index.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns \n",
    "import yaml\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.multiprocessing\n",
    "import torch.nn.functional as F \n",
    "import pytorch_lightning as pl\n",
    "\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "GLOBAL_SEED = 42  # number of life\n",
    "torch.manual_seed(GLOBAL_SEED)\n",
    "np.random.seed(GLOBAL_SEED)\n",
    "torch.multiprocessing.set_sharing_strategy('file_system')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('./dataset/config/age.yaml') as file:\n",
    "    age = yaml.load(file, Loader=yaml.FullLoader)\n",
    "\n",
    "with open('./dataset/config/occupation.yaml') as file:\n",
    "    occupation = yaml.load(file, Loader=yaml.FullLoader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "genre = ['Action', 'Adventure', 'Animation', 'Children\\'s', 'Comedy', 'Crime',\n",
    "\t    'Documentary', 'Drama', 'Fantasy', 'Film-Noir', 'Horror', 'Musical',\n",
    "\t    'Mystery', 'Romance', 'Sci-Fi', 'Thriller', 'War', 'Western']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "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>MovieID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3883.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1986.049446</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1146.778349</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>982.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2010.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2980.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3952.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           MovieID\n",
       "count  3883.000000\n",
       "mean   1986.049446\n",
       "std    1146.778349\n",
       "min       1.000000\n",
       "25%     982.500000\n",
       "50%    2010.000000\n",
       "75%    2980.500000\n",
       "max    3952.000000"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "movies = pd.read_csv('./dataset/movies.csv').drop(columns=['Unnamed: 0'])\n",
    "movies.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "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>UserID</th>\n",
       "      <th>Age</th>\n",
       "      <th>Occupation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>6040.000000</td>\n",
       "      <td>6040.000000</td>\n",
       "      <td>6040.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>3020.500000</td>\n",
       "      <td>30.639238</td>\n",
       "      <td>8.146854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1743.742145</td>\n",
       "      <td>12.895962</td>\n",
       "      <td>6.329511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1510.750000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3020.500000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4530.250000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>14.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6040.000000</td>\n",
       "      <td>56.000000</td>\n",
       "      <td>20.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            UserID          Age   Occupation\n",
       "count  6040.000000  6040.000000  6040.000000\n",
       "mean   3020.500000    30.639238     8.146854\n",
       "std    1743.742145    12.895962     6.329511\n",
       "min       1.000000     1.000000     0.000000\n",
       "25%    1510.750000    25.000000     3.000000\n",
       "50%    3020.500000    25.000000     7.000000\n",
       "75%    4530.250000    35.000000    14.000000\n",
       "max    6040.000000    56.000000    20.000000"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "users = pd.read_csv('./dataset/users.csv').drop(columns=['Unnamed: 0'])\n",
    "users.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "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>UserID</th>\n",
       "      <th>MovieID</th>\n",
