMd"
      },
      "source": [
        "### Sentence Tokenising the short-stories"
      ],
      "id": "Hbp8pUELerMd"
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6c7ad192",
        "outputId": "1be0d044-b6ab-4ff7-c7aa-470c34142982"
      },
      "source": [
        "#performing sentence tokenisation\n",
        "from nltk.tokenize import sent_tokenize\n",
        "sentences = sent_tokenize(stories)\n",
        "len(sentences)"
      ],
      "id": "6c7ad192",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "20516"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eddd5645",
        "outputId": "f3f553b1-5f85-49b7-95fd-6c5ecda5438b"
      },
      "source": [
        "#displays first 3 sentences\n",
        "sentences[:3]"
      ],
      "id": "eddd5645",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['My Dear Sir: I am happy to assure you that your anxiety is without reason.',\n",
              " 'Flemming will be confined to the sofa for three or four weeks, and will have to be careful at first how he uses his leg.',\n",
              " 'A fracture of this kind is always a tedious affair.']"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Wfdcrf5DexP0"
      },
      "source": [
        "### Tokenising the short stories and cleaning the data"
      ],
      "id": "Wfdcrf5DexP0"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0feece6b"
      },
      "source": [
        "#cleaning the dataset\n",
        "def clean_text(sentences):\n",
        "    #creating tokens from the document\n",
        "    tokens = []\n",
        "    for sentence in sentences:\n",
        "      tokens.extend(nltk.word_tokenize(sentence))\n",
        "    punc = str.maketrans('','',string.punctuation)\n",
        "    #this list will remove punctuations\n",
        "    tokens = [ x.translate(punc) for x in tokens]\n",
        "    #this list will remove special characters\n",
        "    tokens = [ x for x in tokens if x.isalpha()]\n",
        "    #this list converts all the words to lowercase\n",
        "    tokens = [ x.lower() for x in tokens]\n",
        "    return tokens"
      ],
      "id": "0feece6b",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2c6e4842"
      },
      "source": [
        "tokens = clean_text(sentences)"
      ],
      "id": "2c6e4842",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9ff718b7",
        "outputId": "c6d9c47f-f2ea-4cdf-92bb-d5546f0edf99"
      },
      "source": [
        "tokens[:10]"
      ],
      "id": "9ff718b7",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['my', 'dear', 'sir', 'i', 'am', 'happy', 'to', 'assure', 'you', 'that']"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "06cc9ae1",
        "outputId": "cb308cd6-3db2-482a-cbe2-f2b9614b0b88"
      },
      "source": [
        "len(tokens) #no of tokens in the 100 short stories"
      ],
      "id": "06cc9ae1",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "364182"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2aa0e7fc",
        "outputId": "a078f038-14e2-4151-f382-142814ba2dba"
      },
      "source": [
        "len(set(tokens)) #no of unique words in the 100 short stories"
      ],
      "id": "2aa0e7fc",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "21957"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "haUZRZ8oe44G"
      },
      "source": [
        "### Generating sequence of tokens"
      ],
      "id": "haUZRZ8oe44G"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3d8ddc0f"
      },
      "source": [
        "#creating the sequence of tokens\n",
        "length = 50 + 1 #first 50 words as input and 1 as the output\n",
        "lines = []\n",
        "\n",
        "for i in range(length,len(tokens)):\n",
        "    seq = tokens[i-length:i]\n",
        "    line = ' '.join(seq)\n",
        "    lines.append(line)\n",
        "    if i > 110000: #we're only taking few tokens as input for the model\n",
        "        break"
      ],
      "id": "3d8ddc0f",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "40baa158",
        "outputId": "7fe6f72a-42a2-4489-ccb4-9718bd4d2717"
      },
      "source": [
        "len(lines)"
      ],
      "id": "40baa158",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "109951"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "185d88a8",
        "outputId": "b5461cf3-49f2-4496-e19a-055e3358629b"
      },
      "source": [
        "lines[:4]"
      ],
      "id": "185d88a8",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['my dear sir i am happy to assure you that your anxiety is without reason flemming will be confined to the sofa for three or four weeks and will have to be careful at first how he uses his leg a fracture of this kind is always a tedious affair fortunately',\n",
              " 'dear sir i am happy to assure you that your anxiety is without reason flemming will be confined to the sofa for three or four weeks and will have to be careful at first how he uses his leg a fracture of this kind is always a tedious affair fortunately the',\n",
              " 'sir i am happy to assure you that your anxiety is without reason flemming will be confined to the sofa for three or four weeks and will have to be careful at first how he uses his leg a fracture of this kind is always a tedious affair fortunately the bone',\n",
              " 'i am happy to assure you that your anxiety is without reason flemming will be confined to the sofa for three or four weeks and will have to be careful at first how he uses his leg a fracture of this kind is always a tedious affair fortunately the bone was']"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "09e5aa91"
      },
      "source": [
        "### Converting tokens to numerical values"
      ],
      "id": "09e5aa91"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "39993bd9"
      },
      "source": [
        "tokenizer = Tokenizer() \n",
        "tokenizer.fit_on_texts(lines) \n",
        "#performing tokenisation on the seq of words (lines) and these seq of words are embedded as integers basically word embedding\n",
        "#every unique token(word) is given an integer as ML model only works on numerical values\n",
        "sequences = tokenizer.texts_to_sequences(lines)"
      ],
      "id": "39993bd9",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7500e2bb",
        "outputId": "b075cb9a-cbed-440b-85b7-83243270e71c"
      },
      "source": [
        "tokenizer.word_index "
      ],
      "id": "7500e2bb",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'the': 1,\n",
              " 'and': 2,\n",
              " 'of': 3,\n",
              " 'a': 4,\n",
              " 'to': 5,\n",
              " 'he': 6,\n",
              " 'in': 7,\n",
              " 'was': 8,\n",
              " 'i': 9,\n",
              " 'his': 10,\n",
              " 'it': 11,\n",
              " 'that': 12,\n",
              " 'had': 13,\n",
              " 'with': 14,\n",
              " 'you': 15,\n",
              " 'she': 16,\n",
              " 'on': 17,\n",
              " 'at': 18,\n",
              " 'her': 19,\n",
              " 'as': 20,\n",
              " 's': 21,\n",
              " 'for': 22,\n",
              " 'him': 23,\n",
              " 'not': 24,\n",
              " 'but': 25,\n",
              " 'is': 26,\n",
              " 'they': 27,\n",
              " 'from': 28,\n",
              " 'were': 29,\n",
              " 'my': 30,\n",
              " 'said': 31,\n",
              " 'there': 32,\n",
              " 'be': 33,\n",
              " 'all': 34,\n",
              " 'by': 35,\n",
              " 'one': 36,\n",
              " 'have': 37,\n",
              " 'me': 38,\n",
              " 'would': 39,\n",
              " 'nt': 40,\n",
              " 'an': 41,\n",
              " 'no': 42,\n",
              " 'out': 43,\n",
              " 'up': 44,\n",
              " 'this': 45,\n",
              " 'which': 46,\n",
              " 'what': 47,\n",
              " 'when': 48,\n",
              " 'man': 49,\n",
              " 'we': 50,\n",
              " 'so': 51,\n",
              " 'do': 52,\n",
              " 'them': 53,\n",
              " 'then': 54,\n",
              " 'into': 55,\n",
              " 'or': 56,\n",
              " 'their': 57,\n",
              " 'could': 58,\n",
              " 'now': 59,\n",
              " 'if': 60,\n",
              " 'been': 61,\n",
              " 'like': 62,\n",
              " 'are': 63,\n",
              " 'did': 64,\n",
              " 'about': 65,\n",
              " 'down': 66,\n",
              " 'who': 67,\n",
              " 'your': 68,\n",
              " 'some': 69,\n",
              " 'little': 70,\n",
              " 'over': 71,\n",
              " 'only': 72,\n",
              " 'will': 73,\n",
              " 'eyes': 74,\n",
              " 'time': 75,\n",
              " 'more': 76,\n",
              " 'old': 77,\n",
              " 'very': 78,\n",
              " 'back': 79,\n",
              " 'made': 80,\n",
              " 'come': 81,\n",
              " 'again': 82,\n",
