MovieLens 1M Recommendations.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using Theano backend.\n",
      "Using gpu device 0: GRID K520 (CNMeM is disabled, cuDNN 5006)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from CFModel import CFModel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define constants\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "RATINGS_CSV_FILE = 'ml1m_ratings.csv'\n",
    "USERS_CSV_FILE = 'ml1m_users.csv'\n",
    "MOVIES_CSV_FILE = 'ml1m_movies.csv'\n",
    "MODEL_WEIGHTS_FILE = 'ml1m_weights.h5'\n",
    "K_FACTORS = 120\n",
    "TEST_USER = 3000"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load MovieLens 1M data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000209 ratings loaded.\n"
     ]
    }
   ],
   "source": [
    "ratings = pd.read_csv(RATINGS_CSV_FILE, sep='\\t', encoding='latin-1', usecols=['userid', 'movieid', 'rating'])\n",
    "max_userid = ratings['userid'].drop_duplicates().max()\n",
    "max_movieid = ratings['movieid'].drop_duplicates().max()\n",
    "print len(ratings), 'ratings loaded.'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6040 descriptions of 6040 users loaded.\n"
     ]
    }
   ],
   "source": [
    "users = pd.read_csv(USERS_CSV_FILE, sep='\\t', encoding='latin-1', usecols=['userid', 'gender', 'zipcode', 'age_desc', 'occ_desc'])\n",
    "print len(users), 'descriptions of', max_userid, 'users loaded.'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3883 descriptions of 3952 movies loaded.\n"
     ]
    }
   ],
   "source": [
    "movies = pd.read_csv(MOVIES_CSV_FILE, sep='\\t', encoding='latin-1', usecols=['movieid', 'title', 'genre'])\n",
    "print len(movies), 'descriptions of', max_movieid, 'movies loaded.'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Make recommendations for a given user"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "trained_model = CFModel(max_userid, max_movieid, K_FACTORS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "trained_model.load_weights(MODEL_WEIGHTS_FILE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>userid</th>\n",
       "      <th>gender</th>\n",
       "      <th>zipcode</th>\n",
       "      <th>age_desc</th>\n",
       "      <th>occ_desc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2999</th>\n",
       "      <td>3000</td>\n",
       "      <td>M</td>\n",
       "      <td>55408</td>\n",
       "      <td>25-34</td>\n",
       "      <td>college/grad student</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      userid gender zipcode age_desc              occ_desc\n",
       "2999    3000      M   55408    25-34  college/grad student"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "users[users['userid'] == TEST_USER]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def predict_rating(userid, movieid):\n",
    "    return trained_model.rate(userid - 1, movieid - 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>userid</th>\n",
       "      <th>movieid</th>\n",
       "      <th>rating</th>\n",
       "      <th>prediction</th>\n",
       "      <th>title</th>\n",
       "      <th>genre</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3000</td>\n",
       "      <td>590</td>\n",
       "      <td>5</td>\n",
       "      <td>3.153018</td>\n",
       "      <td>Dances with Wolves (1990)</td>\n",
       "      <td>Adventure|Drama|Western</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3000</td>\n",
       "      <td>3552</td>\n",
       "      <td>5</td>\n",
       "      <td>3.684154</td>\n",
       "      <td>Caddyshack (1980)</td>\n",
       "      <td>Comedy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3000</td>\n",
       "      <td>2858</td>\n",
       "      <td>5</td>\n",
       "      <td>5.018598</td>\n",
       "      <td>American Beauty (1999)</td>\n",
       "      <td>Comedy|Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3000</td>\n",
       "      <td>3358</td>\n",
       "      <td>5</td>\n",
       "      <td>3.419102</td>\n",
       "      <td>Defending Your Life (1991)</td>\n",
       "      <td>Comedy|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3000</td>\n",
       "      <td>2968</td>\n",
       "      <td>5</td>\n",
       "      <td>3.388493</td>\n",
       "      <td>Time Bandits (1981)</td>\n",
       "      <td>Adventure|Fantasy|Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3000</td>\n",
       "      <td>1307</td>\n",
       "      <td>5</td>\n",
       "      <td>3.841095</td>\n",
       "      <td>When Harry Met Sally... (1989)</td>\n",
       "      <td>Comedy|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>3000</td>\n",
       "      <td>144</td>\n",
       "      <td>5</td>\n",
       "      <td>3.058760</td>\n",
       "      <td>Brothers McMullen, The (1995)</td>\n",
       "      <td>Comedy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3000</td>\n",
       "      <td>1193</td>\n",
       "      <td>5</td>\n",
       "      <td>4.014853</td>\n",
       "      <td>One Flew Over the Cuckoo's Nest (1975)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>3000</td>\n",
       "      <td>1265</td>\n",
       "      <td>5</td>\n",
       "      <td>3.823973</td>\n",
       "      <td>Groundhog Day (1993)</td>\n",
       "      <td>Comedy|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>3000</td>\n",
       "      <td>733</td>\n",
       "      <td>5</td>\n",
       "      <td>3.066499</td>\n",
       "      <td>Rock, The (1996)</td>\n",
       "      <td>Action|Adventure|Thriller</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   userid  movieid  rating  prediction  \\\n",
       "0    3000      590       5    3.153018   \n",
       "1    3000     3552       5    3.684154   \n",
       "2    3000     2858       5    5.018598   \n",
       "3    3000     3358       5    3.419102   \n",
