Collaborative Filtering.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "execution": {
     "iopub.execute_input": "2023-02-16T15:52:48.704179Z",
     "iopub.status.busy": "2023-02-16T15:52:48.703757Z",
     "iopub.status.idle": "2023-02-16T15:52:48.713005Z",
     "shell.execute_reply": "2023-02-16T15:52:48.711953Z",
     "shell.execute_reply.started": "2023-02-16T15:52:48.704144Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import random\n",
    "import matplotlib.pyplot as plt\n",
    "import torch\n",
    "import torch.nn.functional as F\n",
    "import torch.nn as nn\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "from sklearn import preprocessing\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import precision_score,recall_score,accuracy_score,roc_curve,auc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3> Data Preprocessing and Analysis </h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:34:51.831695Z",
     "iopub.status.busy": "2023-02-16T15:34:51.831175Z",
     "iopub.status.idle": "2023-02-16T15:34:56.662811Z",
     "shell.execute_reply": "2023-02-16T15:34:56.661807Z",
     "shell.execute_reply.started": "2023-02-16T15:34:51.831662Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/pandas/util/_decorators.py:311: ParserWarning: Falling back to the 'python' engine because the 'c' engine does not support regex separators (separators > 1 char and different from '\\s+' are interpreted as regex); you can avoid this warning by specifying engine='python'.\n",
      "  return func(*args, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "movies_df = pd.read_table(\"/kaggle/input/movielens-1m-dataset/movies.dat\",sep=\"::\",names=['id','title','genres'],encoding='latin-1')\n",
    "ratings_df = pd.read_table(\"/kaggle/input/movielens-1m-dataset/ratings.dat\",sep=\"::\",names=['user_id','movie_id','rating','timestamp'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:34:56.664972Z",
     "iopub.status.busy": "2023-02-16T15:34:56.664612Z",
     "iopub.status.idle": "2023-02-16T15:34:56.838413Z",
     "shell.execute_reply": "2023-02-16T15:34:56.837234Z",
     "shell.execute_reply.started": "2023-02-16T15:34:56.664938Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>movie_id</th>\n",
       "      <th>rating</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>id</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1193</td>\n",
       "      <td>5</td>\n",
       "      <td>978300760</td>\n",
       "      <td>1193</td>\n",
       "      <td>One Flew Over the Cuckoo's Nest (1975)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>661</td>\n",
       "      <td>3</td>\n",
       "      <td>978302109</td>\n",
       "      <td>661</td>\n",
       "      <td>James and the Giant Peach (1996)</td>\n",
       "      <td>Animation|Children's|Musical</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>914</td>\n",
       "      <td>3</td>\n",
       "      <td>978301968</td>\n",
       "      <td>914</td>\n",
       "      <td>My Fair Lady (1964)</td>\n",
       "      <td>Musical|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>3408</td>\n",
       "      <td>4</td>\n",
       "      <td>978300275</td>\n",
       "      <td>3408</td>\n",
       "      <td>Erin Brockovich (2000)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>2355</td>\n",
       "      <td>5</td>\n",
       "      <td>978824291</td>\n",
       "      <td>2355</td>\n",
       "      <td>Bug's Life, A (1998)</td>\n",
       "      <td>Animation|Children's|Comedy</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id  movie_id  rating  timestamp    id  \\\n",
       "0        1      1193       5  978300760  1193   \n",
       "1        1       661       3  978302109   661   \n",
       "2        1       914       3  978301968   914   \n",
       "3        1      3408       4  978300275  3408   \n",
       "4        1      2355       5  978824291  2355   \n",
       "\n",
       "                                    title                        genres  \n",
       "0  One Flew Over the Cuckoo's Nest (1975)                         Drama  \n",
       "1        James and the Giant Peach (1996)  Animation|Children's|Musical  \n",
       "2                     My Fair Lady (1964)               Musical|Romance  \n",
       "3                  Erin Brockovich (2000)                         Drama  \n",
       "4                    Bug's Life, A (1998)   Animation|Children's|Comedy  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ratings_df = pd.merge(ratings_df, movies_df, how='left', left_on='movie_id', right_on='id')\n",
    "ratings_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:35:02.409172Z",
     "iopub.status.busy": "2023-02-16T15:35:02.408498Z",
     "iopub.status.idle": "2023-02-16T15:35:02.567003Z",
     "shell.execute_reply": "2023-02-16T15:35:02.565897Z",
     "shell.execute_reply.started": "2023-02-16T15:35:02.409138Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>movie_id</th>\n",
       "      <th>rating</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.000209e+06</td>\n",
       "      <td>1.000209e+06</td>\n",
       "      <td>1.000209e+06</td>\n",
       "      <td>1.000209e+06</td>\n",
       "      <td>1.000209e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>3.024512e+03</td>\n",
       "      <td>1.865540e+03</td>\n",
       "      <td>3.581564e+00</td>\n",
       "      <td>9.722437e+08</td>\n",
       "      <td>1.865540e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.728413e+03</td>\n",
       "      <td>1.096041e+03</td>\n",
       "      <td>1.117102e+00</td>\n",
       "      <td>1.215256e+07</td>\n",
       "      <td>1.096041e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>9.567039e+08</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.506000e+03</td>\n",
       "      <td>1.030000e+03</td>\n",
       "      <td>3.000000e+00</td>\n",
       "      <td>9.653026e+08</td>\n",
       "      <td>1.030000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3.070000e+03</td>\n",
       "      <td>1.835000e+03</td>\n",
       "      <td>4.000000e+00</td>\n",
       "      <td>9.730180e+08</td>\n",
       "      <td>1.835000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4.476000e+03</td>\n",
       "      <td>2.770000e+03</td>\n",
       "      <td>4.000000e+00</td>\n",
       "      <td>9.752209e+08</td>\n",
       "      <td>2.770000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6.040000e+03</td>\n",
       "      <td>3.952000e+03</td>\n",
       "      <td>5.000000e+00</td>\n",
       "      <td>1.046455e+09</td>\n",
       "      <td>3.952000e+03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            user_id      movie_id        rating     timestamp            id\n",
       "count  1.000209e+06  1.000209e+06  1.000209e+06  1.000209e+06  1.000209e+06\n",