              " 'know': 83,\n",
              " 'go': 84,\n",
              " 'himself': 85,\n",
              " 'where': 86,\n",
              " 'thought': 87,\n",
              " 'see': 88,\n",
              " 'before': 89,\n",
              " 'our': 90,\n",
              " 'two': 91,\n",
              " 'looked': 92,\n",
              " 'went': 93,\n",
              " 'hand': 94,\n",
              " 'll': 95,\n",
              " 'can': 96,\n",
              " 'am': 97,\n",
              " 'than': 98,\n",
              " 'came': 99,\n",
              " 'after': 100,\n",
              " 'rainsford': 101,\n",
              " 'first': 102,\n",
              " 'how': 103,\n",
              " 'day': 104,\n",
              " 'has': 105,\n",
              " 'upon': 106,\n",
              " 'well': 107,\n",
              " 'good': 108,\n",
              " 'through': 109,\n",
              " 'its': 110,\n",
              " 'long': 111,\n",
              " 'other': 112,\n",
              " 'father': 113,\n",
              " 'general': 114,\n",
              " 'too': 115,\n",
              " 'any': 116,\n",
              " 'here': 117,\n",
              " 'away': 118,\n",
              " 'still': 119,\n",
              " 'night': 120,\n",
              " 'saw': 121,\n",
              " 'head': 122,\n",
              " 'room': 123,\n",
              " 'think': 124,\n",
              " 'boy': 125,\n",
              " 'face': 126,\n",
              " 'right': 127,\n",
              " 'house': 128,\n",
              " 'say': 129,\n",
              " 'off': 130,\n",
              " 'way': 131,\n",
              " 'without': 132,\n",
              " 'must': 133,\n",
              " 'young': 134,\n",
              " 'get': 135,\n",
              " 'men': 136,\n",
              " 'just': 137,\n",
              " 'got': 138,\n",
              " 'much': 139,\n",
              " 'seemed': 140,\n",
              " 'even': 141,\n",
              " 'put': 142,\n",
              " 'look': 143,\n",
              " 'water': 144,\n",
              " 'these': 145,\n",
              " 'mr': 146,\n",
              " 'never': 147,\n",
              " 'something': 148,\n",
              " 'baron': 149,\n",
              " 'heard': 150,\n",
              " 'life': 151,\n",
              " 'took': 152,\n",
              " 'great': 153,\n",
              " 'make': 154,\n",
              " 'should': 155,\n",
              " 'turned': 156,\n",
              " 'us': 157,\n",
              " 'last': 158,\n",
              " 'take': 159,\n",
              " 'door': 160,\n",
              " 'paul': 161,\n",
              " 'nothing': 162,\n",
              " 'knew': 163,\n",
              " 'every': 164,\n",
              " 'moment': 165,\n",
              " 'under': 166,\n",
              " 'sat': 167,\n",
              " 'asked': 168,\n",
              " 'left': 169,\n",
              " 'going': 170,\n",
              " 'hands': 171,\n",
              " 'might': 172,\n",
              " 'home': 173,\n",
              " 'place': 174,\n",
              " 'looking': 175,\n",
              " 'began': 176,\n",
              " 'always': 177,\n",
              " 've': 178,\n",
              " 'mother': 179,\n",
              " 'few': 180,\n",
              " 'because': 181,\n",
              " 'why': 182,\n",
              " 'mrs': 183,\n",
              " 'such': 184,\n",
              " 'new': 185,\n",
              " 'another': 186,\n",
              " 'once': 187,\n",
              " 'three': 188,\n",
              " 'own': 189,\n",
              " 'felt': 190,\n",
              " 'table': 191,\n",
              " 'many': 192,\n",
              " 'small': 193,\n",
              " 'd': 194,\n",
              " 'while': 195,\n",
              " 'marshall': 196,\n",
              " 'against': 197,\n",
              " 'being': 198,\n",
              " 'yes': 199,\n",
              " 'white': 200,\n",
              " 'dark': 201,\n",
              " 'm': 202,\n",
              " 'found': 203,\n",
              " 'side': 204,\n",
              " 'bed': 205,\n",
              " 'perhaps': 206,\n",
              " 'tell': 207,\n",
              " 'light': 208,\n",
              " 'people': 209,\n",
              " 'may': 210,\n",
              " 'told': 211,\n",
              " 'feet': 212,\n",
              " 'same': 213,\n",
              " 'thing': 214,\n",
              " 'street': 215,\n",
              " 'air': 216,\n",
              " 'most': 217,\n",
              " 'lay': 218,\n",
              " 'black': 219,\n",
              " 'seen': 220,\n",
              " 'gave': 221,\n",
              " 'road': 222,\n",
              " 'years': 223,\n",
              " 'stood': 224,\n",
              " 'world': 225,\n",
              " 'those': 226,\n",
              " 'better': 227,\n",
              " 'shall': 228,\n",
              " 'though': 229,\n",
              " 'behind': 230,\n",
              " 'mind': 231,\n",
              " 'morning': 232,\n",
              " 'things': 233,\n",
              " 'each': 234,\n",
              " 'let': 235,\n",
              " 'enough': 236,\n",
              " 'give': 237,\n",
              " 'find': 238,\n",
              " 'window': 239,\n",
              " 'evening': 240,\n",
              " 'trees': 241,\n",
              " 'quite': 242,\n",
              " 'until': 243,\n",
              " 'along': 244,\n",
              " 'camp': 245,\n",
              " 'brennen': 246,\n",
              " 'woman': 247,\n",
              " 'between': 248,\n",
              " 'sound': 249,\n",
              " 'family': 250,\n",
              " 'fell': 251,\n",
              " 'high': 252,\n",
              " 'around': 253,\n",
              " 'both': 254,\n",
              " 'ever': 255,\n",
              " 'almost': 256,\n",
              " 'red': 257,\n",
              " 'whole': 258,\n",
              " 'feeling': 259,\n",
              " 'voice': 260,\n",
              " 'big': 261,\n",
              " 'duvitch': 262,\n",
              " 'anything': 263,\n",
              " 'far': 264,\n",
              " 'heart': 265,\n",
              " 'cold': 266,\n",
              " 'round': 267,\n",
              " 'next': 268,\n",
              " 'fear': 269,\n",
              " 'held': 270,\n",
              " 'wife': 271,\n",
              " 'bet': 272,\n",
              " 'fine': 273,\n",
              " 'coming': 274,\n",
              " 'taken': 275,\n",
              " 'having': 276,\n",
              " 'want': 277,\n",
              " 'later': 278,\n",
              " 'children': 279,\n",
              " 'end': 280,\n",
              " 'john': 281,\n",
              " 'together': 282,\n",
              " 'sun': 283,\n",
              " 'yet': 284,\n",
              " 'days': 285,\n",
              " 'called': 286,\n",
              " 're': 287,\n",
              " 'tom': 288,\n",
              " 'miss': 289,\n",
              " 'dead': 290,\n",
              " 'car': 291,\n",
              " 'francis': 292,\n",
              " 'brought': 293,\n",
              " 'across': 294,\n",
              " 'smile': 295,\n",
              " 'half': 296,\n",
              " 'girl': 297,\n",
              " 'year': 298,\n",
              " 'rather': 299,\n",
              " 'four': 300,\n",
              " 'used': 301,\n",
              " 'feel': 302,\n",
              " 'lost': 303,\n",
              " 'everything': 304,\n",
              " 'walked': 305,\n",
              " 'duvitches': 306,\n",
              " 'passed': 307,\n",
              " 'child': 308,\n",
              " 'dear': 309,\n",
              " 'figure': 310,\n",
              " 'stopped': 311,\n",
              " 'god': 312,\n",
              " 'remembered': 313,\n",
              " 'wanted': 314,\n",
              " 'open': 315,\n",
              " 'hanley': 316,\n",
              " 'dick': 317,\n",
              " 'kind': 318,\n",
              " 'sometimes': 319,\n",
              " 'large': 320,\n",
              " 'daw': 321,\n",
              " 'opened': 322,\n",
              " 'slowly': 323,\n",
              " 'school': 324,\n",
              " 'boys': 325,\n",
              " 'longer': 326,\n",
              " 'ca': 327,\n",
              " 'town': 328,\n",
              " 'sit': 329,\n",
              " 'whom': 330,\n",
              " 'fire': 331,\n",
              " 'nor': 332,\n",
              " 'returned': 333,\n",
              " 'suddenly': 334,\n",
              " 'terenty': 335,\n",
              " 'peter': 336,\n",
              " 'senator': 337,\n",
              " 'sleep': 338,\n",
              " 'part': 339,\n",
              " 'hair': 340,\n",
              " 'silence': 341,\n",
              " 'second': 342,\n",
              " 'myself': 343,\n",
              " 'country': 344,\n",
              " 'de': 345,\n",
              " 'olenka': 346,\n",
              " 'nurse': 347,\n",
              " 'near': 348,\n",
              " 'times': 349,\n",
              " 'hour': 350,\n",
              " 'course': 351,\n",
              " 'soon': 352,\n",