       "      <th>Rating</th>\n",
       "      <th>Timestamp</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",
       "    </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",
       "    </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",
       "    </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",
       "    </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",
       "    </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",
       "    </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",
       "    </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",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             UserID       MovieID        Rating     Timestamp\n",
       "count  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\n",
       "std    1.728413e+03  1.096041e+03  1.117102e+00  1.215256e+07\n",
       "min    1.000000e+00  1.000000e+00  1.000000e+00  9.567039e+08\n",
       "25%    1.506000e+03  1.030000e+03  3.000000e+00  9.653026e+08\n",
       "50%    3.070000e+03  1.835000e+03  4.000000e+00  9.730180e+08\n",
       "75%    4.476000e+03  2.770000e+03  4.000000e+00  9.752209e+08\n",
       "max    6.040000e+03  3.952000e+03  5.000000e+00  1.046455e+09"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ratings = pd.read_csv('./dataset/ratings.csv').drop(columns=['Unnamed: 0'])\n",
    "ratings.describe()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 3883 movies\n",
      "There are 6040 users\n"
     ]
    }
   ],
   "source": [
    "print('There are {} movies'.format(len(movies.MovieID)))\n",
    "print('There are {} users'.format(len(users.UserID)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'No.USER (GENDER)')"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "gender_counts = users.Gender.value_counts()\n",
    "gender_dict = {\n",
    "    'female': gender_counts[0],\n",
    "    'male': gender_counts[1]\n",
    "}\n",
    "\n",
    "plt.figure(figsize = (10, 5))\n",
    "plt.bar(gender_dict.keys(), gender_dict.values(), color ='maroon', width = 0.4)\n",
    "plt.xlabel(\"GENDER\")\n",
    "plt.xticks(rotation=70)\n",
    "plt.ylabel(\"No.USER (GENDER)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='Gender'>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAggAAAGwCAYAAADMjZ3mAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAVWUlEQVR4nO3de5DVdf348deyyy4Iu+AlcNE1Bhy8pWPqSEhmKY1lof6j9tNBbTRrwMa8YBo1OGlIZk0zppZa2pQTWYPmCBqlYXnLEcFIEVPwHpqmsogssPv+/eGwub4g2f2y5+zS4zGzM5xzPmfPa1+ufJ6cc3a2ppRSAgDgPQZUewAAoO8RCABAIhAAgEQgAACJQAAAEoEAACQCAQBI6np6x46Ojnj55ZejsbExampqtuVMAEAvKaVEa2trjBo1KgYM2PLzBD0OhJdffjlaWlp6encAoIpeeOGF2H333bd4e48DobGxsfMBmpqaevppAIAKWr16dbS0tHSex7ekx4Gw6WWFpqYmgQAA/cwHvT3AmxQBgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAUlftAd6rlBLr1q2r9hgAUDWllGhra4uIiGHDhsWAAdX5t3yfCoR169bFZz/72WqPAQB9wq233ho77rhjVR7bSwwAQNKnnkF4rzUH/r8oA/rseADQK2o2rIuhS39T7TH6biCUAXURtQOrPQYAVFRp31DtESLCSwwAwGYIBAAgEQgAQCIQAIBEIAAAiUAAABKBAAAkAgEASAQCAJAIBAAgEQgAQCIQAIBEIAAAiUAAABKBAAAkAgEASAQCAJAIBAAgEQgAQCIQAIBEIAAAiUAAABKBAAAkAgEASAQCAJAIBAAgEQgAQCIQAIBEIAAAiUAAABKBAAAkAgEASAQCAJAIBAAgEQgAQCIQAIBEIAAAiUAAABKBAAAkAgEASAQCAJAIBAAgEQgAQCIQAIBEIAAAiUAAABKBAAAkAgEASAQCAJAIBAAgEQgAQCIQAIBEIAAAiUAAABKBAAAkAgEASAQCAJAIBAAgEQgAQCIQAIBEIAAAiUAAABKBAAAkAgEASAQCAJAIBAAgEQgAQCIQAIBEIAAASZ8KhFLKey9UbxAAqJYup8LqnQv7VCC0tbX950LHxuoNAgDV8p7zX5fzYoX1qUAAAPoGgQAAJAIBAEgEAgCQCAQAIBEIAEAiEACARCAAAIlAAAASgQAAJAIBAEgEAgCQCAQAIBEIAEAiEACARCAAAIlAAAASgQAAJAIBAEgEAgCQCAQAIBEIAEAiEACARCAAAIlAAAASgQAAJAIBAEgEAgCQCAQAIBEIAEAiEACARCAAAIlAAAASgQAAJAIBAEgEAgCQCAQAIBEIAEAiEACARCAAAIlAAAASgQAAJAIBAEgEAgCQCAQAIBEIAEAiEACARCAAAIlAAAASgQAAJAIBAEgEAgCQCAQAIBEIAEAiEACARCAAAIlAAAASgQAAJAIBAEgEAgCQCAQAIBEIAEAiEACARCAAAIlAAAASgQAAJAIBAEgEAgCQCAQAIBEIAEAiEACARCAAAIlAAACSuq09sK2tLdra2jovr169ulcGAgCqb6ufQbj88stj2LBhnR8tLS29ORcAUEVbHQgXX3xxvPXWW50fL7zwQm/OBQBU0Va/xNDQ0BANDQ29OQsA0Ed4kyIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBI+lQgNDQ0/OfCgLrqDQIA1fKe81+X82Klx6jaI29GTU3Ney9UbxAAqJYup8LqnQv7VCAAAH2DQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkNRVe4AtqenYGKXaQwBAhdV0tFd7hIjow4EwdMmvqj0CAPzP8hIDAJD0qWcQBg0aFHfeeWe1xwCAqimlRFtbW0REDBs2rGpz9KlAqKmpicGDB1d7DACoqh122KHaI3iJAQDIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBAIAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBI6np6x1JKRESsXr16mw0DAPSuTeftTefxLelxILS2tkZEREtLS08/BQBQJa2trTFs2LAt3l5TPightqCjoyNefvnlaGxsjJqamh4P+H6rV6+OlpaWeOGFF6KpqWmbfV4yu64Me64Me64Me66M3txzKSVaW1tj1KhRMWDAlt9p0ONnEAYMGBC77757T+/+gZqamnzzVYhdV4Y9V4Y9V4Y9V0Zv7fm/PXOwiTcpAgCJQAAAkj4XCA0NDTFz5sxoaGio9ijbPbuuDHuuDHuuDHuujL6w5x6/SREA2H71uWcQAIDqEwgAQCIQAIBEIAAASVUC4eqrr47Ro0fHoEGDYvz48fHwww//1+N/85vfxN577x2DBg2K/fffP+bPn1+hSfu/7uz6+uuvj8MPPzx23HHH2HHHHWPSpEkf+N+Gd3X3e3qTOXPmRE1NTRx//PG9O+B2ort7fvPNN2PatGnR3NwcDQ0NMW7cOH9/bIXu7vmHP/xh7LXXXjF48OBoaWmJc889N9atW1ehafunP//5zzF58uQYNWpU1NTUxG233faB91m4cGEcdNBB0dDQEHvuuWfcdNNNvTtkqbA5c+aU+vr68rOf/aw8/vjj5Utf+lIZPnx4eeWVVzZ7/P33319qa2vLFVdcUZ544onyzW9+swwcOLAsXbq0wpP3P93d9cknn1yuvvrqsnjx4rJs2bJy+umnl2HDhpUXX3yxwpP3L93d8yYrV64su+22Wzn88MPLcccdV5lh+7Hu7rmtra0ccsgh5Zhjjin33XdfWblyZVm4cGFZsmRJhSfvX7q755tvvrk0NDSUm2++uaxcubL8/ve/L83NzeXcc8+t8OT9y/z588uMGTPK3LlzS0SUW2+99b8ev2LFirLDDjuU8847rzzxxBPlqquuKrW1teWuu+7qtRkrHgiHHnpomTZtWufl9vb2MmrUqHL55Zdv9vgTTzyxfO5zn+ty3fjx48uXv/zlXp1ze9DdXb/fxo0bS2NjY/n5z3/eWyNuF3qy540bN5bDDjus3HDDDeW0004TCFuhu3u+9tpry5gxY8r69esrNeJ2obt7njZtWjnyyCO7XHfeeeeViRMn9uqc25OtCYQLL7yw7Lfffl2uO+mkk8rRRx/da3NV9CWG9evXx6JFi2LSpEmd1w0YMCAmTZoUDz744Gbv8+CDD3Y5PiLi6KOP3uLxvKsnu36/tWvXxoYNG2KnnXbqrTH7vZ7u+dvf/naMGDEizjjjjEqM2e/1ZM+33357TJgwIaZNmxYjR46Mj3zkIzFr1qxob2+v1Nj9Tk/2fNhhh8WiRYs6X4ZYsWJFzJ8/P4455piKzPy/ohrnwh7/sqaeeO2116K9vT1GjhzZ5fqRI0fGk08+udn7rFq1arPHr1q1qtfm3B70ZNfv9/Wvfz1GjRqVvin5j57s+b777ouf/vSnsWTJkgpMuH3oyZ5XrFgR99xzT5xyyikxf/78ePrpp2Pq1KmxYcOGmDlzZiXG7nd6sueTTz45Xnvttfj4xz8epZTYuHFjfOUrX4lvfOMblRj5f8aWzoWrV6+Od955JwYPHrzNH9NPMbBZs2fPjjlz5sStt94agwYNqvY4243W1taYMmVKXH/99bHLLrtUe5ztWkdHR4wYMSKuu+66OPjgg+Okk06KGTNmxI9//ONqj7ZdWbhwYcyaNSuuueaaePTRR2Pu3Lkxb968uPTSS6s9Gv9HFX0GYZdddona2tp45ZVXulz/yiuvxK677rrZ++y6667dOp539WTXm1x55ZUxe/bs+OMf/xgHHHBAb47Z73V3z88880w8++yzMXny5M7rOjo6IiKirq4uli9fHmPHju3dofuhnnw/Nzc3x8CBA6O2trbzun322SdWrVoV69evj/r6+l6duT/qyZ6/9a1vxZQpU+LMM8+MiIj9998/3n777TjrrLNixowZMWCAf4duC1s6FzY1NfXKswcRFX4Gob6+Pg4++OC4++67O6/r6OiIu+++OyZMmLDZ+0yYMKHL8RERf/jDH7Z4PO/qya4jIq644oq49NJL46677opDDjmkEqP2a93d89577x1Lly6NJUuWdH4ce+yx8alPfSqWLFkSLS0tlRy/3+jJ9/PEiRPj6aef7gywiIinnnoqmpubxcEW9GTPa9euTRGwKcqKX/WzzVTlXNhrb3/cgjlz5pSGhoZy0003lSeeeKKcddZZZfjw4WXVqlWllFKmTJlSLrroos7j77///lJXV1euvPLKsmzZsjJz5kw/5riVurvr2bNnl/r6+vLb3/62/POf/+z8aG1trdaX0C90d8/v56cYtk539/z888+XxsbGcvbZZ5fly5eXO+64o4wYMaJcdtll1foS+oXu7nnmzJmlsbGx/OpXvyorVqwoCxYsKGPHji0nnnhitb6EfqG1tbUsXry4LF68uERE+cEPflAWL15cnnvuuVJKKRdddFGZMmVK5/Gbfsxx+vTpZdmyZeXqq6/e/n7MsZRSrrrqqrLHHnuU+vr6cuihh5aHHnqo87YjjjiinHbaaV2Ov+WWW8q4ceNKfX192W+//cq8efMqPHH/1Z1df/jDHy4RkT5mzpxZ+cH7me5+T7+XQNh63d3zAw88UMaPH18aGhrKmDFjyne+852ycePGCk/d/3Rnzxs2bCiXXHJJGTt2bBk0aFBpaWkpU6dOLW+88UblB+9H/vSnP23279tNuz3ttNPKEUccke5z4IEHlvr6+jJmzJhy44039uqMft0zAJB49wgAkAgEACARCABAIhAAgEQgAACJQAAAEoEAACQCAQBIBALQI5/85Cfja1/7WrXHAHqJQIB+bNWqVXHOOefEnnvuGYMGDYqRI0fGxIkT49prr421a9dWezygH6vor3sGtp0VK1bExIkTY/jw4TFr1qzYf//9o6GhIZYuXRrXXXdd7LbbbnHsscdWe8wtam9vj5qaGr8OGPoo/2dCPzV16tSoq6uLRx55JE488cTYZ599YsyYMXHcccfFvHnzYvLkyRER8eabb8aZZ54ZH/rQh6KpqSmOPPLIeOyxxzo/zyWXXBIHHnhg/OIXv4jRo0fHsGHD4gtf+EK0trZ2HvP222/HqaeeGkOHDo3m5ub4/ve/n+Zpa2uLCy64IHbbbbcYMmRIjB8/PhYuXNh5+0033RTDhw+P22+/Pfbdd99oaGiI559/vvcWBPyfCAToh15//fVYsGBBTJs2LYYMGbLZY2pqaiIi4oQTTohXX3017rzzzli0aFEcdNBBcdRRR8W///3vzmOfeeaZuO222+KOO+6IO+64I+69996YPXt25+3Tp0+Pe++9N373u9/FggULYuHChfHoo492ebyzzz47HnzwwZgzZ0787W9/ixNOOCE+85nPxD/+8Y/OY9auXRvf/e5344