       "mean   3.024512e+03  1.865540e+03  3.581564e+00  9.722437e+08  1.865540e+03\n",
       "std    1.728413e+03  1.096041e+03  1.117102e+00  1.215256e+07  1.096041e+03\n",
       "min    1.000000e+00  1.000000e+00  1.000000e+00  9.567039e+08  1.000000e+00\n",
       "25%    1.506000e+03  1.030000e+03  3.000000e+00  9.653026e+08  1.030000e+03\n",
       "50%    3.070000e+03  1.835000e+03  4.000000e+00  9.730180e+08  1.835000e+03\n",
       "75%    4.476000e+03  2.770000e+03  4.000000e+00  9.752209e+08  2.770000e+03\n",
       "max    6.040000e+03  3.952000e+03  5.000000e+00  1.046455e+09  3.952000e+03"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ratings_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:35:03.779537Z",
     "iopub.status.busy": "2023-02-16T15:35:03.779180Z",
     "iopub.status.idle": "2023-02-16T15:35:04.094822Z",
     "shell.execute_reply": "2023-02-16T15:35:04.093890Z",
     "shell.execute_reply.started": "2023-02-16T15:35:03.779493Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:ylabel='Frequency'>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ratings_df.rating.plot.hist(bins=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:35:10.295936Z",
     "iopub.status.busy": "2023-02-16T15:35:10.295018Z",
     "iopub.status.idle": "2023-02-16T15:35:10.344407Z",
     "shell.execute_reply": "2023-02-16T15:35:10.343486Z",
     "shell.execute_reply.started": "2023-02-16T15:35:10.295900Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>#watch</th>\n",
       "      <th>avg_rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Toy Story (1995)</td>\n",
       "      <td>2077</td>\n",
       "      <td>4.146846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Jumanji (1995)</td>\n",
       "      <td>701</td>\n",
       "      <td>3.201141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Grumpier Old Men (1995)</td>\n",
       "      <td>478</td>\n",
       "      <td>3.016736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Waiting to Exhale (1995)</td>\n",
       "      <td>170</td>\n",
       "      <td>2.729412</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Father of the Bride Part II (1995)</td>\n",
       "      <td>296</td>\n",
       "      <td>3.006757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                title  #watch  avg_rating\n",
       "0                    Toy Story (1995)    2077    4.146846\n",
       "1                      Jumanji (1995)     701    3.201141\n",
       "2             Grumpier Old Men (1995)     478    3.016736\n",
       "3            Waiting to Exhale (1995)     170    2.729412\n",
       "4  Father of the Bride Part II (1995)     296    3.006757"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "movie_stats = ratings_df.groupby(\"movie_id\",as_index=False).aggregate(\n",
    "    {'user_id' : 'count', 'rating' : 'mean'}).rename(columns={'user_id' : '#watch', 'rating' : 'avg_rating'})\n",
    "\n",
    "movie_stats = pd.merge(movie_stats,movies_df,how='left',left_on='movie_id',right_on='id')[['title','#watch','avg_rating']]\n",
    "movie_stats.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:35:16.349966Z",
     "iopub.status.busy": "2023-02-16T15:35:16.349371Z",
     "iopub.status.idle": "2023-02-16T15:35:16.821311Z",
     "shell.execute_reply": "2023-02-16T15:35:16.820239Z",
     "shell.execute_reply.started": "2023-02-16T15:35:16.349923Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0, 0, '3428'),\n",
       " Text(0, 0, '2991'),\n",
       " Text(0, 0, '2990'),\n",
       " Text(0, 0, '2883'),\n",
       " Text(0, 0, '2672'),\n",
       " Text(0, 0, '2653'),\n",
       " Text(0, 0, '2649'),\n",
       " Text(0, 0, '2590'),\n",
       " Text(0, 0, '2583'),\n",
       " Text(0, 0, '2578')]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# top 10 most popular movies\n",
    "ax = movie_stats.sort_values(by=['#watch'],ascending=False)[:10].plot(x='title',y='#watch',kind='bar',figsize=(12,6),legend=False,ylabel='#watch')\n",
    "ax.bar_label(ax.containers[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:35:25.338509Z",
     "iopub.status.busy": "2023-02-16T15:35:25.338157Z",
     "iopub.status.idle": "2023-02-16T15:35:25.731311Z",
     "shell.execute_reply": "2023-02-16T15:35:25.730399Z",
     "shell.execute_reply.started": "2023-02-16T15:35:25.338479Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0, 0, '4.55456'),\n",
       " Text(0, 0, '4.52497'),\n",
       " Text(0, 0, '4.51711'),\n",
       " Text(0, 0, '4.51042'),\n",
       " Text(0, 0, '4.47772'),\n",
       " Text(0, 0, '4.47619'),\n",
       " Text(0, 0, '4.45369'),\n",
       " Text(0, 0, '4.44989'),\n",
       " Text(0, 0, '4.41282'),\n",
       " Text(0, 0, '4.40626')]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# top 10 most liked movies\n",
    "ax = movie_stats[movie_stats['#watch'] >= 1000].sort_values(by=['avg_rating'],ascending=False)[:10].plot(x='title',y='avg_rating',kind='bar',figsize=(12,6),legend=False,ylabel='avg-rating')\n",
    "ax.bar_label(ax.containers[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:35:39.458056Z",
     "iopub.status.busy": "2023-02-16T15:35:39.457644Z",
     "iopub.status.idle": "2023-02-16T15:35:39.495547Z",
     "shell.execute_reply": "2023-02-16T15:35:39.494603Z",
     "shell.execute_reply.started": "2023-02-16T15:35:39.458021Z"
    }
   },
   "outputs": [],
   "source": [
    "# ratings are converted to binary since we will be working on ctr or like prediction\n",
    "# 3 and below ratings are considered as dislike, 4 and 5 are considered as like\n",
    "ratings_df.loc[ratings_df.rating <= 3,'rating'] = 0\n",
    "ratings_df.loc[ratings_df.rating > 3,'rating'] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:35:42.651237Z",
     "iopub.status.busy": "2023-02-16T15:35:42.650804Z",
     "iopub.status.idle": "2023-02-16T15:35:42.676112Z",
     "shell.execute_reply": "2023-02-16T15:35:42.675112Z",
     "shell.execute_reply.started": "2023-02-16T15:35:42.651199Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    575281\n",
       "0    424928\n",
       "Name: rating, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ratings_df['rating'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:00.014660Z",
     "iopub.status.busy": "2023-02-16T15:36:00.014283Z",
     "iopub.status.idle": "2023-02-16T15:36:00.019572Z",
     "shell.execute_reply": "2023-02-16T15:36:00.018383Z",
     "shell.execute_reply.started": "2023-02-16T15:36:00.014629Z"
    }
   },
   "outputs": [],
   "source": [
    "# define label encoders for movies and users\n",
    "user_encoder = preprocessing.LabelEncoder()\n",
    "movie_encoder = preprocessing.LabelEncoder()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:00.300795Z",
     "iopub.status.busy": "2023-02-16T15:36:00.299918Z",