              " 'soldier': 353,\n",
              " 'followed': 354,\n",
              " 'cried': 355,\n",
              " 'call': 356,\n",
              " 'lights': 357,\n",
              " 'above': 358,\n",
              " 'bar': 359,\n",
              " 'really': 360,\n",
              " 'love': 361,\n",
              " 'herself': 362,\n",
              " 'hundred': 363,\n",
              " 'alone': 364,\n",
              " 'rose': 365,\n",
              " 'gone': 366,\n",
              " 'caught': 367,\n",
              " 'toward': 368,\n",
              " 'death': 369,\n",
              " 'johnny': 370,\n",
              " 'leg': 371,\n",
              " 'friend': 372,\n",
              " 'sea': 373,\n",
              " 'full': 374,\n",
              " 'name': 375,\n",
              " 'sort': 376,\n",
              " 'talk': 377,\n",
              " 'rain': 378,\n",
              " 'lips': 379,\n",
              " 'glass': 380,\n",
              " 'afraid': 381,\n",
              " 'bank': 382,\n",
              " 'hunting': 383,\n",
              " 'does': 384,\n",
              " 'blue': 385,\n",
              " 'leave': 386,\n",
              " 'legs': 387,\n",
              " 'brother': 388,\n",
              " 'known': 389,\n",
              " 'interest': 390,\n",
              " 'already': 391,\n",
              " 'among': 392,\n",
              " 'sitting': 393,\n",
              " 'garden': 394,\n",
              " 'store': 395,\n",
              " 'whether': 396,\n",
              " 'lived': 397,\n",
              " 'fish': 398,\n",
              " 'oclock': 399,\n",
              " 'set': 400,\n",
              " 'beyond': 401,\n",
              " 'best': 402,\n",
              " 'front': 403,\n",
              " 'ran': 404,\n",
              " 'hard': 405,\n",
              " 'friends': 406,\n",
              " 'play': 407,\n",
              " 'certainly': 408,\n",
              " 'kept': 409,\n",
              " 'reached': 410,\n",
              " 'oh': 411,\n",
              " 'drink': 412,\n",
              " 'chair': 413,\n",
              " 'finger': 414,\n",
              " 'standing': 415,\n",
              " 'ready': 416,\n",
              " 'read': 417,\n",
              " 'turn': 418,\n",
              " 'moved': 419,\n",
              " 'hot': 420,\n",
              " 'hear': 421,\n",
              " 'party': 422,\n",
              " 'stay': 423,\n",
              " 'carried': 424,\n",
              " 'tried': 425,\n",
              " 'smiled': 426,\n",
              " 'horse': 427,\n",
              " 'luck': 428,\n",
              " 'beside': 429,\n",
              " 'fall': 430,\n",
              " 'bad': 431,\n",
              " 'afternoon': 432,\n",
              " 'five': 433,\n",
              " 'livingstone': 434,\n",
              " 'wild': 435,\n",
              " 'ten': 436,\n",
              " 'word': 437,\n",
              " 'women': 438,\n",
              " 'walk': 439,\n",
              " 'holding': 440,\n",
              " 'given': 441,\n",
              " 'shook': 442,\n",
              " 'sky': 443,\n",
              " 'body': 444,\n",
              " 'tree': 445,\n",
              " 'grew': 446,\n",
              " 'added': 447,\n",
              " 'george': 448,\n",
              " 'stumpy': 449,\n",
              " 'poor': 450,\n",
              " 'money': 451,\n",
              " 'jack': 452,\n",
              " 'colonel': 453,\n",
              " 'thinking': 454,\n",
              " 'short': 455,\n",
              " 'became': 456,\n",
              " 'words': 457,\n",
              " 'done': 458,\n",
              " 'doctor': 459,\n",
              " 'watching': 460,\n",
              " 'sign': 461,\n",
              " 'please': 462,\n",
              " 'roaring': 463,\n",
              " 'kitchen': 464,\n",
              " 'consul': 465,\n",
              " 'sure': 466,\n",
              " 'wish': 467,\n",
              " 'letter': 468,\n",
              " 'since': 469,\n",
              " 'banker': 470,\n",
              " 'met': 471,\n",
              " 'game': 472,\n",
              " 'answered': 473,\n",
              " 'show': 474,\n",
              " 'closed': 475,\n",
              " 'quickly': 476,\n",
              " 'need': 477,\n",
              " 'darling': 478,\n",
              " 'ivan': 479,\n",
              " 'uncle': 480,\n",
              " 'state': 481,\n",
              " 'fellow': 482,\n",
              " 'story': 483,\n",
              " 'minutes': 484,\n",
              " 'tea': 485,\n",
              " 'possible': 486,\n",
              " 'forward': 487,\n",
              " 'believe': 488,\n",
              " 'understand': 489,\n",
              " 'business': 490,\n",
              " 'within': 491,\n",
              " 'help': 492,\n",
              " 'whose': 493,\n",
              " 'yellow': 494,\n",
              " 'lying': 495,\n",
              " 'panther': 496,\n",
              " 'wo': 497,\n",
              " 'cry': 498,\n",
              " 'wall': 499,\n",
              " 'outside': 500,\n",
              " 'river': 501,\n",
              " 'empty': 502,\n",
              " 'doing': 503,\n",
              " 'tears': 504,\n",
              " 'books': 505,\n",
              " 'case': 506,\n",
              " 'piece': 507,\n",
              " 'taking': 508,\n",
              " 'keep': 509,\n",
              " 'sent': 510,\n",
              " 'deep': 511,\n",
              " 'several': 512,\n",
              " 'animal': 513,\n",
              " 'office': 514,\n",
              " 'towards': 515,\n",
              " 'continued': 516,\n",
              " 'cut': 517,\n",
              " 'shot': 518,\n",
              " 'hope': 519,\n",
              " 'making': 520,\n",
              " 'watched': 521,\n",
              " 'hannah': 522,\n",
              " 'supper': 523,\n",
              " 'reason': 524,\n",
              " 'certain': 525,\n",
              " 'hours': 526,\n",
              " 'else': 527,\n",
              " 'miles': 528,\n",
              " 'wind': 529,\n",
              " 'dinner': 530,\n",
              " 'late': 531,\n",
              " 'fingers': 532,\n",
              " 'island': 533,\n",
              " 'sense': 534,\n",
              " 'matter': 535,\n",
              " 'dog': 536,\n",
              " 'asleep': 537,\n",
              " 'straight': 538,\n",
              " 'live': 539,\n",
              " 'hotel': 540,\n",
              " 'soft': 541,\n",
              " 'warm': 542,\n",
              " 'neck': 543,\n",
              " 'husband': 544,\n",
              " 'saying': 545,\n",
              " 'six': 546,\n",
              " 'remember': 547,\n",
              " 'floor': 548,\n",
              " 'eat': 549,\n",
              " 'manner': 550,\n",
              " 'line': 551,\n",
              " 'green': 552,\n",
              " 'suppose': 553,\n",
              " 'rest': 554,\n",
              " 'past': 555,\n",
              " 'struck': 556,\n",
              " 'strange': 557,\n",
              " 'run': 558,\n",
              " 'walking': 559,\n",
              " 'wait': 560,\n",
              " 'top': 561,\n",
              " 'mouth': 562,\n",
              " 'knife': 563,\n",
              " 'either': 564,\n",
              " 'low': 565,\n",
              " 'bridge': 566,\n",
              " 'happy': 567,\n",
              " 'silent': 568,\n",
              " 'themselves': 569,\n",
              " 'fact': 570,\n",
              " 'company': 571,\n",
              " 'seeing': 572,\n",
              " 'sorry': 573,\n",
              " 'beautiful': 574,\n",
              " 'point': 575,\n",
              " 'doubt': 576,\n",
              " 'hold': 577,\n",
              " 'ground': 578,\n",
              " 'teeth': 579,\n",
              " 'work': 580,\n",
              " 'darkness': 581,\n",
              " 'also': 582,\n",
              " 'beginning': 583,\n",
              " 'pulled': 584,\n",
              " 'says': 585,\n",
              " 'hardly': 586,\n",
              " 'sharp': 587,\n",
              " 'replied': 588,\n",
              " 'touched': 589,\n",
              " 'wine': 590,\n",
              " 'zaroff': 591,\n",
              " 'admiral': 592,\n",
              " 'pleasure': 593,\n",
              " 'least': 594,\n",
              " 'present': 595,\n",
              " 'character': 596,\n",
              " 'wide': 597,\n",
              " 'move': 598,\n",
              " 'dust': 599,\n",
              " 'idea': 600,\n",
              " 'gentleman': 601,\n",
              " 'pretty': 602,\n",
              " 'host': 603,\n",
              " 'o': 604,\n",
              " 'tall': 605,\n",
              " 'son': 606,\n",
              " 'bright': 607,\n",
              " 'easy': 608,\n",
              " 'spoke': 609,\n",
              " 'talking': 610,\n",
              " 'earth': 611,\n",
              " 'speak': 612,\n",
              " 'leaving': 613,\n",
              " 'gate': 614,\n",
              " 'shoulder': 615,\n",
              " 'threw': 616,\n",
              " 'arms': 617,\n",