YbbojHH388RowYsS3XAmxLvfq7IoFe8dBDD5WIKHPnzu1y/c4771yGDBlShgwZUi688MLyl7/8pTQ1NZV169Z1OW7s2LHlJz/5SSmllJkzZ5YddtihrF69uvP26dOnl/Hjx5dS3v299fX19eWWW27pvP31118vgwcPLuecc04ppZTnnnuu1NbWlpdeeqnL4xx11FHl4osvLqWUcuONN5aIKEuWLNk2SwB6lfcgwHbk4Ycfjo6OjjjllFOira0tHnvssVizZk3svPPOXY5755134plnnum8PHr06GhsbOy83NzcHK+++mpEvPvswvr162P8+PGdt++0006x1157dV5eunRptLe3x7hx47o8TltbW5fHrq+vjwMOOGDbfLFArxII0A/tueeeUVNTE8uXL+9y/ZgxYyIiYvDgwRERsWbNmmhubu7yXoBNhg8f3vnngQMHdrmtpqYmOjo6tnqeNWvWRG1tbSxatChqa2u73DZ06NDOPw8ePLjzpQ+gbxMI0A/tvPPO8elPfzp+9KMfxVe/+tUtvg/hoIMOilWrVkVdXV2MHj26R481duzYGDhwYPz1r3+NPfbYIyIi3njjjXjqqafiiCOOiIiIj370o9He3h6vvvpqHH744T16HKBv8SZF6Keuueaa2LhxYxxyyCHx61//OpYtWxbLly+PX/7yl/Hkk09GbW1tTJo0KSZMmBDHH398LFiwIJ599tl44IEHYsaMGfHII49s1eMMHTo0zjjjjJg+fXrcc8898fe//z1OP/30Lj+eOG7cuDjllFPi1FNPjblz58bKlSvj4YcfjssvvzzmzZvXWysAepFnEKCfGjt2bCxevDhmzZoVF198cbz44ovR0NAQ++67b1xwwQUxderUqKmpifnz58eMGTPii1/8YvzrX/+KXXfdNT7xiU/EyJEjt/qxvve978WaNWti8uTJ0djYGOeff3689dZbXY658cYb47LLLovzzz8/Xnrppdhll13iYx/7WHz+85/f1l86UAE1pZRS7SEAgL7FSwwAQCIQAIBEIAAAiUAAABKBAAAkAgEASAQCAJAIBAAgEQgAQCIQAIBEIAAAyf8HSDEyJ5EafxcAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "gender_numeric = {\n",
    "    'F': 0,\n",
    "    'M': 1\n",
    "}\n",
    "gender_box = users.Gender.map(gender_numeric)\n",
    "sns.boxplot(x=gender_box)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0\n",
       "1    1\n",
       "2    1\n",
       "3    1\n",
       "4    1\n",
       "Name: Gender, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def categorical_to_numeric_dict(col):\n",
    "    dictionary = {}\n",
    "    for value, key in enumerate(col.unique()):\n",
    "        dictionary[key] = value\n",
    "    return dictionary\n",
    "\n",
    "col = users.Gender\n",
    "col_dicionary = categorical_to_numeric_dict(col)\n",
    "sample = col.map(col_dicionary)\n",
    "sample.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'No.RATING')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rating_counts = ratings.Rating.value_counts()\n",
    "rating_dict = {\n",
    "    1: rating_counts[1],\n",
    "    2: rating_counts[2],\n",
    "    3: rating_counts[3],\n",
    "    4: rating_counts[4],\n",
    "    5: rating_counts[5]\n",
    "}\n",
    "\n",
    "plt.figure(figsize = (10, 5))\n",
    "plt.bar(rating_dict.keys(), rating_dict.values(), color ='maroon',\n",
    "        width = 0.4)\n",
    "plt.xlabel(\"SCORE\")\n",
    "plt.xticks(rotation=70)\n",
    "plt.ylabel(\"No.RATING\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'No.CLIENT')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def count_occupation(column, dictionary):\n",
    "    data_dict = {}\n",
    "    for row in column:\n",
    "        if dictionary[row] not in list(data_dict.keys()):\n",
    "            data_dict[dictionary[row]] = 0\n",
    "        else:\n",
    "            data_dict[dictionary[row]] += 1\n",
    "    return data_dict\n",
    "\n",
    "occupation_dict = count_occupation(users.Occupation, occupation)\n",
    "\n",
    "plt.figure(figsize = (10, 5))\n",
    "plt.bar(occupation_dict.keys(), occupation_dict.values(), color ='maroon',\n",
    "        width = 0.4)\n",
    "plt.xlabel(\"OCCUPATION\")\n",
    "plt.xticks(rotation=70)\n",
    "plt.ylabel(\"No.CLIENT\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Content-based "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "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>MovieID</th>\n",
       "      <th>Title</th>\n",
       "      <th>Genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>255</th>\n",
       "      <td>258</td>\n",
       "      <td>Kid in King Arthur's Court, A (1995)</td>\n",
       "      <td>Adventure|Children's|Comedy|Fantasy|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>642</th>\n",
       "      <td>648</td>\n",
       "      <td>Mission: Impossible (1996)</td>\n",