     "iopub.status.idle": "2023-02-16T15:36:00.420002Z",
     "shell.execute_reply": "2023-02-16T15:36:00.419010Z",
     "shell.execute_reply.started": "2023-02-16T15:36:00.300751Z"
    }
   },
   "outputs": [],
   "source": [
    "# make sure that all user and movie ids start from 1\n",
    "ratings_df.user_id = user_encoder.fit_transform(ratings_df.user_id.values)\n",
    "ratings_df.movie_id = movie_encoder.fit_transform(ratings_df.movie_id.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:02.994091Z",
     "iopub.status.busy": "2023-02-16T15:36:02.993407Z",
     "iopub.status.idle": "2023-02-16T15:36:03.019544Z",
     "shell.execute_reply": "2023-02-16T15:36:03.018515Z",
     "shell.execute_reply.started": "2023-02-16T15:36:02.994056Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>movie_id</th>\n",
       "      <th>like</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1104</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>639</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>853</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>3177</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>2162</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id  movie_id  like\n",
       "0        0      1104     1\n",
       "1        0       639     0\n",
       "2        0       853     0\n",
       "3        0      3177     1\n",
       "4        0      2162     1"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = ratings_df[['user_id','movie_id','rating']].rename(columns={'rating' : 'like'})\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3> Model Training and Evaluation </h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:10.614193Z",
     "iopub.status.busy": "2023-02-16T15:36:10.613778Z",
     "iopub.status.idle": "2023-02-16T15:36:10.628793Z",
     "shell.execute_reply": "2023-02-16T15:36:10.627790Z",
     "shell.execute_reply.started": "2023-02-16T15:36:10.614161Z"
    }
   },
   "outputs": [],
   "source": [
    "class MovieLensDataset(Dataset):\n",
    "    \n",
    "    def __init__(self, users, movies, ratings):\n",
    "        self.users = users\n",
    "        self.movies = movies\n",
    "        self.ratings = ratings\n",
    "        \n",
    "    def __len__(self):\n",
    "        return len(self.users)\n",
    "    \n",
    "    def __getitem__(self,idx):\n",
    "        user = self.users[idx]\n",
    "        movie = self.movies[idx]\n",
    "        rating = self.ratings[idx]\n",
    "        \n",
    "        return {\n",
    "               \"users\" : torch.tensor(user), \n",
    "               \"movies\" : torch.tensor(movie), \n",
    "               \"ratings\" : torch.tensor(rating)\n",
    "                }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:15.036678Z",
     "iopub.status.busy": "2023-02-16T15:36:15.036300Z",
     "iopub.status.idle": "2023-02-16T15:36:15.044319Z",
     "shell.execute_reply": "2023-02-16T15:36:15.043355Z",
     "shell.execute_reply.started": "2023-02-16T15:36:15.036648Z"
    }
   },
   "outputs": [],
   "source": [
    "class CollaborativeFilteringModel(nn.Module):\n",
    "    def __init__(self,n_users,n_movies,emb_dim=32):\n",
    "        super().__init__()\n",
    "        self.user_embeddings = nn.Embedding(n_users,emb_dim)\n",
    "        self.movie_embeddings = nn.Embedding(n_movies,emb_dim)\n",
    "        \n",
    "        self.cosine_sim = nn.CosineSimilarity(dim=1, eps=1e-6)\n",
    "        \n",
    "    def forward(self,user,movie):\n",
    "        user_emb = self.user_embeddings(user)\n",
    "        movie_emb = self.movie_embeddings(movie)\n",
    "        \n",
    "        # apply cosine similarity to user-movie embeddings and scale it between 0-1\n",
    "        # if user and movie embeddings are similar, that means higher chance to user to like the movie\n",
    "        output = (self.cosine_sim(user_emb,movie_emb) + 1) / 2\n",
    "\n",
    "        return output "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:15.221724Z",
     "iopub.status.busy": "2023-02-16T15:36:15.220917Z",
     "iopub.status.idle": "2023-02-16T15:36:15.298949Z",
     "shell.execute_reply": "2023-02-16T15:36:15.298024Z",
     "shell.execute_reply.started": "2023-02-16T15:36:15.221668Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "device(type='cuda', index=0)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "device"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:21.871489Z",
     "iopub.status.busy": "2023-02-16T15:36:21.871124Z",
     "iopub.status.idle": "2023-02-16T15:36:22.250256Z",
     "shell.execute_reply": "2023-02-16T15:36:22.249268Z",
     "shell.execute_reply.started": "2023-02-16T15:36:21.871458Z"
    }
   },
   "outputs": [],
   "source": [
    "train_data, test_data = train_test_split(df, test_size=0.2, random_state=42, stratify=df.like.values)\n",
    "test_data, val_data = train_test_split(test_data, test_size=0.5, random_state=42, stratify=test_data.like.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:22.252648Z",
     "iopub.status.busy": "2023-02-16T15:36:22.252248Z",
     "iopub.status.idle": "2023-02-16T15:36:22.260006Z",
     "shell.execute_reply": "2023-02-16T15:36:22.258890Z",
     "shell.execute_reply.started": "2023-02-16T15:36:22.252610Z"
    }
   },
   "outputs": [],
   "source": [
    "train_dataset = MovieLensDataset(train_data.user_id.values, train_data.movie_id.values, train_data.like.values)\n",
    "val_dataset = MovieLensDataset(val_data.user_id.values, val_data.movie_id.values, val_data.like.values)\n",
    "test_dataset = MovieLensDataset(test_data.user_id.values, test_data.movie_id.values, test_data.like.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:27.992677Z",
     "iopub.status.busy": "2023-02-16T15:36:27.992302Z",
     "iopub.status.idle": "2023-02-16T15:36:27.998100Z",
     "shell.execute_reply": "2023-02-16T15:36:27.997050Z",
     "shell.execute_reply.started": "2023-02-16T15:36:27.992647Z"
    }
   },
   "outputs": [],
   "source": [
    "train_dl = DataLoader(train_dataset, batch_size=1024,shuffle=True)\n",
    "val_dl = DataLoader(val_dataset, batch_size=1024,shuffle=True)\n",
    "test_dl = DataLoader(test_dataset, batch_size=1024,shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:33.993131Z",
     "iopub.status.busy": "2023-02-16T15:36:33.992478Z",
     "iopub.status.idle": "2023-02-16T15:36:34.011531Z",
     "shell.execute_reply": "2023-02-16T15:36:34.010650Z",
     "shell.execute_reply.started": "2023-02-16T15:36:33.993096Z"
    }
   },
   "outputs": [],
   "source": [
    "user_count = len(ratings_df['user_id'].unique())\n",
    "movie_count = len(ratings_df['movie_id'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:34.311914Z",
     "iopub.status.busy": "2023-02-16T15:36:34.311243Z",
     "iopub.status.idle": "2023-02-16T15:36:37.321987Z",