              " 'soldiers': 618,\n",
              " 'showed': 619,\n",
              " 'noise': 620,\n",
              " 'steps': 621,\n",
              " 'less': 622,\n",
              " 'answer': 623,\n",
              " 'food': 624,\n",
              " 'repeated': 625,\n",
              " 'hunt': 626,\n",
              " 'american': 627,\n",
              " 'von': 628,\n",
              " 'castle': 629,\n",
              " 'tennessee': 630,\n",
              " 'fallen': 631,\n",
              " 'become': 632,\n",
              " 'summer': 633,\n",
              " 'delaney': 634,\n",
              " 'houses': 635,\n",
              " 'bring': 636,\n",
              " 'broken': 637,\n",
              " 'often': 638,\n",
              " 'mean': 639,\n",
              " 'marjorie': 640,\n",
              " 'pay': 641,\n",
              " 'distance': 642,\n",
              " 'strong': 643,\n",
              " 'eye': 644,\n",
              " 'moving': 645,\n",
              " 'flowers': 646,\n",
              " 'stand': 647,\n",
              " 'trouble': 648,\n",
              " 'bottle': 649,\n",
              " 'fresh': 650,\n",
              " 'clouds': 651,\n",
              " 'music': 652,\n",
              " 'sudden': 653,\n",
              " 'stream': 654,\n",
              " 'foot': 655,\n",
              " 'captain': 656,\n",
              " 'dodd': 657,\n",
              " 'laughed': 658,\n",
              " 'english': 659,\n",
              " 'number': 660,\n",
              " 'yacht': 661,\n",
              " 'lighter': 662,\n",
              " 'ought': 663,\n",
              " 'prisoner': 664,\n",
              " 'soul': 665,\n",
              " 'except': 666,\n",
              " 'write': 667,\n",
              " 'arm': 668,\n",
              " 'shoulders': 669,\n",
              " 'windows': 670,\n",
              " 'send': 671,\n",
              " 'bear': 672,\n",
              " 'rich': 673,\n",
              " 'winter': 674,\n",
              " 'nice': 675,\n",
              " 'order': 676,\n",
              " 'gold': 677,\n",
              " 'laugh': 678,\n",
              " 'plain': 679,\n",
              " 'means': 680,\n",
              " 'during': 681,\n",
              " 'stop': 682,\n",
              " 'led': 683,\n",
              " 'ah': 684,\n",
              " 'conscious': 685,\n",
              " 'fool': 686,\n",
              " 'below': 687,\n",
              " 'chest': 688,\n",
              " 'brain': 689,\n",
              " 'creek': 690,\n",
              " 'smiling': 691,\n",
              " 'tired': 692,\n",
              " 'begin': 693,\n",
              " 'happened': 694,\n",
              " 'explained': 695,\n",
              " 'board': 696,\n",
              " 'hall': 697,\n",
              " 'married': 698,\n",
              " 'bit': 699,\n",
              " 'hennefalcon': 700,\n",
              " 'ask': 701,\n",
              " 'flemming': 702,\n",
              " 'care': 703,\n",
              " 'pines': 704,\n",
              " 'breath': 705,\n",
              " 'nearly': 706,\n",
              " 'ill': 707,\n",
              " 'thousand': 708,\n",
              " 'silver': 709,\n",
              " 'able': 710,\n",
              " 'attention': 711,\n",
              " 'sick': 712,\n",
              " 'fair': 713,\n",
              " 'pass': 714,\n",
              " 'lady': 715,\n",
              " 'hat': 716,\n",
              " 'cigar': 717,\n",
              " 'pleasant': 718,\n",
              " 'moments': 719,\n",
              " 'dropped': 720,\n",
              " 'real': 721,\n",
              " 'service': 722,\n",
              " 'presence': 723,\n",
              " 'palm': 724,\n",
              " 'hill': 725,\n",
              " 'human': 726,\n",
              " 'further': 727,\n",
              " 'ears': 728,\n",
              " 'lifted': 729,\n",
              " 'gray': 730,\n",
              " 'tomorrow': 731,\n",
              " 'thank': 732,\n",
              " 'perfect': 733,\n",
              " 'started': 734,\n",
              " 'dressed': 735,\n",
              " 'lodge': 736,\n",
              " 'loved': 737,\n",
              " 'yard': 738,\n",
              " 'fyokla': 739,\n",
              " 'pond': 740,\n",
              " 'surgeon': 741,\n",
              " 'natural': 742,\n",
              " 'follow': 743,\n",
              " 'living': 744,\n",
              " 'lie': 745,\n",
              " 'weather': 746,\n",
              " 'ago': 747,\n",
              " 'week': 748,\n",
              " 'daughter': 749,\n",
              " 'killed': 750,\n",
              " 'church': 751,\n",
              " 'indeed': 752,\n",
              " 'pale': 753,\n",
              " 'listening': 754,\n",
              " 'true': 755,\n",
              " 'habit': 756,\n",
              " 'lot': 757,\n",
              " 'wrong': 758,\n",
              " 'local': 759,\n",
              " 'laid': 760,\n",
              " 'officer': 761,\n",
              " 'horses': 762,\n",
              " 'use': 763,\n",
              " 'free': 764,\n",
              " 'strength': 765,\n",
              " 'terrible': 766,\n",
              " 'memory': 767,\n",
              " 'enemy': 768,\n",
              " 'form': 769,\n",
              " 'waited': 770,\n",
              " 'raised': 771,\n",
              " 'finished': 772,\n",
              " 'wondered': 773,\n",
              " 'trying': 774,\n",
              " 'tonight': 775,\n",
              " 'others': 776,\n",
              " 'valley': 777,\n",
              " 'edge': 778,\n",
              " 'quietly': 779,\n",
              " 'sergeant': 780,\n",
              " 'train': 781,\n",
              " 'clothes': 782,\n",
              " 'stories': 783,\n",
              " 'ahineev': 784,\n",
              " 'vane': 785,\n",
              " 'peaches': 786,\n",
              " 'swamp': 787,\n",
              " 'porto': 788,\n",
              " 'koëldwethout': 789,\n",
              " 'pain': 790,\n",
              " 'change': 791,\n",
              " 'impossible': 792,\n",
              " 'knows': 793,\n",
              " 'rooms': 794,\n",
              " 'girls': 795,\n",
              " 'dress': 796,\n",
              " 'till': 797,\n",
              " 'staring': 798,\n",
              " 'months': 799,\n",
              " 'expect': 800,\n",
              " 'mine': 801,\n",
              " 'war': 802,\n",
              " 'remained': 803,\n",
              " 'nature': 804,\n",
              " 'noticed': 805,\n",
              " 'visit': 806,\n",
              " 'return': 807,\n",
              " 'heavy': 808,\n",
              " 'seat': 809,\n",
              " 'paper': 810,\n",
              " 'probably': 811,\n",
              " 'notice': 812,\n",
              " 'entirely': 813,\n",
              " 'following': 814,\n",
              " 'knees': 815,\n",
              " 'direction': 816,\n",
              " 'desert': 817,\n",
              " 'sight': 818,\n",
              " 'land': 819,\n",
              " 'faces': 820,\n",
              " 'middle': 821,\n",
              " 'terror': 822,\n",
              " 'blood': 823,\n",
              " 'pointed': 824,\n",
              " 'running': 825,\n",
              " 'quiet': 826,\n",
              " 'shut': 827,\n",
              " 'observed': 828,\n",
              " 'learned': 829,\n",
              " 'bare': 830,\n",
              " 'coat': 831,\n",
              " 'wood': 832,\n",
              " 'inside': 833,\n",
              " 'fast': 834,\n",
              " 'exclaimed': 835,\n",
              " 'snow': 836,\n",
              " 'guests': 837,\n",
              " 'millions': 838,\n",
              " 'nails': 839,\n",
              " 'lincoln': 840,\n",
              " 'banos': 841,\n",
              " 'baroness': 842,\n",
              " 'christiana': 843,\n",
              " 'simpson': 844,\n",
              " 'jovita': 845,\n",
              " 'sir': 846,\n",
              " 'break': 847,\n",
              " 'familiar': 848,\n",
              " 'touch': 849,\n",
              " 'drew': 850,\n",
              " 'beauty': 851,\n",
              " 'weak': 852,\n",
              " 'talked': 853,\n",
              " 'played': 854,\n",
              " 'afterwards': 855,\n",
              " 'glance': 856,\n",
              " 'remarked': 857,\n",
              " 'desire': 858,\n",
              " 'usual': 859,\n",
              " 'faint': 860,\n",
              " 'itself': 861,\n",
              " 'giving': 862,\n",
              " 'york': 863,\n",
              " 'duty': 864,\n",
              " 'question': 865,\n",
              " 'stepped': 866,\n",
              " 'entered': 867,\n",
              " 'expression': 868,\n",
              " 'thin': 869,\n",
              " 'minute': 870,\n",
              " 'close': 871,\n",
              " 'filled': 872,\n",
              " 'watch': 873,\n",
              " 'covered': 874,\n",
              " 'playing': 875,\n",
              " 'corner': 876,\n",
              " 'someone': 877,\n",
              " 'none': 878,\n",
              " 'twice': 879,\n",