       "      <td>Action|Adventure|Mystery</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>662</th>\n",
       "      <td>668</td>\n",
       "      <td>Pather Panchali (1955)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1171</th>\n",
       "      <td>1188</td>\n",
       "      <td>Strictly Ballroom (1992)</td>\n",
       "      <td>Comedy|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1783</th>\n",
       "      <td>1852</td>\n",
       "      <td>Love Walked In (1998)</td>\n",
       "      <td>Drama|Thriller</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1968</th>\n",
       "      <td>2037</td>\n",
       "      <td>Candleshoe (1977)</td>\n",
       "      <td>Adventure|Children's|Comedy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>2070</td>\n",
       "      <td>Tender Mercies (1983)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2056</th>\n",
       "      <td>2125</td>\n",
       "      <td>Ever After: A Cinderella Story (1998)</td>\n",
       "      <td>Drama|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2491</th>\n",
       "      <td>2560</td>\n",
       "      <td>Ravenous (1999)</td>\n",
       "      <td>Drama|Horror</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3016</th>\n",
       "      <td>3085</td>\n",
       "      <td>Living Dead Girl, The (La Morte Vivante) (1982)</td>\n",
       "      <td>Horror</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3301</th>\n",
       "      <td>3370</td>\n",
       "      <td>Betrayed (1988)</td>\n",
       "      <td>Drama|Thriller</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3302</th>\n",
       "      <td>3371</td>\n",
       "      <td>Bound for Glory (1976)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      MovieID                                            Title  \\\n",
       "255       258             Kid in King Arthur's Court, A (1995)   \n",
       "642       648                       Mission: Impossible (1996)   \n",
       "662       668                           Pather Panchali (1955)   \n",
       "1171     1188                         Strictly Ballroom (1992)   \n",
       "1783     1852                            Love Walked In (1998)   \n",
       "1968     2037                                Candleshoe (1977)   \n",
       "2001     2070                            Tender Mercies (1983)   \n",
       "2056     2125            Ever After: A Cinderella Story (1998)   \n",
       "2491     2560                                  Ravenous (1999)   \n",
       "3016     3085  Living Dead Girl, The (La Morte Vivante) (1982)   \n",
       "3301     3370                                  Betrayed (1988)   \n",
       "3302     3371                           Bound for Glory (1976)   \n",
       "\n",
       "                                           Genres  \n",
       "255   Adventure|Children's|Comedy|Fantasy|Romance  \n",
       "642                      Action|Adventure|Mystery  \n",
       "662                                         Drama  \n",
       "1171                               Comedy|Romance  \n",
       "1783                               Drama|Thriller  \n",
       "1968                  Adventure|Children's|Comedy  \n",
       "2001                                        Drama  \n",
       "2056                                Drama|Romance  \n",
       "2491                                 Drama|Horror  \n",
       "3016                                       Horror  \n",
       "3301                               Drama|Thriller  \n",
       "3302                                        Drama  "
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class UserProfile():\n",
    "    def __init__(self, user_prefer_list:pd.DataFrame):\n",
    "        if type(user_prefer_list) != pd.DataFrame:\n",