     "shell.execute_reply": "2023-02-16T15:36:37.321115Z",
     "shell.execute_reply.started": "2023-02-16T15:36:34.311876Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CollaborativeFilteringModel(\n",
       "  (user_embeddings): Embedding(6040, 64)\n",
       "  (movie_embeddings): Embedding(3706, 64)\n",
       "  (cosine_sim): CosineSimilarity()\n",
       ")"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = CollaborativeFilteringModel(user_count,movie_count,emb_dim=64)\n",
    "model.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:48.238055Z",
     "iopub.status.busy": "2023-02-16T15:36:48.237690Z",
     "iopub.status.idle": "2023-02-16T15:36:48.244237Z",
     "shell.execute_reply": "2023-02-16T15:36:48.243142Z",
     "shell.execute_reply.started": "2023-02-16T15:36:48.238026Z"
    }
   },
   "outputs": [],
   "source": [
    "optimizer = torch.optim.Adam(model.parameters(),lr=0.1)\n",
    "loss_function = nn.BCELoss()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:36:51.110012Z",
     "iopub.status.busy": "2023-02-16T15:36:51.109656Z",
     "iopub.status.idle": "2023-02-16T15:37:58.404614Z",
     "shell.execute_reply": "2023-02-16T15:37:58.403571Z",
     "shell.execute_reply.started": "2023-02-16T15:36:51.109983Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1 - Training Loss: 0.6008950721882188, Validation Loss: 0.5400366412133587\n",
      "Epoch 2 - Training Loss: 0.5150530220526258, Validation Loss: 0.529147284979723\n",
      "Epoch 3 - Training Loss: 0.5040982698692995, Validation Loss: 0.5262442833306838\n",
      "Epoch 4 - Training Loss: 0.499428129295254, Validation Loss: 0.5240477764484833\n",
      "Epoch 5 - Training Loss: 0.4966589735292108, Validation Loss: 0.5233470359627082\n"
     ]
    }
   ],
   "source": [
    "epochs = 5\n",
    "training_losses = []\n",
    "validation_losses = []\n",
    "for epoch in range(epochs):\n",
    "\n",
    "    train_loss = 0.0\n",
    "    \n",
    "    model.train()\n",
    "    # training loop\n",
    "    for i, data in enumerate(train_dl):\n",
    "        users = data['users'].to(device)\n",
    "        movies = data['movies'].to(device)\n",
    "        \n",
    "        ratings = data['ratings'].to(device)\n",
    "        ratings = ratings.view(ratings.shape[0],-1).to(torch.float32)\n",
    "\n",
    "        optimizer.zero_grad()\n",
    "\n",
    "        outputs = model(users,movies)\n",
    "        loss = loss_function(outputs, ratings.squeeze())\n",
    "        train_loss += loss.item()\n",
    "\n",
    "        loss.backward()\n",
    "\n",
    "        optimizer.step()\n",
    "       \n",
    "    train_loss /= (i+1)\n",
    "    training_losses.append(train_loss)\n",
    "    \n",
    "    model.eval()\n",
    "    val_loss = 0.0\n",
    "    for i, data in enumerate(val_dl):\n",
    "        users = data['users'].to(device)\n",
    "        movies = data['movies'].to(device)\n",
    "        \n",
    "        ratings = data['ratings'].to(device)\n",
    "        ratings = ratings.view(ratings.shape[0],-1).to(torch.float32)\n",
    "        \n",
    "        outputs = model(users,movies)\n",
    "        loss = loss_function(outputs, ratings.squeeze())\n",
    "        val_loss += loss.item()\n",
    "\n",
    "    val_loss /= (i+1) \n",
    "    validation_losses.append(val_loss)\n",
    "    print(f\"Epoch {epoch+1} - Training Loss: {train_loss}, Validation Loss: {val_loss}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:39:57.153695Z",
     "iopub.status.busy": "2023-02-16T15:39:57.153298Z",
     "iopub.status.idle": "2023-02-16T15:39:57.386383Z",
     "shell.execute_reply": "2023-02-16T15:39:57.385487Z",
     "shell.execute_reply.started": "2023-02-16T15:39:57.153656Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(epochs),training_losses,label='Training loss')\n",
    "plt.plot(range(epochs),validation_losses,label='Validation loss')\n",
    "plt.xlabel(\"Epochs\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.title(\"Collaborative Filtering Loss\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:38:07.450548Z",
     "iopub.status.busy": "2023-02-16T15:38:07.449490Z",
     "iopub.status.idle": "2023-02-16T15:38:09.119067Z",
     "shell.execute_reply": "2023-02-16T15:38:09.118007Z",
     "shell.execute_reply.started": "2023-02-16T15:38:07.450476Z"
    }
   },
   "outputs": [],
   "source": [
    "predictions = []\n",
    "targets = []\n",
    "\n",
    "# evaluate the model on the test set\n",
    "with torch.no_grad():\n",
    "    for i, data in enumerate(test_dl):\n",
    "        users = data['users'].to(device)\n",
    "        movies = data['movies'].to(device)\n",
    "\n",
    "        ratings = data['ratings'].to(device)\n",
    "        ratings = ratings.view(ratings.shape[0],-1).to(torch.float32)\n",
    "        targets.extend(ratings.squeeze(dim=1).tolist())\n",
    "\n",
    "        outputs = model(users,movies)\n",
    "        predictions.extend(outputs.tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:38:12.267747Z",
     "iopub.status.busy": "2023-02-16T15:38:12.267388Z",
     "iopub.status.idle": "2023-02-16T15:38:12.287031Z",
     "shell.execute_reply": "2023-02-16T15:38:12.284896Z",
     "shell.execute_reply.started": "2023-02-16T15:38:12.267716Z"
    }
   },
   "outputs": [],
   "source": [
    "predictions = np.array(predictions)\n",
    "targets = np.array(targets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:41:34.299691Z",
     "iopub.status.busy": "2023-02-16T15:41:34.299316Z",
     "iopub.status.idle": "2023-02-16T15:41:34.400502Z",
     "shell.execute_reply": "2023-02-16T15:41:34.399391Z",
     "shell.execute_reply.started": "2023-02-16T15:41:34.299661Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision: 0.7470013466537221\n",
      "Recall: 0.829248366013072\n",
      "Accuracy: 0.7402545465452255\n"
     ]
    }
   ],
   "source": [
    "precision = precision_score(targets,np.where(predictions >= 0.5,1,0))\n",
    "recall = recall_score(targets,np.where(predictions >= 0.5,1,0))\n",
    "accuracy = accuracy_score(targets,np.where(predictions >= 0.5,1,0))\n",
    "\n",
    "print(f\"Precision: {precision}\\nRecall: {recall}\\nAccuracy: {accuracy}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:38:24.531555Z",
     "iopub.status.busy": "2023-02-16T15:38:24.531180Z",
     "iopub.status.idle": "2023-02-16T15:38:24.735169Z",
     "shell.execute_reply": "2023-02-16T15:38:24.734295Z",