              " 'wore': 880,\n",
              " 'nodded': 881,\n",
              " 'birds': 882,\n",
              " 'thick': 883,\n",
              " 'soap': 884,\n",
              " 'won': 885,\n",
              " 'ship': 886,\n",
              " 'bass': 887,\n",
              " 'veterinary': 888,\n",
              " 'dmitritch': 889,\n",
              " 'president': 890,\n",
              " 'trail': 891,\n",
              " 'maloney': 892,\n",
              " 'bullen': 893,\n",
              " 'kentuck': 894,\n",
              " 'appeared': 895,\n",
              " 'hurt': 896,\n",
              " 'suit': 897,\n",
              " 'nobody': 898,\n",
              " 'nose': 899,\n",
              " 'fancy': 900,\n",
              " 'somehow': 901,\n",
              " 'glad': 902,\n",
              " 'evil': 903,\n",
              " 'city': 904,\n",
              " 'twelve': 905,\n",
              " 'washington': 906,\n",
              " 'surprise': 907,\n",
              " 'lieutenant': 908,\n",
              " 'smoke': 909,\n",
              " 'getting': 910,\n",
              " 'sister': 911,\n",
              " 'cook': 912,\n",
              " 'conversation': 913,\n",
              " 'sunday': 914,\n",
              " 'however': 915,\n",
              " 'person': 916,\n",
              " 'seem': 917,\n",
              " 'awake': 918,\n",
              " 'neither': 919,\n",
              " 'leaves': 920,\n",
              " 'peace': 921,\n",
              " 'position': 922,\n",
              " 'drove': 923,\n",
              " 'try': 924,\n",
              " 'opinion': 925,\n",
              " 'dream': 926,\n",
              " 'explain': 927,\n",
              " 'worth': 928,\n",
              " 'heads': 929,\n",
              " 'died': 930,\n",
              " 'slept': 931,\n",
              " 'wet': 932,\n",
              " 'movement': 933,\n",
              " 'suggested': 934,\n",
              " 'anyone': 935,\n",
              " 'single': 936,\n",
              " 'grass': 937,\n",
              " 'chance': 938,\n",
              " 'glanced': 939,\n",
              " 'today': 940,\n",
              " 'effort': 941,\n",
              " 'power': 942,\n",
              " 'abroad': 943,\n",
              " 'downward': 944,\n",
              " 'bottom': 945,\n",
              " 'comes': 946,\n",
              " 'quick': 947,\n",
              " 'wheel': 948,\n",
              " 'shore': 949,\n",
              " 'nearer': 950,\n",
              " 'ahead': 951,\n",
              " 'anyway': 952,\n",
              " 'somewhere': 953,\n",
              " 'wants': 954,\n",
              " 'pack': 955,\n",
              " 'christmas': 956,\n",
              " 'helped': 957,\n",
              " 'worse': 958,\n",
              " 'fifteen': 959,\n",
              " 'forty': 960,\n",
              " 'rubbing': 961,\n",
              " 'vankin': 962,\n",
              " 'pipe': 963,\n",
              " 'jungle': 964,\n",
              " 'cadillac': 965,\n",
              " 'cairns': 966,\n",
              " 'partner': 967,\n",
              " 'yer': 968,\n",
              " 'weeks': 969,\n",
              " 'evidently': 970,\n",
              " 'deal': 971,\n",
              " 'object': 972,\n",
              " 'throw': 973,\n",
              " 'watkins': 974,\n",
              " 'imagine': 975,\n",
              " 'month': 976,\n",
              " 'health': 977,\n",
              " 'forest': 978,\n",
              " 'opposite': 979,\n",
              " 'roof': 980,\n",
              " 'golden': 981,\n",
              " 'fashion': 982,\n",
              " 'drop': 983,\n",
              " 'drawn': 984,\n",
              " 'partly': 985,\n",
              " 'eight': 986,\n",
              " 'dozen': 987,\n",
              " 'neighbors': 988,\n",
              " 'meet': 989,\n",
              " 'picture': 990,\n",
              " 'changed': 991,\n",
              " 'group': 992,\n",
              " 'huge': 993,\n",
              " 'grown': 994,\n",
              " 'account': 995,\n",
              " 'written': 996,\n",
              " 'dreamed': 997,\n",
              " 'friendly': 998,\n",
              " 'greater': 999,\n",
              " 'reply': 1000,\n",
              " ...}"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "74be593b"
      },
      "source": [
        "sequences = np.array(sequences)"
      ],
      "id": "74be593b",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e3d05ca0",
        "outputId": "96c0f540-1be1-4011-bbc7-169eb42e4cef"
      },
      "source": [
        "sequences[:2]\n",
        "#it is a 2d array\n",
        "#rows represent the line of the short stories\n",
        "#column represent the words (there are 51 cols)"
      ],
      "id": "e3d05ca0",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[   30,   309,   846,     9,    97,   567,     5,  1890,    15,\n",
              "           12,    68,  3705,    26,   132,   524,   702,    73,    33,\n",
              "        11636,     5,     1,  2266,    22,   188,    56,   300,   969,\n",
              "            2,    73,    37,     5,    33,  5505,    18,   102,   103,\n",
              "            6, 11633,    10,   371,     4, 11630,     3,    45,   318,\n",
              "           26,   177,     4,  5504,  3703,  2805],\n",
              "       [  309,   846,     9,    97,   567,     5,  1890,    15,    12,\n",
              "           68,  3705,    26,   132,   524,   702,    73,    33, 11636,\n",
              "            5,     1,  2266,    22,   188,    56,   300,   969,     2,\n",
              "           73,    37,     5,    33,  5505,    18,   102,   103,     6,\n",
              "        11633,    10,   371,     4, 11630,     3,    45,   318,    26,\n",
              "          177,     4,  5504,  3703,  2805,     1]])"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "b50d2c9b",
        "outputId": "b0ef6e3a-f835-4e7b-e110-c1053db30d69"
      },
      "source": [
        "sequences.shape #51 cols are words where we take first 50 cols(words) as x and last col(word) as y"
      ],
      "id": "b50d2c9b",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(109951, 51)"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6v4T2YpDfQEr"
      },
      "source": [
        "### Preparing data to that has to fed into the model"
      ],
      "id": "6v4T2YpDfQEr"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6b96ea34"
      },
      "source": [
        "#first 50 tokens(words) are x and last tokens is y\n",
        "X,y = sequences[:,:-1] , sequences[:,-1]\n",
        "#sequences[:,-1] selects only the last col\n",
        "#sequences[:,:-1] selects all cols except the last col"
      ],
      "id": "6b96ea34",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cbf98e7d",
        "outputId": "25d0a864-fbd6-4905-87e3-341fcd685cbe"
      },
      "source": [
        "X[:2]"
      ],
      "id": "cbf98e7d",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[   30,   309,   846,     9,    97,   567,     5,  1890,    15,\n",
              "           12,    68,  3705,    26,   132,   524,   702,    73,    33,\n",
              "        11636,     5,     1,  2266,    22,   188,    56,   300,   969,\n",
              "            2,    73,    37,     5,    33,  5505,    18,   102,   103,\n",
              "            6, 11633,    10,   371,     4, 11630,     3,    45,   318,\n",
              "           26,   177,     4,  5504,  3703],\n",
              "       [  309,   846,     9,    97,   567,     5,  1890,    15,    12,\n",
              "           68,  3705,    26,   132,   524,   702,    73,    33, 11636,\n",
              "            5,     1,  2266,    22,   188,    56,   300,   969,     2,\n",
              "           73,    37,     5,    33,  5505,    18,   102,   103,     6,\n",
              "        11633,    10,   371,     4, 11630,     3,    45,   318,    26,\n",
              "          177,     4,  5504,  3703,  2805]])"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "63e6944a",
        "outputId": "11ee206f-7a62-426b-eba6-e199623b00ad"
      },
      "source": [
        "y[:2]"
      ],
      "id": "63e6944a",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([2805,    1])"
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "44f3287b",
        "outputId": "0fed7699-835c-4496-cfe7-3c375e316f28"
      },
      "source": [
        "vocabulary_size = len(tokenizer.word_index) + 1 #1 as it's stating from 1 (that word_index is from 1)\n",
        "vocabulary_size"
      ],
      "id": "44f3287b",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "11638"
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b1dd2241"
      },
      "source": [
        "y = to_categorical(y,num_classes=vocabulary_size)\n",
        "# a numpy array/vector which has integers that represent different categories is converted into a numpy array/matrix \n",
        "#which has binary values and has columns equal to the number of categories in the data.\n",
        "#X.shape[1] is 50\n",
        "seq_length = X.shape[1]"