    "            raise \"THERE IS A PROBLEM FROM THE SYSTEM\"\n",
    "        self.movies = pd.read_csv('./dataset/movies.csv').drop(columns=['Unnamed: 0'])\n",
    "        self.ratings = pd.read_csv('./dataset/ratings.csv').drop(columns=['Unnamed: 0'])\n",
    "        self.users = pd.read_csv('./dataset/users.csv').drop(columns=['Unnamed: 0'])\n",
    "        self.genre = ['Action', 'Adventure', 'Animation', 'Children\\'s', 'Comedy', 'Crime',\n",
    "\t                    'Documentary', 'Drama', 'Fantasy', 'Film-Noir', 'Horror', 'Musical',\n",
    "\t                    'Mystery', 'Romance', 'Sci-Fi', 'Thriller', 'War', 'Western']\n",
    "        \n",
    "        self.user_prefer_list = user_prefer_list\n",
    "        self.movie_one_hot = pd.read_csv('./dataset/movie_one_hot.csv').drop(columns=['Unnamed: 0'])\n",
    "\n",
    "    def analyzed_user(self, user_prefer_list):\n",
    "        # User prefer list consists of <movie_name, ratings> => <{movie_genres}, ratings> -> LinearRegression/RidgeRegression\n",
    "        movie_mask = self.movies.Title.isin(user_prefer_list.Titles.values)\n",
    "        user_prefer_movie_id = self.movies.loc[movie_mask]['MovieID']\n",
    "        genre_mask = self.movie_one_hot.MovieID.isin(user_prefer_movie_id)\n",
    "        movie_genres = self.movie_one_hot.loc[genre_mask]\n",
    "        user_profile = LinearRegression().fit(movie_genres, user_prefer_list.Ratings)\n",
    "        return user_profile\n",
    "    \n",
    "    def predict(self, user, vector):\n",
    "        output = user(vector)\n",
    "        output = max(min(output, 1), 5)\n",
    "        return output\n",
    "        \n",
    "    def get_user_taste(self, analyzed_user):\n",
    "        user_taste = {}\n",
    "        for genre, coef in zip(self.genre, analyzed_user.coef_):\n",
    "            user_taste[genre] = coef\n",
    "        return user_taste   \n",
    "\n",
    "    def get_prefer_genres(self, analysed_user, num_genre=5):\n",
    "        prefer_genres = []\n",
    "        user_taste = self.get_user_taste(analysed_user)\n",
    "        key_list = list(user_taste.keys())\n",
    "        val_list = list(user_taste.values())\n",
    "\n",
    "        if num_genre > len(self.genre):\n",
    "            return \"No.Genre is way too much!!!\"\n",
    "        else:\n",
    "            sorted_coef = sorted(val_list.copy())[::-1]\n",
    "            genres_coef = sorted_coef[:num_genre]\n",
    "            for genre_coef in genres_coef:\n",
    "                position = val_list.index(genre_coef)\n",
    "                prefer_genres.append(key_list[position])\n",
    "            return prefer_genres\n",
    "        \n",
    "    def recommend(self, n_films=20, threshhold=3):\n",
    "        \"\"\"\n",
    "            1. Chọn riêng rẽ (theo tổ hợp) các phim có thể loại dựa trên thông tin của người dùng đã được xử lý.\n",
    "            2. Với mỗi tổ hợp, chọn ra n phim để đề xuất cho người dùng. \n",
    "        \"\"\"\n",
    "        recommend_idx = []\n",
    "        if threshhold > 5:\n",
    "            raise \"INVALID THRESH-HOLD\"\n",
    "        user_prefer_genres = self.get_prefer_genres(self.analyzed_user(self.user_prefer_list))\n",
    "        chosen_films = self.movie_one_hot[user_prefer_genres]\n",
    "        for idx in range(len(chosen_films)):\n",
    "            film_info = chosen_films.iloc[idx, :]\n",
    "            if film_info.sum() < threshhold:\n",
    "                continue\n",
    "            else:\n",
    "                recommend_idx.append(idx + 1)\n",
    "        recommend_idx = np.random.choice(recommend_idx, n_films)\n",
    "        mask = self.movies.MovieID.isin(recommend_idx)\n",
    "        recommend_movies = self.movies.loc[mask]\n",
    "        return recommend_movies\n",
    "        \n",
    "movie_mask = np.random.randint(0, 3883, 10)\n",