     "shell.execute_reply.started": "2023-02-16T15:38:24.531505Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEWCAYAAAB42tAoAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAAsTAAALEwEAmpwYAABCuElEQVR4nO3dd3gUVffA8e9Jh5BACBAgdOm9hCIKgghSVFQQsAIWQFF/ihWsYANFFEVBXlBeXxQLKIhiARQBEem9E1qoIUB63/v7YxaMMQlLyGayyfk8zz7Z2WlnJsmcvffO3CvGGJRSSqnceNkdgFJKqaJNE4VSSqk8aaJQSimVJ00USiml8qSJQimlVJ40USillMqTJgqllFJ50kSh/kVElonIWRHxz+Hz+7N91kVEorJMi4g8KiLbRCRRRKJE5GsRaZbLvvxF5GMRiROREyIy6iKxPSIiB5zLrxORq7PMGyAiq0QkSUSW5bDudBHZLSIOERmSbZ6IyKsiclREYp3H2iTL/Foissh5Xk6IyBQR8XHO6yAii0XkjIhEO4+3SpZ1fxSRhCyvNBHZmmX+KyKyVUQyROTlbHH1EZGVInLOud//iEiQq+fvIsc8LVtcqSISn2V+QrZXpoi875znJyJzReSgiBgR6ZJt27kes4jUyGHbRkSe+PdvXBUFmijUP4hILaATYICb8rGJycD/AY8C5YH6wHygTy7LvwzUA2oCXYGnRaRnLrG1B8YD/YGywEzgWxHxdi5yBnjXuUxONgMPARtymHcbcC/WsZcH/gT+l2X+h8ApoArQErjGuS2AEGA6UMt5HPHAJ+dXNMb0MsaUOf8CVgFfZ9n2PuBp4Icc4ioLvApUBRoB1YC3ssx/mbzPX67HbIwZkS2uOVnjyjYvDEjOFvdK4C7gRA7bzvWYjTGHs81rBjiAeTkcvyoKjDH60teFF/Ai8AcwCfg+27xlwP3ZPusCRDnf1wMygXaXsL+jQI8s068AX+Sy7EBgTZbpQKyEViXbcvcDy/LY50pgSLbPngG+yjLdBEjJMr0T6J1l+i3go1y23xqIz2VeLec5qp3DvNnAyxc5X7cCWy/1/OV0zNnmB2IluGtymT8YiAQkh3lRQJc8tp3rMTvnvwT8Vph/5/q6tJeWKFR29wCfOV/Xi0jYJazbDStprHFlYREJwfqmvDnLx5uxLtI5+RHwFpH2zlLEvcAmcvhGmw9fAHVFpL6I+GJdGH/KMn8yMEhESotIONAr2/ysOgPbc5l3D7DCGHMgn3Fe2HY+zl9e+gHRwPJc5g8GPjXOK/slutgx3wP8Nx/bVYXEx+4AVNHhrO+vifXN+rSI7AfuAN5xcROhwPFL2GUZ58/YLJ/FAkE5LAvWN955WN+OBTgH9MrnxSu748AKYDfWt98jwLVZ5v8OPADEAd5YF7b52TciIs2xSmV9c9nPPVhVSZdMRLpjXbDbOz+61POXl1wTgYjUwKpquy8f24U8jllEOmFVa83N57ZVIdAShcpqMPCLMea0c/pz52fnZQC+2dbxBdKd72Ow6vBdleD8GZzls2CshJCT+7FKEU0AP6z68e9FpOol7DM3LwFtgepAADAW+NVZgvACfga+waqiqYDVLjEh6wZEpC5Wqef/jDErsu/AmYgrk4+Looh0wPp99DfG7HF+fKnnL7dtV8dKBJ/mssg9wMr8lIJcOObBwDxjTEIu81URoIlCASAipYABwDXOu2dOAI8DLUSkhXOxw1j1zVnVBg453y8FqolIhCv7NMacxfom3yLLxy3IvdqmBbDQGLPHGOMwxvzkXL+jK/u7iBbAl8aYKGNMhjFmFlYyaIzVuF0dmGKMSTXGxGA1Vvc+v7KI1ASWAK8YY/73r61bBgPfXOpFUURaAd8B9xpjlp7/PB/nLzf3AKuMMZF5zM9v1VCux+z8m7vtMratCokmCnXezVhVLo2x7uppiXWXzQqsCwXAl8BQEWnnvJ20PlYy+QLAGLMX6+6gOWLdNusnIgEiMkhEns1lv58Cz4tIiIg0xKremZXLsmuBPiJSx7n/7lh3VW0DEBFvEQnAqlL1cu77QgnofDxY1Va+zvleWbZ9m4iEiYiXiNyNVVra5yxhHQAeFBEfESmHdQHc7NxuOPAr8IExZlpOgWe5KP7r2ETE1xmXF+DjjMvbOa8pVlvII8aYhZd6/i5yzOfdk1NczvU7AuH8826n8/P8ndsGOP+7FleO2ekWrOrD33KZr4oKu1vT9VU0XlgXo7dz+HwAVmOxj3P6XqxvrHFYt3U+C3hlWV6wbo/dDiRh3ZXzJdAkl/36Ax87t3cSGJVtfgLQKcu2x2GVbOKx7kS6O8uyQ7Dugsr6mpVl/rIc5ndxzgsAPsD6hh6HdTtpzyzrtnSufxY4jXXhrOSc95JzWwlZX9mO43askldOdw3NyiGuIc55n2DdOpp129sv4fzleszO+VcCiUBQLr+fj4D/5TLvYA7bruXKMTvn/4xVArP9719feb/E+QtTSimlcqRVT0oppfKkiUIppVSeNFEopZTKkyYKpZRSefK4J7MrVKhgatWqZXcYSinlUdavX3/aGFMxP+t6XKKoVasW69atszsMpZTyKCJy6OJL5UyrnpRSSuVJE4VSSqk8aaJQSimVJ00USiml8qSJQimlVJ40USillMqT2xKFiHwsIqdEZFsu80VE3hORfSKyRURauysWpZRS+efOEsUsoGce83sB9ZyvYcBUN8ailFIlhjGGTIchPdNBakYmKemZl7U9tz1wZ4xZLiK18likL3+P0btaRMqJSBVjzKWMuayUUkVCeqaDpNRMktMzSUrLICnNukCnpDtITMsgJT2T6PhUjAGHMaRmODhyJomypXxJy3SQnungUEwSIYF+pGU4/n5lOsjIdJCeaV34MxyGtAzHhffpmQ4yMg3JuSSDlMNbiVu34LKOzc4ns8OxBrA/L8r52b8ShYgMwyp1UKNGjUIJTilVchhjXWjPJaWTkJpBfEo6cckZxKWkE5eSQWxSGueS0jmblE5scjoJqekkpmYSl5JOfIq1fHpm/sf2KVfaF19vL/y8vdh4+Bw1Q0vj7+OFn48XAb5e+Pj74Ost+Hh54evjha+X4OvthY+39dPX2xpYMCE1g4pBAXgJJMWeZcH0Caz55VtCK1cj+TLOj52JQnL4LMczbYyZDkwHiIiI0JGWlFIXlZKeydFzyZyMS2H/qQRSMxzEJKYRm5xObFI6MYmpxCZncDYxjbNJaaRmOPLcXmk/b8qW8qVsKV+CA3ypUMaP2hUCCQrwoYy/9TLO5ULL+FHK15vSfj6U8vOmlK83pfy8CfD1xt/Hi1LOnz7e7qv979evHxt+Xcjo0aN5/vnnCQwMzPe27EwUUVgD1p9XDThmUyxKqSLMGMPZpHSi41M5EZdy4eJ+NimduOR0ziWlsfN4PL4+wjnnZ/GpGWQfwNPXWwgO8KVcaV9CSvsRXq4UTasGExLoR0hpP8qV9r1w4S9bynof7EwMAb7e9hz8Jdi+fTvlypUjPDycCRMmMG7cOJo0aXLZ27UzUXwHPCwiXwDtgVhtn1Cq5DHGcCYxjeOxKZxNsn6ejE0hOiGV6PhUjsWmEHkqgfjUjH+tKwJB/tbFPDTQjwyHoW2t8pQtZSWDmqGlCQsKoJwzKQSX8kEkp8oMz5aYmMgrr7zC22+/zZ133smsWbOoW7dugW3fbYlCROYAXYAKIhKFNQC9L4AxZhqwCOgN7AOSgKHuikUpZR+Hw3AqPpVjscnsPRlPTGIaR88mcyo+laizyUSdTSI+5d9JoGwpq3qnctkAbmkdTs3QQCoF+RMWHED5QD9CSvtSrrQf3l7F78J/KX744QdGjhzJoUOHuPfee5kwYUKB78Oddz3dfpH5Bhjprv0rpQpPfEo6h2KS2B+dwIHTiRyOSeLouWROxKVw9GwyGY5/1gGVK+1LpSB/qoWUJqJmCLUqBBJeLoCQ0lZiCAsO8IiqHrt9+OGHjBw5ksaNG7N8+XI6derklv143HgUSil7pGU4OBiTyM7jcRw4nUjU2WQOxyRxMCaRU/Gp/1i2StkAwsuVoll4WXo3q0J4uVJUKRtA5bIBVC9fmuAAX5uOwvNlZGQQHR1NlSpVGDBgAMnJyTzyyCP4+fm5bZ9isrf2FHERERFGBy5Syn3SMx3sPhHPrhPx7D4Rx47jcew/lcjJ+JR/NA5XCvKnRvnS1KoQSJ2KgdQKDaS2872/j5YG3GHNmjUMHz4cHx8fVq9ejbe36+dZRNYbYyLys18tUShVQjkchsjTCew9mcCWo7HsPRlPZHQiR84mXXgmwM/HiwZhQXS8IpTq5UtTu0Ig9cLKULdSGU0GhejcuXOMGTOGadOmUaVKFSZPnoyXV+F11aeJQqkSID3TwYHTVrXRjuNxrD94lh3H40hKs57m9fUWaoUGUj8siOubVqZh5SCahpelZvnSbr3XX13c1q1b6d69O9HR0Tz66KOMGzeO4ODgQo1BE4VSxUxMQioHTiey60Q8247GsuN4HLtOxJPmfKDM11toXLUsAyKq06hKEDVDA2lVo5yWEIqY9PR0fH19qV+/Pl27duWpp56idWt7+k7VRKGUB0vPdLD9WBy7jsex7Vgsq/bHEBmdeGF+cIAPzauV454ONWkSHkyDsGDqhZXBV0sJRVZqaioTJkxg9uzZbNiwgTJlyjBnzhxbY9JEoZSHMMZw5Ewym6LOseHQWTZHnWP3ifgL1UeBft60rV2e29pUp0HlMjSoHEyV4AC8SvhzBp7k119/5cEHH2TPnj0MHDiQ1NRUypQpY3dYmiiUKqocDsOuE/FsjjrH2oNnWL0/hmOxKYDVyNyyejkGRFSnba3yNA0PpnpIaU0KHio5OZlhw4Yxe/Zs6tSpw08//cT1119vd1gXaKJQqogwxrDjeBwr9p5m69FY1h08w8k46/mEkNK+tK1Vnge7XEGrGiHUDwvCz0erj4qLgIAATp8+zfPPP8+YMWMoVaqU3SH9gyYKpWyUkp7J0p2nWLE3mpX7ThN11uoMulpIKdrUDKFrg0q0rVWemqGli2UfRSXZli1beOqpp5g5cybVqlXjhx9+KNRbXi+FJgqlClFGpoMNh62qpOV7otl05BypGQ6CAnxoXzuUh7vW5dpGlagUFGB3qMpNEhMTefnll3nnnXcICQlh7969VKtWrcgmCdBEoZTbHTmTxLLdp1i1P4bVkTGcTUoHoFGVYO5sX5MuDSrS8YpQfV6hBPjuu+945JFHOHz4MA888ADjx4+nfPnydod1UZoolCpgqRmZrD1wlp+2H+f3PdEcOWNVJ1UODqBrw0pc1yiMK+uEEhLovr55VNE0f/58goODWblyJVdddZXd4bhME4VSBeBwTBJLd53kj32n+SvyDPGpGQT4etG5XkWGdKzNtQ0rUUvbGUqc9PR03nvvPbp27Urr1q2ZPHkyAQEB+Pp6VqeImiiUyqcTsSnM2xDFtxuPsu9UAgA1ypemV7PKdG9cmavqhlLaT//FSqrVq1czfPhwtmzZwjPPPEPr1q0JCgqyO6x80b9ipVxkjOHwmSR+2naCP/bH8Me+02Q6DBE1Q3jhhsZ0aVCROhUCtdRQwp09e5bRo0czffp0wsPD+fbbb+nbt6/dYV0WTRRKXcS2o7Es2nqcJTtPsuekVXKoUyGQBzrVoX+bcOpW8sxvico9pk+fzowZM3j88cd5+eWXPbYUkZWOR6FUDqLjU/l2YxQLNh1j+7E4fLyENjVDuL5JZau9oUKg3SGqImT37t1ER0dz9dVXk5qayu7du2nevLndYf2DjkehVAGITUrn5x0nmLs+irUHz2AMtKxejuf7NOLW1tUor3cpqWxSUlJ44403GD9+PA0bNmTTpk34+/sXuSRxuTRRqBItOj6Vn7Yd54etx1l78CyZDkOt0NI83LUufZpXoWHlwu33X3mOxYsX89BDD7Fv3z7uuOMO3n777WLbPqWJQpU4Dodh3aGzfLzyAIt3niTTYahTMZDhnetwXeMwWlUvV2z/4VXBWL58OT169KBevXosXryY6667zu6Q3EoThSox9kcn8PW6KOZvPMqJuBSC/H14oFMdbmkVTv2wMpocVJ4yMzPZsWMHzZo1o1OnTsycOZM77riDgIDi392KJgpVrO07lcDSnSf5ZcdJ1h86i7eX0LleBUb3bsi1DSsRFOBZDz4pe2zcuJERI0awc+dO9u7dS1hYGPfee6/dYRUaTRSq2DkZl8KcNYf5fsvxCw/CNa4SzJM96jOgbXXtcE+5LD4+npdeeonJkydToUIFpk6dSqVKlewOq9BpolDFgjGGjUfO8Z/lkfyy4yQOY2hbqzxjb2pCt0aVqBZS2u4QlYeJjY2lWbNmHDlyhOHDh/PGG28QEhJid1i20EShPNqp+BR+3HqC//55kMjoRMr4+3D/1bW5vV0NfdZB5UtcXBzBwcGULVuWYcOG0a1bN6688kq7w7KVPnCnPNLRc8n8d9VBZv1xkLRMB42rBDOkYy2ub1KZsqW13UFduvT0dN555x1effVVli1bRuvWre0OqUDpA3eqRDDG8GdkDF+uPcKPW0+QaQy9m1Xh4a519a4ldVn++OMPRowYwbZt27j55pupWLGi3SEVKZooVJHncBj+OnCGN3/excbD5yhbypcBbasx4portO1BXbZHHnmEKVOmUL16dRYsWMBNN91kd0hFjiYKVWQ5HIZfdpzg7V/2sPdUAhWD/Hn15qb0a12NUn7edoenPJgx5kIJtHLlyjz55JO89NJLlClTxubIiiZNFKrIOZeUxhdrj/D5X4c5fCaJupXK8Gb/5vRqWlmfe1CXbdeuXYwYMYLHH3+cvn378txzz9kdUpGniUIVGeeS0pi6bD//W32IpLRM2tUuz2PX1ePGFlXx1fGk1WVKTk7m9ddfZ8KECQQGBpKcnGx3SB7DrYlCRHoCkwFvYIYxZny2+WWB2UANZywTjTGfuDMmVfQcO5fMR7/vZ+76KJLSM+ndrAr3X12bVjVK5j3rquAtXbqU4cOHs3//fu6++24mTpxYIh+cyy+3JQoR8QY+ALoDUcBaEfnOGLMjy2IjgR3GmBtFpCKwW0Q+M8akuSsuVXQcPJ3ItN/3s2DTMdIyHfRuVoWRXa/QHltVgYuKisLHx4elS5dy7bXX2h2Ox3FniaIdsM8