      ],
      "id": "b1dd2241",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "n1ZkhB7ygVR2"
      },
      "source": [
        "### Creating a function to generate stories from a given seed text"
      ],
      "id": "n1ZkhB7ygVR2"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QCGE-VaE4eR7"
      },
      "source": [
        "def generate_story(model,tokenizer,text_seq_len,seed_text,n_words):\n",
        "  text = []\n",
        "  for _ in range(n_words):\n",
        "    encoded = tokenizer.texts_to_sequences([seed_text])[0]\n",
        "    #the model expects the input(sentences) to be of same length hence we use the pad_sequences\n",
        "    #pre it adds 0 before the sentence\n",
        "    encoded = pad_sequences([encoded],maxlen=text_seq_len,padding='pre')\n",
        "\n",
        "    pred = model.predict(encoded) \n",
        "    y_pred = np.argmax(pred,axis=1)\n",
        "    predicted_word = ''\n",
        "    for word,index in tokenizer.word_index.items():\n",
        "      if index == y_pred:\n",
        "        predicted_word = word\n",
        "        break\n",
        "    seed_text = seed_text + ' ' + predicted_word\n",
        "    text.append(predicted_word)\n",
        "  return ' '.join(text)"
      ],
      "id": "QCGE-VaE4eR7",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_N8I83obhKPC"
      },
      "source": [
        "### Training the Bi-directional GRU Model"
      ],
      "id": "_N8I83obhKPC"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PeRXjaB2hPLm"
      },
      "source": [
        "bi_di_gru = Sequential()\n",
        "#input layer\n",
        "bi_di_gru.add(Embedding(vocabulary_size,50,input_length=seq_length))\n",
        "#hidden layer 1 (LSTM layer)\n",
        "bi_di_gru.add(Bidirectional(GRU(100,return_sequences=True)))\n",
        "bi_di_gru.add(Dropout(0.2))\n",
        "#hidden layer 2 (LSTM layer)\n",
        "bi_di_gru.add(GRU(120))\n",
        "#hidden layer 3 (dense layer)\n",
        "bi_di_gru.add(Dense(140,activation='relu'))\n",
        "#output layer\n",
        "bi_di_gru.add(Dense(vocabulary_size,activation='softmax'))#units=vocab size bcoz we want probability for each predicted word"
      ],
      "id": "PeRXjaB2hPLm",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WfeK6-5OhPUc",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a48dcbcc-a8f3-48ea-ebf6-0085c37f1e24"
      },
      "source": [
        "bi_di_gru.summary()"
      ],
      "id": "WfeK6-5OhPUc",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " embedding (Embedding)       (None, 50, 50)            581900    \n",
            "                                                                 \n",
            " bidirectional (Bidirectiona  (None, 50, 200)          91200     \n",
            " l)                                                              \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 50, 200)           0         \n",
            "                                                                 \n",
            " gru_1 (GRU)                 (None, 120)               115920    \n",
            "                                                                 \n",
            " dense (Dense)               (None, 140)               16940     \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 11638)             1640958   \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 2,446,918\n",
            "Trainable params: 2,446,918\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MV-pOsxbhPd8"
      },
      "source": [
        "bi_di_gru.compile(loss='categorical_crossentropy',optimizer='adam',metrics=['accuracy'])"
      ],
      "id": "MV-pOsxbhPd8",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "H9gctIvYhPoM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5bbeec21-6b87-487b-d517-24f3597bb64d"
      },
      "source": [
        "bidi_gru_history = bi_di_gru.fit(X,y,batch_size=256,epochs=150)"
      ],
      "id": "H9gctIvYhPoM",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/150\n",
            "430/430 [==============================] - 36s 68ms/step - loss: 6.9204 - accuracy: 0.0636\n",
            "Epoch 2/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 6.3982 - accuracy: 0.0801\n",
            "Epoch 3/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 6.0772 - accuracy: 0.0997\n",
            "Epoch 4/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 5.8242 - accuracy: 0.1168\n",
            "Epoch 5/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 5.6297 - accuracy: 0.1285\n",
            "Epoch 6/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 5.4616 - accuracy: 0.1364\n",
            "Epoch 7/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 5.3014 - accuracy: 0.1445\n",
            "Epoch 8/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 5.1444 - accuracy: 0.1519\n",
            "Epoch 9/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 4.9924 - accuracy: 0.1589\n",
            "Epoch 10/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 4.8442 - accuracy: 0.1662\n",
            "Epoch 11/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 4.6984 - accuracy: 0.1729\n",
            "Epoch 12/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 4.5598 - accuracy: 0.1793\n",
            "Epoch 13/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 4.4223 - accuracy: 0.1874\n",
            "Epoch 14/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 4.2888 - accuracy: 0.1944\n",
            "Epoch 15/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 4.1617 - accuracy: 0.2053\n",
            "Epoch 16/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 4.0361 - accuracy: 0.2175\n",
            "Epoch 17/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 3.9127 - accuracy: 0.2308\n",
            "Epoch 18/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 3.7962 - accuracy: 0.2449\n",
            "Epoch 19/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 3.6847 - accuracy: 0.2586\n",
            "Epoch 20/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 3.5821 - accuracy: 0.2712\n",
            "Epoch 21/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 3.4817 - accuracy: 0.2843\n",
            "Epoch 22/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 3.3910 - accuracy: 0.2966\n",
            "Epoch 23/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 3.3039 - accuracy: 0.3087\n",
            "Epoch 24/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 3.2251 - accuracy: 0.3193\n",
            "Epoch 25/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 3.1508 - accuracy: 0.3320\n",
            "Epoch 26/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 3.0788 - accuracy: 0.3408\n",
            "Epoch 27/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 3.0095 - accuracy: 0.3524\n",
            "Epoch 28/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.9445 - accuracy: 0.3621\n",
            "Epoch 29/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.8834 - accuracy: 0.3722\n",
            "Epoch 30/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 2.8241 - accuracy: 0.3809\n",
            "Epoch 31/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.7677 - accuracy: 0.3898\n",
            "Epoch 32/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.7153 - accuracy: 0.3979\n",
            "Epoch 33/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.6594 - accuracy: 0.4081\n",
            "Epoch 34/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.6123 - accuracy: 0.4144\n",
            "Epoch 35/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 2.5617 - accuracy: 0.4241\n",
            "Epoch 36/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.5153 - accuracy: 0.4321\n",
            "Epoch 37/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.4700 - accuracy: 0.4383\n",
            "Epoch 38/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.4326 - accuracy: 0.4456\n",
            "Epoch 39/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.3895 - accuracy: 0.4526\n",
            "Epoch 40/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.3479 - accuracy: 0.4597\n",
            "Epoch 41/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.3066 - accuracy: 0.4682\n",
            "Epoch 42/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.2717 - accuracy: 0.4741\n",
            "Epoch 43/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 2.2359 - accuracy: 0.4808\n",
            "Epoch 44/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.1971 - accuracy: 0.4882\n",
            "Epoch 45/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.1672 - accuracy: 0.4926\n",
            "Epoch 46/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.1331 - accuracy: 0.5014\n",
            "Epoch 47/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.0978 - accuracy: 0.5065\n",
            "Epoch 48/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.0673 - accuracy: 0.5120\n",
