first_bot.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "# eden = open('static/data/Eden.txt', encoding='utf8').read()\n",
    "# eden = open('static/data/Large.txt', encoding='utf8').read()\n",
    "eden = open('static/data/Disney.txt', encoding='utf8').read()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1873"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corpus = eden.split()\n",
    "len(corpus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_pairs(corpus):\n",
    "    for i in range(len(corpus)-1):\n",
    "        yield (corpus[i], corpus[i+1])\n",
    "        \n",
    "pairs = make_pairs(corpus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "word_dict = {}\n",
    "for word_1, word_2 in pairs:\n",
    "    if word_1 in word_dict.keys():\n",
    "        word_dict[word_1].append(word_2)\n",
    "    else:\n",
    "        word_dict[word_1] = [word_2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"That for rotten judgement, I knew But when did You And I go, there's just no\""
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "first_word = np.random.choice(corpus)\n",
    "chain = [first_word]\n",
    "n_words = 15\n",
    "for i in range(n_words):\n",
    "    chain.append(np.random.choice(word_dict[chain[-1]]))\n",
    "' '.join(chain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.6.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}