pp_2dict.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ea6fa89b",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "from __future__ import print_function, division\n",
    "from builtins import range, input\n",
    "# Note: you may need to update your version of future\n",
    "# sudo pip install -U future\n",
    "\n",
    "import pickle\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.utils import shuffle\n",
    "\n",
    "# load in the data\n",
    "# https://www.kaggle.com/grouplens/movielens-20m-dataset\n",
    "df = pd.read_csv('C://Users//Admin//Documents//GitHub//small_rating.csv')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c33887bb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Calling: update_user2movie_and_movie2user\n",
      "processed: 0.023\n",
      "processed: 0.046\n",
      "processed: 0.070\n",
      "processed: 0.093\n",
      "processed: 0.116\n",
      "processed: 0.139\n",
      "processed: 0.162\n",
      "processed: 0.185\n",
      "processed: 0.209\n",
      "processed: 0.232\n",
      "processed: 0.255\n",
      "processed: 0.278\n",
      "processed: 0.301\n",
      "processed: 0.325\n",
      "processed: 0.348\n",
      "processed: 0.371\n",
      "processed: 0.394\n",
      "processed: 0.417\n",
      "processed: 0.440\n",
      "processed: 0.464\n",
      "processed: 0.487\n",
      "processed: 0.510\n",
      "processed: 0.533\n",
      "processed: 0.556\n",
      "processed: 0.580\n",
      "processed: 0.603\n",
      "processed: 0.626\n",
      "processed: 0.649\n",
      "processed: 0.672\n",
      "processed: 0.695\n",
      "processed: 0.719\n",
      "processed: 0.742\n",
      "processed: 0.765\n",
      "processed: 0.788\n",
      "processed: 0.811\n",
      "processed: 0.835\n",
      "processed: 0.858\n",
      "processed: 0.881\n",
      "processed: 0.904\n",
      "processed: 0.927\n",
      "processed: 0.950\n",
      "processed: 0.974\n",
      "processed: 0.997\n",
      "Calling: update_usermovie2rating_test\n",
      "processed: 0.093\n",
      "processed: 0.185\n",
      "processed: 0.278\n",
      "processed: 0.371\n",
      "processed: 0.464\n",
      "processed: 0.556\n",
      "processed: 0.649\n",
      "processed: 0.742\n",
      "processed: 0.835\n",
      "processed: 0.927\n"
     ]
    }
   ],
   "source": [
    "N = df.userId.max() + 1 # number of users\n",
    "M = df.movie_idx.max() + 1 # number of movies\n",
    "\n",
    "# split into train and test\n",
    "df = shuffle(df)\n",
    "cutoff = int(0.8*len(df))\n",
    "df_train = df.iloc[:cutoff]\n",
    "df_test = df.iloc[cutoff:]\n",
    "\n",
    "# a dictionary to tell us which users have rated which movies\n",
    "user2movie = {}\n",
    "# a dicationary to tell us which movies have been rated by which users\n",
    "movie2user = {}\n",
    "# a dictionary to look up ratings\n",
    "usermovie2rating = {}\n",
    "print(\"Calling: update_user2movie_and_movie2user\")\n",
    "count = 0\n",
    "def update_user2movie_and_movie2user(row):\n",
    "  global count\n",
    "  count += 1\n",
    "  if count % 100000 == 0:\n",
    "    print(\"processed: %.3f\" % (float(count)/cutoff))\n",
    "\n",
    "  i = int(row.userId)\n",
    "  j = int(row.movie_idx)\n",
    "  if i not in user2movie:\n",
    "    user2movie[i] = [j]\n",
    "  else:\n",
    "    user2movie[i].append(j)\n",
    "\n",
    "  if j not in movie2user:\n",
    "    movie2user[j] = [i]\n",
    "  else:\n",
    "    movie2user[j].append(i)\n",
    "\n",
    "  usermovie2rating[(i,j)] = row.rating\n",
    "df_train.apply(update_user2movie_and_movie2user, axis=1)\n",
    "\n",
    "# test ratings dictionary\n",
    "usermovie2rating_test = {}\n",
    "print(\"Calling: update_usermovie2rating_test\")\n",
    "count = 0\n",
    "def update_usermovie2rating_test(row):\n",
    "  global count\n",
    "  count += 1\n",
    "  if count % 100000 == 0:\n",
    "    print(\"processed: %.3f\" % (float(count)/len(df_test)))\n",
    "\n",
    "  i = int(row.userId)\n",
    "  j = int(row.movie_idx)\n",
    "  usermovie2rating_test[(i,j)] = row.rating\n",
    "df_test.apply(update_usermovie2rating_test, axis=1)\n",
    "\n",
    "# note: these are not really JSONs\n",
    "with open('user2movie.json', 'wb') as f:\n",
    "  pickle.dump(user2movie, f)\n",
    "\n",
    "with open('movie2user.json', 'wb') as f:\n",
    "  pickle.dump(movie2user, f)\n",
    "\n",
    "with open('usermovie2rating.json', 'wb') as f:\n",
    "  pickle.dump(usermovie2rating, f)\n",
    "\n",
    "with open('usermovie2rating_test.json', 'wb') as f:\n",
    "  pickle.dump(usermovie2rating_test, f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "64a3934a",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}