            "Epoch 49/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 2.0345 - accuracy: 0.5183\n",
            "Epoch 50/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 2.0109 - accuracy: 0.5221\n",
            "Epoch 51/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.9819 - accuracy: 0.5259\n",
            "Epoch 52/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.9504 - accuracy: 0.5331\n",
            "Epoch 53/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.9197 - accuracy: 0.5386\n",
            "Epoch 54/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.8947 - accuracy: 0.5455\n",
            "Epoch 55/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.8683 - accuracy: 0.5499\n",
            "Epoch 56/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.8434 - accuracy: 0.5527\n",
            "Epoch 57/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.8167 - accuracy: 0.5602\n",
            "Epoch 58/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.7956 - accuracy: 0.5631\n",
            "Epoch 59/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.7667 - accuracy: 0.5692\n",
            "Epoch 60/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.7475 - accuracy: 0.5733\n",
            "Epoch 61/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.7269 - accuracy: 0.5753\n",
            "Epoch 62/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.7002 - accuracy: 0.5827\n",
            "Epoch 63/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.6829 - accuracy: 0.5861\n",
            "Epoch 64/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.6591 - accuracy: 0.5909\n",
            "Epoch 65/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.6342 - accuracy: 0.5941\n",
            "Epoch 66/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.6125 - accuracy: 0.5985\n",
            "Epoch 67/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.5987 - accuracy: 0.6013\n",
            "Epoch 68/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.5814 - accuracy: 0.6039\n",
            "Epoch 69/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.5558 - accuracy: 0.6122\n",
            "Epoch 70/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.5451 - accuracy: 0.6126\n",
            "Epoch 71/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.5231 - accuracy: 0.6170\n",
            "Epoch 72/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.5035 - accuracy: 0.6213\n",
            "Epoch 73/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.4865 - accuracy: 0.6253\n",
            "Epoch 74/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.4640 - accuracy: 0.6292\n",
            "Epoch 75/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.4525 - accuracy: 0.6321\n",
            "Epoch 76/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.4364 - accuracy: 0.6347\n",
            "Epoch 77/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.4296 - accuracy: 0.6361\n",
            "Epoch 78/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.4014 - accuracy: 0.6424\n",
            "Epoch 79/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.3895 - accuracy: 0.6441\n",
            "Epoch 80/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.3707 - accuracy: 0.6485\n",
            "Epoch 81/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.3526 - accuracy: 0.6524\n",
            "Epoch 82/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.3471 - accuracy: 0.6534\n",
            "Epoch 83/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.3351 - accuracy: 0.6568\n",
            "Epoch 84/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.3148 - accuracy: 0.6597\n",
            "Epoch 85/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.3033 - accuracy: 0.6632\n",
            "Epoch 86/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.2891 - accuracy: 0.6643\n",
            "Epoch 87/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.2783 - accuracy: 0.6682\n",
            "Epoch 88/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.2614 - accuracy: 0.6713\n",
            "Epoch 89/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.2523 - accuracy: 0.6723\n",
            "Epoch 90/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.2508 - accuracy: 0.6724\n",
            "Epoch 91/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.2309 - accuracy: 0.6783\n",
            "Epoch 92/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.2200 - accuracy: 0.6775\n",
            "Epoch 93/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.1997 - accuracy: 0.6843\n",
            "Epoch 94/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.1899 - accuracy: 0.6866\n",
            "Epoch 95/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.1785 - accuracy: 0.6892\n",
            "Epoch 96/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.1755 - accuracy: 0.6904\n",
            "Epoch 97/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.1602 - accuracy: 0.6934\n",
            "Epoch 98/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.1424 - accuracy: 0.6972\n",
            "Epoch 99/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.1363 - accuracy: 0.6984\n",
            "Epoch 100/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.1277 - accuracy: 0.7018\n",
            "Epoch 101/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.1181 - accuracy: 0.7016\n",
            "Epoch 102/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.1051 - accuracy: 0.7049\n",
            "Epoch 103/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0998 - accuracy: 0.7056\n",
            "Epoch 104/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0966 - accuracy: 0.7049\n",
            "Epoch 105/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0781 - accuracy: 0.7096\n",
            "Epoch 106/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0708 - accuracy: 0.7122\n",
            "Epoch 107/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0647 - accuracy: 0.7131\n",
            "Epoch 108/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0548 - accuracy: 0.7159\n",
            "Epoch 109/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0402 - accuracy: 0.7205\n",
            "Epoch 110/150\n",
            "430/430 [==============================] - 29s 67ms/step - loss: 1.0383 - accuracy: 0.7205\n",
            "Epoch 111/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0332 - accuracy: 0.7211\n",
            "Epoch 112/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0236 - accuracy: 0.7230\n",
            "Epoch 113/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0147 - accuracy: 0.7246\n",
            "Epoch 114/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 1.0082 - accuracy: 0.7266\n",
            "Epoch 115/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9947 - accuracy: 0.7278\n",
            "Epoch 116/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9886 - accuracy: 0.7299\n",
            "Epoch 117/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9741 - accuracy: 0.7340\n",
            "Epoch 118/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9716 - accuracy: 0.7344\n",
            "Epoch 119/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9655 - accuracy: 0.7352\n",
            "Epoch 120/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9594 - accuracy: 0.7361\n",
            "Epoch 121/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9575 - accuracy: 0.7368\n",
            "Epoch 122/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9497 - accuracy: 0.7384\n",
            "Epoch 123/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9329 - accuracy: 0.7451\n",
            "Epoch 124/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9261 - accuracy: 0.7449\n",
            "Epoch 125/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9200 - accuracy: 0.7457\n",
            "Epoch 126/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9152 - accuracy: 0.7471\n",
            "Epoch 127/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9103 - accuracy: 0.7474\n",
            "Epoch 128/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.9104 - accuracy: 0.7479\n",
            "Epoch 129/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8982 - accuracy: 0.7519\n",
            "Epoch 130/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8970 - accuracy: 0.7504\n",
            "Epoch 131/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8914 - accuracy: 0.7525\n",
            "Epoch 132/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8833 - accuracy: 0.7556\n",
            "Epoch 133/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8654 - accuracy: 0.7604\n",
            "Epoch 134/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8635 - accuracy: 0.7597\n",
            "Epoch 135/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8609 - accuracy: 0.7614\n",