    "prefer_movie_title = movies.loc[movie_mask].Title\n",
    "movie_ratings = np.random.randint(0, 5, 10)\n",
    "user_prefer_list = {\n",
    "    'Titles': prefer_movie_title.values, \n",
    "    'Ratings': movie_ratings\n",
    "}\n",
    "user_profile = pd.DataFrame(data=user_prefer_list, index=None)\n",
    "\n",
    "recommender_system = UserProfile(user_profile)\n",
    "recommender_system.recommend()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Matrix Factorization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "ratings[\"Rating\"] = ratings[\"Rating\"] - 3  # rating range (-2, 2)\n",
    "train_ratings, validation_ratings = train_test_split(\n",
    "    ratings, test_size=0.1, random_state=GLOBAL_SEED\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# map movie id and user id to indexes.\n",
    "movie_index_by_id = {id: idx for idx, id in enumerate(movies[\"MovieID\"])}\n",
    "user_index_by_id = {id: idx for idx, id in enumerate(users[\"UserID\"])}\n",
    "\n",
    "\n",
    "class MLDataset(Dataset):\n",
    "    def __init__(self, ratings: pd.DataFrame):\n",
    "        self.ratings = ratings\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.ratings)\n",
    "\n",
    "    def __getitem__(self, index):\n",
    "        user_id = self.ratings[\"UserID\"].iloc[index]\n",
    "        movie_id = self.ratings[\"MovieID\"].iloc[index]\n",
    "        rating = self.ratings[\"Rating\"].iloc[index]\n",
    "        user_index = user_index_by_id[user_id]\n",
    "        movie_index = movie_index_by_id[movie_id]\n",
    "        return user_index, movie_index, rating\n",
    "\n",
    "\n",
    "training_data = MLDataset(train_ratings)\n",
    "validation_data = MLDataset(validation_ratings)\n",
    "batch_size = 1024\n",
    "\n",
    "train_dataloader = DataLoader(\n",
    "                    training_data, \n",
    "                    batch_size=batch_size, \n",
    "                    shuffle=True, \n",
    "                    num_workers=10\n",
    "                    )\n",
    "\n",
    "validation_dataloader = DataLoader(\n",
    "                        validation_data, \n",
    "                        batch_size=batch_size, \n",
    "                        shuffle=False, \n",
    "                        num_workers=10\n",
    "                        )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "import jdc\n",
    "\n",
    "LR = 1\n",
    "WEIGHT_DECAY = 5e-5\n",
    "\n",
    "\n",
    "class MatrixFactorization(pl.LightningModule):\n",
    "    \"\"\"Pytorch lighting class for Matrix Factorization training.\n",
    "\n",
    "    Attributes:\n",
    "        n_users: number of users.\n",
    "        n_items: number of items.\n",
    "        n_factors: number of latent factors (or embedding size)\n",
    "    \"\"\"\n",
    "\n",
    "    def __init__(self, n_users: int, n_items: int, n_factors: int = 40):\n",
    "        super().__init__()\n",
    "        self.n_users = n_users\n",
    "        self.n_items = n_items\n",
    "        self.n_factors = n_factors\n",
    "        self.user_biases = nn.Embedding(n_users, 1)\n",
    "        self.item_biases = nn.Embedding(n_items, 1)\n",
    "        self.bias = nn.Parameter(data=torch.rand(1))\n",
    "        self.user_embeddings = nn.Embedding(n_users, n_factors)\n",
    "        self.item_embeddings = nn.Embedding(n_items, n_factors)\n",
    "\n",
    "    def forward(self, users, items):\n",
    "        \"\"\"\n",
    "        Forward pass through the model. For a single user and item, this\n",
    "        looks like:\n",
    "        bias + user_bias + item_bias + user_embeddings.dot(item_embeddings)\n",
    "\n",
    "        Arguments:\n",
    "            users: Array of user indices\n",
    "            items : Array of item indices\n",
    "        Returns:\n",