       "4    3000     2968       5    3.388493   \n",
       "5    3000     1307       5    3.841095   \n",
       "6    3000      144       5    3.058760   \n",
       "7    3000     1193       5    4.014853   \n",
       "8    3000     1265       5    3.823973   \n",
       "9    3000      733       5    3.066499   \n",
       "\n",
       "                                    title                      genre  \n",
       "0               Dances with Wolves (1990)    Adventure|Drama|Western  \n",
       "1                       Caddyshack (1980)                     Comedy  \n",
       "2                  American Beauty (1999)               Comedy|Drama  \n",
       "3              Defending Your Life (1991)             Comedy|Romance  \n",
       "4                     Time Bandits (1981)   Adventure|Fantasy|Sci-Fi  \n",
       "5          When Harry Met Sally... (1989)             Comedy|Romance  \n",
       "6           Brothers McMullen, The (1995)                     Comedy  \n",
       "7  One Flew Over the Cuckoo's Nest (1975)                      Drama  \n",
       "8                    Groundhog Day (1993)             Comedy|Romance  \n",
       "9                        Rock, The (1996)  Action|Adventure|Thriller  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_ratings = ratings[ratings['userid'] == TEST_USER][['userid', 'movieid', 'rating']]\n",
    "user_ratings['prediction'] = user_ratings.apply(lambda x: predict_rating(TEST_USER, x['movieid']), axis=1)\n",
    "user_ratings.sort_values(by='rating', \n",
    "                         ascending=False).merge(movies, \n",
    "                                                on='movieid', \n",
    "                                                how='inner', \n",
    "                                                suffixes=['_u', '_m']).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieid</th>\n",
       "      <th>prediction</th>\n",
       "      <th>title</th>\n",
       "      <th>genre</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2959</td>\n",
       "      <td>4.624823</td>\n",
       "      <td>Fight Club (1999)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>296</td>\n",
       "      <td>4.522243</td>\n",
       "      <td>Pulp Fiction (1994)</td>\n",
       "      <td>Crime|Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3160</td>\n",
       "      <td>4.459617</td>\n",
       "      <td>Magnolia (1999)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1136</td>\n",
       "      <td>4.371700</td>\n",
       "      <td>Monty Python and the Holy Grail (1974)</td>\n",
       "      <td>Comedy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>50</td>\n",
       "      <td>4.365819</td>\n",
       "      <td>Usual Suspects, The (1995)</td>\n",
       "      <td>Crime|Thriller</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2395</td>\n",
       "      <td>4.290950</td>\n",
       "      <td>Rushmore (1998)</td>\n",
       "      <td>Comedy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2762</td>\n",
       "      <td>4.255275</td>\n",
       "      <td>Sixth Sense, The (1999)</td>\n",
       "      <td>Thriller</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>745</td>\n",
       "      <td>4.232957</td>\n",
       "      <td>Close Shave, A (1995)</td>\n",
       "      <td>Animation|Comedy|Thriller</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>593</td>\n",
       "      <td>4.197603</td>\n",
       "      <td>Silence of the Lambs, The (1991)</td>\n",
       "      <td>Drama|Thriller</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>608</td>\n",
       "      <td>4.188972</td>\n",
       "      <td>Fargo (1996)</td>\n",
       "      <td>Crime|Drama|Thriller</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieid  prediction                                   title  \\\n",
       "0     2959    4.624823                       Fight Club (1999)   \n",
       "1      296    4.522243                     Pulp Fiction (1994)   \n",
       "2     3160    4.459617                         Magnolia (1999)   \n",
       "3     1136    4.371700  Monty Python and the Holy Grail (1974)   \n",
       "4       50    4.365819              Usual Suspects, The (1995)   \n",
       "5     2395    4.290950                         Rushmore (1998)   \n",
       "6     2762    4.255275                 Sixth Sense, The (1999)   \n",
       "7      745    4.232957                   Close Shave, A (1995)   \n",
       "8      593    4.197603        Silence of the Lambs, The (1991)   \n",
       "9      608    4.188972                            Fargo (1996)   \n",
       "\n",
       "                       genre  \n",
       "0                      Drama  \n",
       "1                Crime|Drama  \n",
       "2                      Drama  \n",
       "3                     Comedy  \n",
       "4             Crime|Thriller  \n",
       "5                     Comedy  \n",
       "6                   Thriller  \n",
       "7  Animation|Comedy|Thriller  \n",
       "8             Drama|Thriller  \n",
       "9       Crime|Drama|Thriller  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recommendations = ratings[ratings['movieid'].isin(user_ratings['movieid']) == False][['movieid']].drop_duplicates()\n",
    "recommendations['prediction'] = recommendations.apply(lambda x: predict_rating(TEST_USER, x['movieid']), axis=1)\n",
    "recommendations.sort_values(by='prediction',\n",
    "                          ascending=False).merge(movies,\n",
    "                                                 on='movieid',\n",
    "                                                 how='inner',\n",
    "                                                 suffixes=['_u', '_m']).head(10)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}