            "Epoch 136/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8603 - accuracy: 0.7611\n",
            "Epoch 137/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8495 - accuracy: 0.7635\n",
            "Epoch 138/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8531 - accuracy: 0.7629\n",
            "Epoch 139/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8444 - accuracy: 0.7636\n",
            "Epoch 140/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8272 - accuracy: 0.7676\n",
            "Epoch 141/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8309 - accuracy: 0.7667\n",
            "Epoch 142/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8359 - accuracy: 0.7668\n",
            "Epoch 143/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8250 - accuracy: 0.7674\n",
            "Epoch 144/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8250 - accuracy: 0.7684\n",
            "Epoch 145/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8111 - accuracy: 0.7709\n",
            "Epoch 146/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.8016 - accuracy: 0.7750\n",
            "Epoch 147/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.7959 - accuracy: 0.7758\n",
            "Epoch 148/150\n",
            "430/430 [==============================] - 30s 69ms/step - loss: 0.7933 - accuracy: 0.7742\n",
            "Epoch 149/150\n",
            "430/430 [==============================] - 30s 69ms/step - loss: 0.7907 - accuracy: 0.7781\n",
            "Epoch 150/150\n",
            "430/430 [==============================] - 29s 68ms/step - loss: 0.7849 - accuracy: 0.7783\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d2s0sOAahPyb"
      },
      "source": [
        "bi_di_gru.save('./Bi_di_GRU.h5')"
      ],
      "id": "d2s0sOAahPyb",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "buKqh5Zai8Zj",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "aeceea6c-1ba2-437b-a868-aec171e23a27"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "epochs = range(len(bidi_gru_history.history['accuracy']))\n",
        "plt.plot(epochs,bidi_gru_history.history['accuracy'])\n",
        "plt.title(\"Bi-directional GRU Model Training Accuracy\")\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel('Accuracy')\n",
        "plt.show()"
      ],
      "id": "buKqh5Zai8Zj",
      "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": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IBFardS2i8cU",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "2e24b0ec-7fc6-4311-c7ad-a00fb778f252"
      },
      "source": [
        "plt.plot(epochs,bidi_gru_history.history['loss'])\n",
        "plt.title(\"Bi-directional GRU Model Training Loss\")\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel('Loss')\n",
        "plt.show()"
      ],
      "id": "IBFardS2i8cU",
      "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": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lLb-psmskbDp"
      },
      "source": [
        "### Generating Stories from Bi-directional GRU Model"
      ],
      "id": "lLb-psmskbDp"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YUd7pkjOjNvm",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "9a863ea3-6e40-4454-b54e-02d987d86776"
      },
      "source": [
        "seed_text = \"The country was in chaos but\"\n",
        "generate_story(bi_di_gru, tokenizer, seq_length, seed_text, 50) #generate next 50 words"
      ],
      "id": "YUd7pkjOjNvm",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'they had helped they were back for a mask when he remembered was the child to it had a small patch of congealed blood and the dead s company was letting him midnight the piano on a federal photographer and were out of sorts and social kingdoms your books have'"
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xS_fHOzvjNyp",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "596d043e-ec37-47ad-8b99-4f2f05cd56fd"
      },
      "source": [
        "seed_text = \"I walked out of the store dissatisfied and it\"\n",
        "generate_story(bi_di_gru, tokenizer, seq_length, seed_text, 50) #generate next 50 words"
      ],
      "id": "xS_fHOzvjNyp",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'had been devoured grave he cried that he had lost within the character which was three years he could hear johnny called up and with tears and then he dozed offone day when it was like a midshipman and walked in the island of the fort and a piece of'"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2DDomz4yjN1K",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "c1112751-ddb4-4f31-8503-00acdf64e6e7"
      },
      "source": [
        "seed_text = \"The lady was known as a soothsayer - a psychic of sorts. The rubbing of her crystal ball was finalized as she completed her prophecy. Her hands were wrinkled and dirty, a strange companion to the polish of the ball she held in her hands.\"\n",
        "generate_story(bi_di_gru, tokenizer, seq_length, seed_text, 50) #generate next 50 words"
      ],
      "id": "2DDomz4yjN1K",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'on every year and that afternoon or until no doubt leapt from my own determination for off de american boy held his champagne he must cook it hundreds of times but i did not know anything he announced gleefully then remembering the family insisted that he and admiral remain behindit'"
            ]
          },
          "metadata": {},
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0g0WHUx3jN3q",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        },
        "outputId": "8f54182d-7ed8-45a2-e9e0-38f9bcacd784"
      },
      "source": [
        "seed_text = \"Little Chandler's thoughts ever since lunch-time had been of his meeting with Gallaher, of Gallaher's invitation, and of the great city London where Gallaher lived. He was called Little Chandler because\"\n",
        "generate_story(bi_di_gru, tokenizer, seq_length, seed_text, 100) #generate next 100 words"
      ],
      "id": "0g0WHUx3jN3q",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'of course we were s means in melancholy in deep and it turned the cliff and his hair was still more as though it were the elder by a matter of minutes and then they were conscious of a small family but no nightlight to make a blessed breach no i ll get up he said somewhat all right we re feeling whom he wondered if he had ever been remembered of course i spoke in a muffled whisper and no doubt flemming these stories oh henry daw was free to see he said he thought malignantly i thought about'"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6NJGKzpnjN6i",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 90
        },
        "outputId": "5efac043-5cd6-4da3-ebe7-bbfffa21067a"
      },
      "source": [
        "seed_text = \"On the morning of the fifth of September, in uniform, his revolver on the table, the colonel addressed his soliders and\"\n",
        "generate_story(bi_di_gru, tokenizer, seq_length, seed_text, 100) #generate next 100 words"
      ],
      "id": "6NJGKzpnjN6i",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'stand quite still more than a baron she proffered no explanation his silence was irritating and made her own plans and myself are haunting me for now that all the first time about whose ferocity was rendered gentle which it was late to be he did not go home in the drama that one could derive true enjoyment and become cultivated and humane but do you suppose the public understands that she used to say what they want to know what i was asked he remarked somewhat later ef i was asked expectations i thought he d throw over some'"
            ]
          },
          "metadata": {},
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KJKn0J0IjN9P",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 90
        },
        "outputId": "ab13383e-8cc0-42a7-bb98-00b7188d0ffc"
      },
      "source": [
        "seed_text = \"On Christmas Eve around 1978, when I was a young boy, after having the traditional pastry and cookies and after having sung around the fire, I, with my sister, my mother and my aunts, and Emma Jackman and her son, got into Emma Jackman's car and then later\"\n",
        "generate_story(bi_di_gru, tokenizer, seq_length, seed_text, 100) #generate next 100 words"
      ],
      "id": "KJKn0J0IjN9P",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'ivan dmitritch thought was little few showing me proffered a poor window the things of silence the band of a small smug town coming to him as if what also could be set advantage of the middle of the big manufacturers of the while and you know but there briefly no longer clearly reply when the commander s men out of human human mother are impossible for me came and s pause i know the incident of the sturgeon marfa went up to the mogilev wilderness he s only a confirmed place of his regiment and i learned this and'"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    }
  ]
}