    "            preds: Predicted ratings.\n",
    "        \"\"\"\n",
    "        # select users and items from the batch\n",
    "        batch_user_embs = self.user_embeddings(users)\n",
    "        batch_item_embs = self.item_embeddings(items)\n",
    "\n",
    "        preds = torch.reshape(\n",
    "            torch.diag(\n",
    "                torch.matmul(batch_user_embs, torch.transpose(batch_item_embs, 0, 1))\n",
    "            ),\n",
    "            (-1, 1),\n",
    "        )\n",
    "        # add bias\n",
    "        preds += self.user_biases(users) + self.item_biases(items) + self.bias\n",
    "\n",
    "        return torch.clip(preds.squeeze(), min=-2, max=2)\n",
    "\n",
    "    def training_step(self, batch, batch_idx):\n",
    "        users, items, rating = batch\n",
    "        rating = rating.to(torch.float32)\n",
    "        output = self.forward(users, items)\n",
    "        loss = F.mse_loss(rating, output)\n",
    "        self.log(\"batch_loss\", loss)\n",
    "        return {\"loss\": loss}  # for computing avg_loss in training_epoch_end\n",
    "\n",
    "    def configure_optimizers(self):\n",
    "        optimizer = torch.optim.SGD(self.parameters(), lr=LR, weight_decay=WEIGHT_DECAY)\n",
    "        return optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: True (cuda), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "IPU available: False, using: 0 IPUs\n",
      "HPU available: False, using: 0 HPUs\n",
      "c:\\Users\\ABC\\anaconda3\\envs\\torch_gpu\\lib\\site-packages\\pytorch_lightning\\trainer\\configuration_validator.py:70: UserWarning: You passed in a `val_dataloader` but have no `validation_step`. Skipping val loop.\n",
      "  rank_zero_warn(\"You passed in a `val_dataloader` but have no `validation_step`. Skipping val loop.\")\n",
      "Missing logger folder: mf_tb_logs\\lr1_wd5e-05\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
      "\n",
      "  | Name            | Type      | Params\n",
      "----------------------------------------------\n",
      "0 | user_biases     | Embedding | 6.0 K \n",
      "1 | item_biases     | Embedding | 3.9 K \n",
      "2 | user_embeddings | Embedding | 241 K \n",
      "3 | item_embeddings | Embedding | 155 K \n",
      "----------------------------------------------\n",
      "406 K     Trainable params\n",
      "0         Non-trainable params\n",
      "406 K     Total params\n",
      "1.627     Total estimated model params size (MB)\n"
     ]
    }
   ],
   "source": [
    "# for tensorboard\n",
    "from pytorch_lightning.loggers import TensorBoardLogger\n",
    "logger = TensorBoardLogger(\"mf_tb_logs\", name=f\"lr{LR}_wd{WEIGHT_DECAY}\")\n",
    "AVAIL_GPUS = min(1, torch.cuda.device_count())\n",
    "\n",
    "n_users = len(user_index_by_id)\n",
    "n_movies = len(movie_index_by_id)\n",
    "n_factors = 40\n",
    "\n",
    "model = MatrixFactorization(n_users=n_users, n_items=n_movies, n_factors=n_factors)\n",
    "\n",
    "trainer = pl.Trainer(max_epochs=100, devices=1, accelerator=\"gpu\", logger=logger)\n",
    "trainer.fit(model, train_dataloader, validation_dataloader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def eval_model(model, train_dataloader):\n",
    "    loss = 0\n",
    "    for users, items, rating in train_dataloader:\n",
    "        pred = model(users, items)\n",
    "        loss += F.mse_loss(pred, rating)\n",
    "    RMSE = (loss / len(train_dataloader))\n",
    "    return RMSE\n",
    "    \n",
    "print(\"Train RMSE: {:.3f}\".format(eval_model(model, train_dataloader)))\n",
    "print(\"Validation RMSE: {:.3f}\".format(eval_model(model, validation_dataloader)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "microservice",
   "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": "3.10.10"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}