YEwkgIl8AfYGsicIAQWK1KpUBzgAZboxJFQFrD55h2rL9LN11Cj8fL25uWZWRXetSM1QfkFMFIzMzk2nTpuHn58cDDzzAPffcw6BBg/D397c7NI/kzkQRDhzJMh0FtM+2zBTgO+AYEAQMNMY4sm9IRIYBwwBq1KjhlmCVexlj3eL6zuI9/HXgDOUD/Xi0Wz3ual+DSsHa95IqOBs2bGD48OGsW7eOfv368cADDyAimiQugzsTRU5PP2V/DPx6YBNwLXAFsFhEVhhj4v6xkjHTgelgPZld8KEqdzHGsDry72cgKpTx47nejbizQw1K++m9FKrgxMXF8cILLzBlyhQqVqzInDlzGDhwoN1hFQvu/E+NAqpnma6GVXLIaigw3lj9iOwTkQNAQ2CNG+NShcDhMCzadpz/LI9kc1QsFYP8eeXmpvTXZyCUm2zevJkpU6YwYsQIXnvtNcqVK2d3SMWGOxPFWqCeiNQGjgKDgDuyLXMY6AasEJEwoAEQ6caYVCFYc+AMry3ayeYj56hdIVAThHKbAwcO8Ntvv3HvvffSqVMn9u3bR+3ate0Oq9hxW6IwxmSIyMPAz1i3x35sjNkuIiOc86cBrwCzRGQrVlXVM8aY0+6KSbnXidgU3vhxJws2HaNSkD9v9mtO/zbV8PLSPphUwUpLS+Ptt99m3LhxBAQEcMsttxASEqJJwk3cWklsjFkELMr22bQs748BPdwZg3K/6PhUJv68m283HcXhMDzQqTajujfQEoRyixUrVjBixAh27NjBrbfeyuTJk0vsOBGFRVsTVb6lpGcyc+UBpi7bT1qGg/4R1RjR+QpqhGpHfco9oqOj6dGjB2FhYSxcuJAbbrjB7pBKBE0U6pIZY5i7PopJi/dwPDaFbg0rMbp3I+pW0g7VVMEzxrBkyRK6d+9OxYoV+f777+nQoQOBgfrcTWHRRKEuSWR0AuO+38Gy3dE0DQ/mrf4tuLpeBbvDUsXU9u3befDBB1mxYgW//fYbXbp0oVu3bnaHVeJoolAucTgMk5fu5YPf9uHr7cVzvRtxf6faOliQcoukpCReffVV3nrrLYKDg5kxYwadO3e2O6wSSxOFuqiNh88y7vsdbDx8jptaVGVM70ZULqtPUyv3MMbQtWtX1qxZw+DBg3nrrbd0xDmbaaJQuToVl8KLC7bz0/YThAb68catzRjUtrqWIpRbHD9+nEqVKuHt7c2YMWMoW7YsXbp0sTssBWgn/ypHK/eepvd7K1iy8ySPXluX35/uyu3tamiSUAUuMzOT9957jwYNGvDhhx8C0LdvX00SRYiWKNQ/pGU4+Oj3/byzZA+1KgTy6b3taVxVu/1W7rFu3TqGDx/Ohg0buP766+ndu7fdIakcuJwoRCTQGJPozmCUvdYfOsNz325j14l4+jSrwoT+zSnjr98llHu8+eabPPvss1SuXJkvv/yS2267TUusRdRFq55EpKOI7AB2OqdbiMiHbo9MFZr4lHQmL9nLbdP+5ExiGtPuasMHd7bWJKEKnDGG9PR0ANq1a8fIkSPZuXMnAwYM0CRRhLlyJXgHqzvw7wCMMZtFRO9TKya+XneEcQt3EJ+aQZ9mVRjfrxlBAb52h6WKof379/PQQw/RtGlT3n77bbp06aLtEB7Cpa+Mxpgj2bJ9pnvCUYUlNjmd8T/uZM6aI7SvXZ7RvRvRsno5u8NSxVBqaipvvfUWr732Gr6+vvTt29fukNQlciVRHBGRjoARET/gUZzVUMozbY2KZfj/1nEsNoUhHWsxpncj/Hz0BjhV8NavX89dd93Frl27uO2223j33XepWrWq3WGpS+RKohgBTMYa2jQK+AV4yJ1BKfdwOAz/W32I1xftpEIZf+Y9eCVtapa3OyxVjJUpUwYRYdGiRfTq1cvucFQ+uZIoGhhj7sz6gYhcBfzhnpCUO8SlpDPqy80s2XmSq+qG8s6AljpWtSpwDoeDTz75hD///JMZM2bQoEEDtm3bhpeXllg9mSu/vfdd/EwVUftOJdB78gqW7T7F830aMfu+9pokVIHbtm0bnTt35v7772fv3r0kJlp302uS8Hy5lihE5EqgI1BRREZlmRWMNWKdKuKMMcxadZCJP+/Gy0v4/IEOtKutVU2qYCUmJjJu3DgmTZpE2bJl+eSTTxg8eLDe7lqM5FX15AeUcS4TlOXzOKC/O4NSl8/hMDw9bwtz10dxdd0KvHFrM6qX1wGFVMFLSUnhk08+4Z577uHNN98kNDTU7pBUAcs1URhjfgd+F5FZxphDhRiTukxxKek89+02Fm4+xrDOdRjdq6F+u1MFKioqivfee4833niD0NBQdu3aRfnyWlotrlxpzE4SkbeAJsCFim1jzLVui0rl22+7TvHMvC1EJ6TyaLd6PH5dPU0SqsBkZGTw/vvv8+KLL5KZmcnAgQNp06aNJolizpVWps+AXUBtYCxwEFjrxphUPpxJTOO+WWsZOmstIaX9+Hr4lYzqXl+ThCowf/31FxEREYwaNYrOnTuzfft22rRpY3dYqhC4UqIINcbMFJH/y1Id9bu7A1Ou23TkHCM/20B0fCoPdbmCR7vVI8BX7zdQBcfhcDB06FBiY2OZO3cut956q34JKUFcSRTpzp/HRaQPcAyo5r6Q1KVYHRnDfbPWUraUL1+NuFK74VAFxhjD3Llz6dmzJ0FBQXzzzTeEh4cTFBR08ZVVseJK1dOrIlIWeAJ4EpgBPObOoJRr5q2P4u6ZfxFWNkCThCpQe/fu5frrr2fAgAFMnz4dgIYNG2qSKKEuWqIwxnzvfBsLdIULT2Yrmzgchim/7WPS4j10vCKUqXe2oWxp7fFVXb7U1FQmTJjA66+/jr+/P1OmTGHEiBF2h6VsltcDd97AAKw+nn4yxmwTkRuAMUApoFXhhKiyOhWfwsOfb2TNgTPc2KIqE29rjr+PtkeogjFy5EhmzpzJoEGDmDRpElWqVLE7JFUEiDEm5xkis4DqwBqgPXAIuBJ41hgzv5Di+5eIiAizbt06u3Zvq6Pnkuk7ZSXxKRmM69uEARHVtUFRXbZTp07hcDioXLkye/fuJTIykuuvv97usFQBE5H1xpiI/KybV9VTBNDcGOMQkQDgNFDXGHMiPztSl+fYuWTu/M9qktIymfdgR5qGl7U7JOXhHA4HM2bM4JlnnqFHjx58+eWX1KtXj3r16tkdmipi8mrMTjPGOACMMSnAHk0S9th5PI6bP/iD6PhUZg1tp0lCXbYtW7Zw9dVXM3z4cFq2bMnYsWPtDkkVYXmVKBqKyBbnewGucE4LYIwxzd0enWLj4bPcO2stXiJ8OfxKTRLqss2dO5dBgwYREhLCp59+yl133aVVmCpPeSWKRoUWhcrRgk1HGfXVZioHB/C/+9pRp2IZu0NSHiwuLo7g4GC6dOnCyJEjeemll7TrDeWSvDoF1I4AbfTbrlM8+fVmmoWXZcbgCCqU8bc7JOWhDh8+zCOPPMKxY8dYvXo1FSpUYPLkyXaHpTyIW0cUEZGeIrJbRPaJyLO5LNNFRDaJyHbtGsTy7cYohv1vHVdULMN/722nSULlS3p6OhMnTqRRo0YsWbKEAQMGkNtdjkrlxZUuPPLF+RzGB0B3rLG214rId8aYHVmWKQd8CPQ0xhwWkUruisdTzFgRyas/7KR1jXLMHNyWsqX0QTp16Q4dOsRNN93Eli1buPHGG3n//fepWbOm3WEpD+VSohCRUkANY8zuS9h2O2CfMSbSuY0vgL7AjizL3AF8Y4w5DGCMOXUJ2y9WjDG8u2Qvk5fupVvDSnx4V2t9kE5dMmMMIkLlypUJCwvj22+/pW/fvtpYrS7LRaueRORGYBPwk3O6pYh858K2w4EjWaajnJ9lVR8IEZFlIrJeRO5xKepi6MNl+5m8dC+3tgpn2t1tNEmoS2KMYfbs2bRt25aEhAT8/f355ZdfuPnmmzVJqMvmShvFy1ilg3MAxphNQC0X1svprzN7BakP0AboA1wPvCAi9f+1IZFhIrJORNZFR0e7sGvPMn35ft76eTe9m1Vm4m0t8PXWweiV63bv3k23bt24++678fHxISYmxu6QVDHjyhUpwxgTm49tR2F1AXJeNawuyrMv85MxJtEYcxpYDrTIviFjzHRjTIQxJqJixYr5CKXomrnyAK8v2kXvZpWZPKgVXl767U+5JiMjg5deeonmzZuzYcMGpk6dyqpVq7QtQhU4VxLFNhG5A/AWkXoi8j6wyoX11gL1RKS2iPgBg4DsVVYLgE4i4iMipbH6lNp5CfF7LGMMr3y/g1e+30GPxmFMHtRKSxLqknh7e7NixQr69+/P7t27GTFiBF5e+jekCp4rf1WPYI2XnQp8jtXd+GMXW8kYkwE8DPyMdfH/yhizXURGiMgI5zI7sdo+tmB1PjjDGLMtH8fhcf676iAzVx7gzvY1mHpXG00SyiUnTpzg3nvv5ciRI4gIixYt4rPPPiMsLMzu0FQxlmvvsRcWEGlljNlYSPFcVHHoPXbF3miGfrKWq+pW4JMhbbW6SV1UZmYm06dPZ/To0SQnJzN79mxuu+02u8NSHuRyeo915WvsJBHZJSKviEiT/OxE/W1rVCzDPl1PjdDSvKdtEsoFGzdupGPHjjz00ENERESwdetWTRKqUF00URhjugJdgGhguohsFZHn3R1YcbQl6hy3/2c1IaV9mX1fex2VTrlkypQpHDx4kM8++4zFixdTv/6/bgxUyq0uWvX0j4VFmgFPAwONMX5uiyoPnlr1FBmdwO3/WY23CHMf7EjVcqXsDkkVUcYY5s+fT61atWjVqhVnz54FICQkxObIlCdza9WTiDQSkZdFZBswBeuOp2r52VlJFZuczpBP1pKeaZgxuK0mCZWrgwcPctNNN3Hrrbfy7rvvAlaC0CSh7ORKFx6fAHOAHsaY7M9BqItwOAxPfLWZo+eSmX1fexpXDbY7JFUEpaenM2nSJMaOHYuXlxcTJ07k//7v/+wOSynAhURhjOlQGIEUVxN+3sWSnSd5pmdDrrwi1O5wVBH10Ucf8eyzz3LzzTczefJkatSoYXdISl2Qa6IQka+MMQNEZCv/7HpDR7hz0Vdrj/DR75EMjKjOiGvq2B2OKmJiYmI4ePAgbdq04YEHHqBu3br07NnT7rCU+pe8ShTny703FEYgxc1P244z+tutXFU3lNduaaods6kLjDF8+umnPPnkkwQFBbFnzx78/f01SagiK9fGbGPMcefbh4wxh7K+gIcKJzzPtD86gSe/3kK9SmWYelcbfPSpa+W0c+dOunbtypAhQ6hXrx7z58/Hx8dtw8IoVSBcuYJ1z+GzXgUdSHFxLimNh2ZvwMdb+HhIW4ID9FkJZdm8eTMtWrRgy5YtTJ8+nZUrV9K8udbgqqIvrzaKB7FKDnVEZEuWWUHAH+4OzBM5HIaHPtvAvugEZtwTobfBKgCioqKoVq0azZs3Z+zYsdx3331UqlTiB3NUHiSvEsXnwI1YPb7emOXVxhhzVyHE5nGm/r6fVftjeL5PI7o21AtBSXfs2DEGDhxIo0aNOHr0KCLC6NGjNUkoj5NXojDGmIPASCA+ywsRKe/+0DzLH/tOM/EXa/ChIR1r2R2OslFmZiZTpkyhUaNGLFiwgKeffpoKFSrYHZZS+ZZXK9rnWHc8rce6PTbrbTsG0Ps9nbZGxTL8f+upUb404/s11zucSrCUlBQ6d+7M2rVr6d69Ox9++CF169a1OyylLkuuicIYc4PzZ+3CC8fz7DuVwJBP1lDG34c5D3TQxusSKj09HV9fXwICAujatSujRo1i4MCB+qVBFQuu9PV0lYgEOt/fJSKTREQfGwWOnUvm9v+sJtMY/ntvO228LoGMMcydO5e6deuyYcMGACZMmMCgQYM0Sahiw5XbY6cCSSLSAqvn2EPA/9walQdIy3Bw33/XkZyWyef3d6BB5SC7Q1KFLDIykj59+nDbbbcRGhqqw5CqYsuVv+wMY/VF3heYbIyZjHWLbIn2nxWR7Dwex8TbWmhHfyXQpEmTaNKkCStWrODdd99lzZo1tGzZ0u6wlHILVx4JjReR0cDdQCcR8QZKdEX8rhNxTPl1H9c1qkTPppXtDkfZICEhgd69ezN58mSqVdNe91Xx5kqJYiCQCtxrjDkBhANvuTWqIizTYRj15WYC/b157ZZmdoejCsnp06cZOnQo3333HQDPP/888+bN0yShSgRXhkI9AXwGlBWRG4AUY8ynbo+siHrr593sOB7HSzc2ISw4wO5wlJs5HA4+/vhjGjRowOzZs9m3bx+AtkeoEsWVu54GAGuA24ABwF8i0t/dgRVFi3ecZNrv+xkQUY0bmlexOxzlZjt27KBLly7cd999NG7cmE2bNjFq1Ci7w1Kq0LnSRvEc0NYYcwpARCoCS4C57gysqDkUk8gTX22iXqUyjOur3YaXBOvWrWP79u3MnDmTIUOGaClClViuJAqv80nCKQbX2jaKjfRMB09+vRmHgY/ubkOAr7fdISk3WbRoETExMdx9993cfffd3HDDDZQvrz3WqJLNlQv+TyLys4gMEZEhwA/AIveGVbRM+XUfaw+eZexNTahTsYzd4Sg3iIqKon///vTp04cpU6ZgjEFENEkohWuN2U8BHwHNgRbAdGPMM+4OrKjYfiyWD37bR59mVejXRu9wKW4yMjKYPHkyjRo14ocffuC1115jxYoVWrWoVBZ5jUdRD5gIXAFsBZ40xhwtrMCKAmMMLy7YTpkAH168sbHd4Sg3WL9+PY899hg9e/bkgw8+oE4d7etSqezyKlF8DHwP9MPqQfb9QomoCPl5+0nWHzrLY93q6a2wxUhsbCzffPMNAO3bt+evv/5i0aJFmiSUykVejdlBxpj/ON/vFpENhRFQUZGe6WD8jzupUzGQQe20D8TiwBjDV199xWOPPUZMTAwHDx6katWqtGvXzu7QlCrS8ipRBIhIKxFpLSKtgVLZpou1acv2czAmiTG9GuldTsXA/v376dWrF4MGDSI8PJxVq1ZRtWpVu8NSyiPkVaI4DkzKMn0iy7QBrnVXUHY7ciaJKb/to1fTylzXOMzucNRlio+Pp02bNjgcDt577z0eeughvL01+SvlqrwGLupamIEUJe8u2Ysx8MIN2oDtybZs2ULz5s0JCgpi5syZdOjQgfDwcLvDUsrjlKgH51yx7Wgs32yMYshVtXQgIg8VHR3N4MGDadGiBYsWWY/89OvXT5OEUvnk1kQhIj1FZLeI7BORZ/NYrq2IZNrdh5Qxhue+3Ur50n6M7KLjHHsah8PBjBkzaNCgAXPmzGHMmDF06dLF7rCU8niudOGRL85xKz4AugNRwFoR+c4YsyOH5SYAP7srFld9t/kYm6NimdCvGWVLl+ghNzxSv379mD9/Pp07d2bq1Kk0bqxVh0oVBFd6jxXnWNkvOqdriIgr9xO2A/YZYyKNMWnAF1ij5GX3CDAPOJXDvEKTluHgzZ9207ByEP3bVLczFHUJEhMTycjIAOD2229n1qxZLFu2TJOEUgXIlaqnD4Ergdud0/FYJYWLCQeOZJmOcn52gYiEA7cA0/LakIgME5F1IrIuOjrahV1funkbojh6LpknejTA20u7b/AECxcupHHjxnz44YcADBgwgMGDB2v3G0oVMFcSRXtjzEggBcAYcxbwc2G9nP5bTbbpd4FnjDGZeW3IGDPdGBNhjImoWLGiC7u+NCnpmUz8eTeta5TjukaVCnz7qmAdOXKEW2+9lZtuuomgoCDatGljd0hKFWuutFGkO9sRDFwYj8LhwnpRQNY6nGrAsWzLRABfOL8BVgB6i0iGMWa+C9svMJ/+eZCYxDSm3NFav40WcbNnz2bEiBE4HA7Gjx/P448/jp+fK99blFL55UqieA/4FqgkIq8B/YHnXVhvLVBPRGoDR4FBwB1ZFzDG1D7/XkRmAd8XdpI4m5jG27/soVO9CnSoo11KF1Xnu/2uVq0aXbp04f3336d27doXX1EpddkumiiMMZ+JyHqgG1Z10s3GmJ0urJchIg9j3c3kDXxsjNkuIiOc8/NslygsHy2PJDXDwXN9Gmlpogg6d+4co0ePJjAwkIkTJ9KlSxe95VWpQnbRRCEiNYAkYGHWz4wxhy+2rjFmEdkGOcotQRhjhlxsewXtZFwKn/xxgL4tq9KwcnBh717lwRjDnDlzGDVqFNHR0Tz++OMXShVKqcLlStXTD1jtEwIEALWB3UATN8ZVKN7+ZTepGQ5Gda9vdygqiwMHDjBs2DCWLFlC27Zt+fHHH2nVqpXdYSlVYrlS9dQs67Sz59jhbouokCSmZvD9luN0a1iJmqGBdoejskhPT2fLli188MEHDB8+XDvwU8pml/xktjFmg4i0dUcwhWnu+iiS0jJ5sMsVdoeigKVLl/LDDz8wadIk6tevz6FDhwgI0MGilCoKXGmjGJVl0gtoDbjnqbdCNHd9FA0rBxFRS+90stPJkyd54okn+Oyzz7jiiit47rnnCA0N1SShVBHiygN3QVle/lhtFjl1xeExNhw+y9ajsQxqq1112MXhcPDRRx/RsGFDvvrqK1544QW2bt1KaGio3aEppbLJs0ThfNCujDHmqUKKp1BM/z2SIH8f+rWpZncoJVZsbCzPP/88LVu2ZOrUqTRs2NDukJRSuci1RCEiPs6uNYrVsKdHziTx665TDGhbnaAA7SG2MCUkJDBp0iQyMzMJCQnhr7/+4tdff9UkoVQRl1eJYg1WktgkIt8BXwOJ52caY75xc2xu8emfB0l3OBh6VS27QylRFixYwCOPPMKRI0do2bIl1157LXXq1LE7LKWUC1xpoygPxGCNkX0DcKPzp8c5k5jGnDVH6N4ojGohpe0Op0Q4dOgQffv25eabb6ZcuXL88ccfXHttsR1uXaliKa8SRSXnHU/b+PuBu/Oy9wLrET754wAJqRmM6qEP2BUGYwz9+/dnx44dvPnmmzz22GP4+mp1n1KeJq9E4Q2UwbXuwos8YwwLNx/jqrqh2l2Hm61evZomTZoQFBTE9OnTKV++PDVr1rQ7LKVUPuWVKI4bY8YVWiRutunIOQ7GJDHiGn3Azl3OnDnD6NGjmT59Oi+++CJjx47VrjeUKgbyShTFqve1ueuj8PPxolezKnaHUuwYY5g9ezZPPPEEZ86c4YknnuCpp4rVHdVKlWh5JYpuhRaFm51LSuPbjUfp06wKZUtpHXlBGzNmDOPHj6dDhw4sXryYFi1a2B2SUqoA5ZoojDFnCjMQd5q58gDJ6ZkMv0ZvxywoKSkpJCQkUKFCBYYOHUrNmjUZNmwYXl6u3EinlPIkxf6/OiU9k1mrDnJdozBtxC4gixcvplmzZjzwwAMA1K9fnxEjRmiSUKqYKvb/2Qs3HyM+JYMhHWvZHYrHO3HiBHfccQc9evRARHj44YftDkkpVQguuZtxT2KM4b9/HqRWaGk6XqGdzV2O3377jVtuuYXk5GRefvllnnnmGe3hVakSoliXKNYfOsu2o3EM7lhLh9DMp/T0dACaN29O9+7d2bp1Ky+99JImCaVKkGKdKOZtOIq/jxcDIrQ78UsVHx/P448/TqdOncjMzCQ0NJSvv/6a+vX1qXalSppimyjiUtKZtyGKG1tUJdC/WNewFShjDN988w2NGjVi8uTJtGrVitTUVLvDUkrZqNgmim/WR5GW4eDuDtp1hKtOnz7NjTfeSL9+/ahQoQKrVq1i6tSplC6tHSgqVZIV20Qxf9Mx6oeVoUX1cnaH4jGCgoI4efIkkyZNYt26dXTo0MHukJRSRUCxTBSHYhLZdOQcfVuG2x1Kkbdy5Up69epFQkIC/v7+/PXXXzz++OP4+Gh1nVLKUiwTxY/bTgBwU4uqNkdSdMXExHD//ffTqVMnduzYQWRkJIA+NKeU+pdieVVYvieaOhUDqV5e69azM8Ywa9YsGjRowKxZs3jqqafYsWMHzZs3tzs0pVQRVezqFxJTM9hw+Cy3tdFbYnPz6aef0qBBA6ZNm0azZs3sDkcpVcQVuxLFr7tOkZLuoFezynaHUmQkJyfz0ksvERUVhYgwb948VqxYoUlCKeWSYpco5m88SliwP+1ra5cdAD///DNNmzZl3LhxLFiwAICQkBBti1BKuaxYXS12n4hn6a5TDIiojrdXye6y49ixYwwcOJCePXvi6+vLr7/+ysiRI+0OSynlgYpVopiz5jBeAvdcWcvuUGz36quvsmDBAsaNG8fmzZvp2rWr3SEppTxUsWrMXrnvNJ3rV6RikL/dodhi/fr1+Pr60rx5c1555RVGjRpF3bp17Q5LKeXh3FqiEJGeIrJbRPaJyLM5zL9TRLY4X6tEJN9jaJ5OSGXfqQTa1ip/eUF7oLi4OB599FHatWvHmDFjAAgNDdUkoZQqEG5LFCLiDXwA9AIaA7eLSONsix0ArjHGNAdeAabnd3+r9scAcGUJGnfCGMPXX39Nw4YNmTJlCg8++CCzZ8+2OyylVDHjzqqndsA+Y0wkgIh8AfQFdpxfwBizKsvyq4Fq+d3Zyr3RBPp50zy8bH434XE+//xz7rrrLlq1asWCBQto27at3SEppYohdyaKcOBIlukooH0ey98H/JjTDBEZBgwDqFGjxr/mG2P4ddcpujSohI93sWqf/5e0tDQiIyNp2LAh/fv3Jzk5mSFDhmjfTEopt3HnVTWn+1NNjguKdMVKFM/kNN8YM90YE2GMiahYseK/5u85mcDphDQ61i3e1U7Lly+nZcuW9OjRg5SUFPz9/bn//vs1SSil3MqdiSIKyNqPRjXgWPaFRKQ5MAPoa4yJyc+Oftt9CoBr6v87iRQHp0+fZujQoVxzzTUkJyczbdo0HYpUKVVo3PlVdC1QT0RqA0eBQcAdWRcQkRrAN8Ddxpg9+d3Rj9tO0LByENVCil8ngJGRkbRt25a4uDieffZZXnjhBR1ISClVqNyWKIwxGSLyMPAz4A18bIzZLiIjnPOnAS8CocCHIgKQYYyJuJT9nE5IZfORc/xft3oFewA2i4uLIzg4mNq1azN06FCGDBlC06ZN7Q5LKVUCubVy2xizCFiU7bNpWd7fD9x/OfvYcyIegIhaIZezmSIjKSmJV155henTp7N582aqVavGxIkT7Q5LKVWCeXwr6KaocwA0CAuyN5AC8MMPP/Dwww9z8OBBhg4dSqlSpewOSSmlPD9R/Lk/hnqVylAp2HMbdzMyMrj99tuZO3cujRo14vfff6dz5852h6WUUoCHdwqYkelg3cGzXFW3gt2h5Isx1t3CPj4+hIWF8frrr7Np0yZNEkqpIsWjE8XhM0kkp2fS1AOfxl67di3t27dnw4YNAEyZMoXRo0fj5+dnc2RKKfVPHp0oDsUkAVDDg8bGjo2N5eGHH6Z9+/ZERUURE5OvR0eUUqrQeHSi2HkiDoCGVTyjIft8B35Tp07l4YcfZteuXXTv3t3usJRSKk8e3Zi992QCYcH+BAf42h2KS3bu3El4eDgLFy4kIuKSHhdRSinbeHSJYt+pBOpWKmN3GLlKTU3l1VdfZeHChQCMHj2av/76S5OEUsqjeGyiyHQY9pyMp2HlYLtDydFvv/1GixYteOGFF1i6dCkAvr6+eHt72xyZUkpdGo9NFLtPxJOa4aBpeNFKFKdOnWLw4MFce+21pKen8+OPP/Luu+/aHZZSSuWbxyaKbUdjAWherZy9gWTzyy+/MGfOHJ577jm2bdtGz5497Q5JKaUui8c2Zu89FY+fjxc1i8CtsVu3bmX37t3079+fO++8k44dO1KnTh27w1JKqQLhwSWKOBpWDrJ1RLvExESefvppWrVqxdNPP016ejoioklCKVWseGSicDgMW6LO0czGJ7IXLlxI48aNeeuttxgyZAhr167F19czbtNVSqlL4ZFVT4fPJJGYlmlboti2bRs33XQTTZo0YcWKFVx99dW2xKGUUoXBI0sUu09aY1DUK8SuxTMyMli2bBkATZs25fvvv2fjxo2aJJRSxZ5HJoqtUbF4ewmNCqnrjvMPyXXr1o29e/cC0KdPH61qUkqVCB6ZKFbuO03tCoGU9nNvzdnZs2d58MEHufLKKzl9+jRff/01devWdes+lVKqqPHINoqE1AwC/dz7hHNqaiqtWrXiyJEjPPbYY4wdO5agIM/ofFAppQqSxyUKh7H6eHqoyxVu2f7Ro0cJDw/H39+fl19+mRYtWtCqVSu37EsppTyBx1U9pWVkAlCrQmCBbjclJYWxY8dSp04dFixYAMCQIUM0SSilSjyPK1GkZ1rDh15RseB6jV26dCkPPvgge/fu5fbbb6d9+/YFtm2llPJ0HleiSM90ABAW7F8g23vssce47rrrMMbwyy+/8Pnnn1O5cuUC2bZSShUHHpcoktOtqqdKQQH53obD4SAz09pOu3btePHFF9m6dauONqeUUjnwuESBVfOEn0/+Qt+8eTMdO3bkgw8+AOCOO+5g7NixBATkP/EopVRx5nGJIsNh8jWqXUJCAk888QRt2rQhMjJSq5eUUspFHtiY7aBy8KV9+1+yZAlDhw4lKiqKYcOGMX78eEJCQtwUoVJKFS8emSiqlL20ROHn50f58uX58ssv6dixo5siU0qp4snjEkWGw1C1XKk8l0lPT+fdd98lNjaWV199lc6dO7Nx40a8vDyupk0ppWznkVdOX2/Jdd6qVato06YNTz/9NDt37sThsG6n1SShlFL545FXzxqh/34q+8yZMwwbNoyrrrqKc+fOMX/+fObNm6cJQimlLpNHXkXLlfp3994xMTF8/vnnPPnkk+zYsYO+ffvaEJlSShU/HtdGARDob4W9e/duvvzyS1588UXq1avHoUOHCA0NtTk6pZQqXtxaohCRniKyW0T2icizOcwXEXnPOX+LiLR2ZbuBXpm8+OKLNG/enHfeeYcjR44AaJJQSik3cFuJQkS8gQ+A7kAUsFZEvjPG7MiyWC+gnvPVHpjq/JkrR2oSvbt04EDkfu68807efvttwsLC3HMQSiml3Fr11A7YZ4yJBBCRL4C+QNZE0Rf41BhjgNUiUk5Eqhhjjue20YzYk/hUqMOSJUvo1q2bG8NXSikF7k0U4cCRLNNR/Lu0kNMy4cA/EoWIDAOGOSdT9+7du+26664r2Gg9UwXgtN1BFBF6Lv6m5+Jvei7+1iC/K7ozUeT0sIPJxzIYY6YD0wFEZJ0xJuLyw/N8ei7+pufib3ou/qbn4m8isi6/67qzMTsKqJ5luhpwLB/LKKWUspE7E8VaoJ6I1BYRP2AQ8F22Zb4D7nHe/dQBiM2rfUIppVThc1vVkzEmQ0QeBn4GvIGPjTHbRWSEc/40YBHQG9gHJAFDXdj0dDeF7In0XPxNz8Xf9Fz8Tc/F3/J9LsS64UgppZTKmUd24aGUUqrwaKJQSimVpyKbKNzV/YcncuFc3Ok8B1tEZJWItLAjzsJwsXORZbm2IpIpIv0LM77C5Mq5EJEuIrJJRLaLyO+FHWNhceF/pKyILBSRzc5z4Up7qMcRkY9F5JSIbMtlfv6um8aYIvfCavzeD9QB/IDNQONsy/QGfsR6FqMD8Jfdcdt4LjoCIc73vUryuciy3K9YN0v0tztuG/8uymH1hFDDOV3J7rhtPBdjgAnO9xWBM4Cf3bG74Vx0BloD23KZn6/rZlEtUVzo/sMYkwac7/4jqwvdfxhjVgPlRKRKYQdaCC56Lowxq4wxZ52Tq7GeRymOXPm7AHgEmAecKszgCpkr5+IO4BtjzGEAY0xxPR+unAsDBImIAGWwEkVG4YbpfsaY5VjHlpt8XTeLaqLIrWuPS12mOLjU47wP6xtDcXTRcyEi4cAtwLRCjMsOrvxd1AdCRGSZiKwXkXsKLbrC5cq5mAI0wnqgdyvwf8YYR+GEV6Tk67pZVMejKLDuP4oBl49TRLpiJYqr3RqRfVw5F+8CzxhjMq0vj8WWK+fCB2gDdANKAX+KyGpjzB53B1fIXDkX1wObgGuBK4DFIrLCGBPn5tiKmnxdN4tqotDuP/7m0nGKSHNgBtDLGBNTSLEVNlfORQTwhTNJVAB6i0iGMWZ+oURYeFz9HzltjEkEEkVkOdACKG6JwpVzMRQYb6yK+n0icgBoCKwpnBCLjHxdN4tq1ZN2//G3i54LEakBfAPcXQy/LWZ10XNhjKltjKlljKkFzAUeKoZJAlz7H1kAdBIRHxEpjdV7885CjrMwuHIuDmOVrBCRMKyeVCMLNcqiIV/XzSJZojDu6/7D47h4Ll4EQoEPnd+kM0wx7DHTxXNRIrhyLowxO0XkJ2AL4ABmGGNyvG3Sk7n4d/EKMEtEtmJVvzxjjCl23Y+LyBygC1BBRKKAlwBfuLzrpnbhoZRSKk9FtepJKaVUEaGJQimlVJ40USillMqTJgqllFJ50kShlFIqT5ooVJHk7Pl1U5ZXrTyWTSiA/c0SkQPOfW0QkSvzsY0ZItLY+X5MtnmrLjdG53bOn5dtzt5Qy11k+ZYi0rsg9q1KLr09VhVJIpJgjClT0MvmsY1ZwPfGmLki0gOYaIxpfhnbu+yYLrZdEfkvsMcY81oeyw8BIowxDxd0LKrk0BKF8ggiUkZEljq/7W8VkX/1GisiVURkeZZv3J2cn/cQkT+d634tIhe7gC8H6jrXHeXc1jYRecz5WaCI/OAc22CbiAx0fr5MRCJEZDxQyhnHZ855Cc6fX2b9hu8syfQTEW8ReUtE1oo1TsBwF07Lnzg7dBORdmKNRbLR+bOB8ynlccBAZywDnbF/7NzPxpzOo1L/Ynf/6frSV04vIBOrE7dNwLdYvQgEO+dVwHqy9HyJOMH58wngOed7byDIuexyIND5+TPAiznsbxbOsSuA24C/sDrU2woEYnVNvR1oBfQD/pNl3bLOn8uwvr1fiCnLMudjvAX4r/O9H1ZPnqWAYcDzzs/9gXVA7RziTMhyfF8DPZ3TwYCP8/11wDzn+yHAlCzrvw7c5XxfDqvfp0C7f9/6KtqvItmFh1JAsjGm5fkJEfEFXheRzljdUYQDYcCJLOusBT52LjvfGLNJRK4BGgN/OLs38cP6Jp6Tt0TkeSAaqxfebsC3xupUDxH5BugE/ARMFJEJWNVVKy7huH4E3hMRf6AnsNwYk+ys7mouf4/IVxaoBxzItn4pEdkE1ALWA4uzLP9fEamH1Ruoby777wHcJCJPOqcDgBoUzz6gVAHRRKE8xZ1YI5O1Mcaki8hBrIvcBcaY5c5E0gf4n4i8BZwFFhtjbndhH08ZY+aenxCR63JayBizR0TaYPWZ84aI/GKMGefKQRhjUkRkGVa31wOBOed3BzxijPn5IptINsa0FJGywPfASOA9rL6MfjPG3OJs+F+Wy/oC9DPG7HYlXqVA2yiU5ygLnHImia5AzewLiEhN5zL/AWZiDQm5GrhKRM63OZQWkfou7nM5cLNznUCsaqMVIlIVSDLGzAYmOveTXbqzZJOTL7A6Y+uE1ZEdzp8Pnl9HROo795kjY0ws8CjwpHOdssBR5+whWRaNx6qCO+9n4BFxFq9EpFVu+1DqPE0UylN8BkSIyDqs0sWuHJbpAmwSkY1Y7QiTjTHRWBfOOSKyBStxNHRlh8aYDVhtF2uw2ixmGGM2As2ANc4qoOeAV3NYfTqw5Xxjdja/YI1tvMRYQ3eCNZbIDmCDiGwDPuIiJX5nLJuxutV+E6t08wdW+8V5vwGNzzdmY5U8fJ2xbXNOK5UnvT1WKaVUnrREoZRSKk+aKJRSSuVJE4VSSqk8aaJQSimVJ00USiml8qSJQimlVJ40USillMrT/wOYku4iSyQBvQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# calculate the auc score\n",
    "fpr,tpr,_ = roc_curve(targets,predictions)\n",
    "score = auc(fpr,tpr)\n",
    "plt.plot(fpr, tpr)\n",
    "plt.plot([0, 1], [0, 1], 'k--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title(f\"AUC {score}\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3> Recommend Movies </h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:59:46.116095Z",
     "iopub.status.busy": "2023-02-16T15:59:46.115423Z",
     "iopub.status.idle": "2023-02-16T15:59:46.127807Z",
     "shell.execute_reply": "2023-02-16T15:59:46.126898Z",
     "shell.execute_reply.started": "2023-02-16T15:59:46.116060Z"
    }
   },
   "outputs": [],
   "source": [
    "def recommend_movies(user_id,top_k=10):\n",
    "    # get all the movie_ids\n",
    "    unique_movies = ratings_df.movie_id.unique()\n",
    "    # get the movie_ids that are watched by the user\n",
    "    watched_movies_by_user = ratings_df[ratings_df['user_id'] == user_id]['movie_id'].values\n",
    "    # get the movie_ids that are not watched by the user\n",
    "    not_watched_movies = torch.tensor(list(set(unique_movies) - set(watched_movies_by_user)))\n",
    "    # create a tensor filled with user_id to fed the model\n",
    "    user = torch.full((1,not_watched_movies.shape[0]),user_id)\n",
    "    user = torch.squeeze(user)\n",
    "    \n",
    "    with torch.no_grad():\n",
    "        # make the prediction\n",
    "        user = user.to(device)\n",
    "        not_watched_movies = not_watched_movies.to(device)\n",
    "        movies = model(user,not_watched_movies)\n",
    "        \n",
    "    movies = torch.unsqueeze(movies,1)\n",
    "    movies = torch.cat([torch.unsqueeze(not_watched_movies,1),movies],dim=1)\n",
    "    # get the top k movies\n",
    "    movies = movies[movies[:, 1].sort(descending=True)[1]]\n",
    "    top_k_movies = movies[:top_k,:]\n",
    "    movie_ids = top_k_movies[:,0].to(int).tolist()\n",
    "    \n",
    "    # create a recommendation dataframe\n",
    "    recommended_movies = pd.DataFrame()\n",
    "    for idx,m_id in enumerate(movie_ids):\n",
    "        movie_row = ratings_df[ratings_df['movie_id']==m_id].iloc[0,:]\n",
    "        title = movie_row.title\n",
    "        genres = movie_row.genres\n",
    "\n",
    "        recommended_movies.at[idx,'movie_id'] = m_id\n",
    "        recommended_movies.at[idx,'score'] = top_k_movies[idx,1].item()\n",
    "        recommended_movies.at[idx,'title'] = title\n",
    "        recommended_movies.at[idx,'genres'] = genres\n",
    "        \n",
    "    return recommended_movies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-02-16T15:59:47.174824Z",
     "iopub.status.busy": "2023-02-16T15:59:47.174446Z",
     "iopub.status.idle": "2023-02-16T15:59:47.216476Z",
     "shell.execute_reply": "2023-02-16T15:59:47.215540Z",
     "shell.execute_reply.started": "2023-02-16T15:59:47.174793Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recommendations for user: 1154\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movie_id</th>\n",
       "      <th>score</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>828.0</td>\n",
       "      <td>0.898708</td>\n",
       "      <td>Talk of Angels (1998)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2850.0</td>\n",
       "      <td>0.890968</td>\n",
       "      <td>Ten Benny (1997)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2910.0</td>\n",
       "      <td>0.890200</td>\n",
       "      <td>End of the Affair, The (1955)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2309.0</td>\n",
       "      <td>0.882843</td>\n",
       "      <td>Apple, The (Sib) (1998)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2521.0</td>\n",
       "      <td>0.880965</td>\n",
       "      <td>Killing, The (1956)</td>\n",
       "      <td>Crime|Film-Noir</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movie_id     score                          title           genres\n",
       "0     828.0  0.898708          Talk of Angels (1998)            Drama\n",
       "1    2850.0  0.890968               Ten Benny (1997)            Drama\n",
       "2    2910.0  0.890200  End of the Affair, The (1955)            Drama\n",
       "3    2309.0  0.882843        Apple, The (Sib) (1998)            Drama\n",
       "4    2521.0  0.880965            Killing, The (1956)  Crime|Film-Noir"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_k_movies = 5\n",
    "user_id = random.randint(0,user_count+1)\n",
    "\n",
    "print(f\"Recommendations for user: {user_id}\")\n",
    "recommend_movies(user_id,top_k_movies)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.9.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}