reco_system.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "gh9wkEgdoTT1",
    "outputId": "a2f7380a-b529-4d1d-d6e6-08f352f1421d"
   },
   "outputs": [],
   "source": [
    "# !pip install pytorch_lightning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "execution": {
     "iopub.execute_input": "2023-03-18T07:28:49.890282Z",
     "iopub.status.busy": "2023-03-18T07:28:49.889911Z",
     "iopub.status.idle": "2023-03-18T07:28:49.898562Z",
     "shell.execute_reply": "2023-03-18T07:28:49.897695Z",
     "shell.execute_reply.started": "2023-03-18T07:28:49.890249Z"
    },
    "id": "BCaqnkE9oH02"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from tqdm.notebook import tqdm\n",
    "import matplotlib.pyplot as plt\n",
    "import re\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "import pytorch_lightning as pl\n",
    "\n",
    "from datetime import datetime\n",
    "\n",
    "np.random.seed(42)\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "from tqdm.auto import tqdm\n",
    "\n",
    "from random import sample\n",
    "\n",
    "import torch\n",
    "from torch.utils.data import Dataset\n",
    "import torch.nn as nn\n",
    "import pytorch_lightning as pl\n",
    "from torch.utils.data import DataLoader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True\n"
     ]
    }
   ],
   "source": [
    "print(torch.cuda.is_available())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([0.], device='cuda:0')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.zeros(1).cuda()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "qDp_O2iJoH04"
   },
   "source": [
    "### Import Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "_kg_hide-input": false,
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:52:15.340423Z",
     "iopub.status.busy": "2023-03-18T06:52:15.339536Z",
     "iopub.status.idle": "2023-03-18T06:52:24.281710Z",
     "shell.execute_reply": "2023-03-18T06:52:24.280633Z",
     "shell.execute_reply.started": "2023-03-18T06:52:15.340381Z"
    },
    "id": "P5SFyGQdoH05",
    "outputId": "3d089c1a-d256-44a4-8733-ec00eacc92c2"
   },
   "outputs": [],
   "source": [
    "# # Download MovieLens data.\n",
    "# print(\"Downloading movielens data...\")\n",
    "# from urllib.request import urlretrieve\n",
    "# import zipfile\n",
    "\n",
    "# urlretrieve(\"http://files.grouplens.org/datasets/movielens/ml-25m.zip\", \"movielens.zip\")\n",
    "# zip_ref = zipfile.ZipFile('movielens.zip', \"r\")\n",
    "# zip_ref.extractall()\n",
    "# print(\"Done!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 314
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:52:37.212106Z",
     "iopub.status.busy": "2023-03-18T06:52:37.211668Z",
     "iopub.status.idle": "2023-03-18T06:53:03.681604Z",
     "shell.execute_reply": "2023-03-18T06:53:03.680413Z",
     "shell.execute_reply.started": "2023-03-18T06:52:37.212073Z"
    },
    "id": "0vyHGRDAoH06",
    "outputId": "3c9496e5-f8fd-48ec-cc97-4b0ae86cf9cd"
   },
   "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>Adventure</th>\n",
       "      <th>Animation</th>\n",
       "      <th>Children</th>\n",
       "      <th>Comedy</th>\n",
       "      <th>Fantasy</th>\n",
       "      <th>Romance</th>\n",
       "      <th>Drama</th>\n",
       "      <th>Action</th>\n",
       "      <th>...</th>\n",
       "      <th>Horror</th>\n",
       "      <th>Mystery</th>\n",
       "      <th>Sci-Fi</th>\n",
       "      <th>IMAX</th>\n",
       "      <th>Documentary</th>\n",
       "      <th>War</th>\n",
       "      <th>Musical</th>\n",
       "      <th>Western</th>\n",
       "      <th>Film-Noir</th>\n",
       "      <th>(no genres listed)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story (1995)</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Jumanji (1995)</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Grumpier Old Men (1995)</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Waiting to Exhale (1995)</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Father of the Bride Part II (1995)</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId                               title  Adventure  Animation  \\\n",
       "0        1                    Toy Story (1995)        1.0        1.0   \n",
       "1        2                      Jumanji (1995)        1.0        0.0   \n",
       "2        3             Grumpier Old Men (1995)        0.0        0.0   \n",
       "3        4            Waiting to Exhale (1995)        0.0        0.0   \n",
       "4        5  Father of the Bride Part II (1995)        0.0        0.0   \n",
       "\n",
       "   Children  Comedy  Fantasy  Romance  Drama  Action  ...  Horror  Mystery  \\\n",
       "0       1.0     1.0      1.0      0.0    0.0     0.0  ...     0.0      0.0   \n",
       "1       1.0     0.0      1.0      0.0    0.0     0.0  ...     0.0      0.0   \n",
       "2       0.0     1.0      0.0      1.0    0.0     0.0  ...     0.0      0.0   \n",
       "3       0.0     1.0      0.0      1.0    1.0     0.0  ...     0.0      0.0   \n",
       "4       0.0     1.0      0.0      0.0    0.0     0.0  ...     0.0      0.0   \n",
       "\n",
       "   Sci-Fi  IMAX  Documentary  War  Musical  Western  Film-Noir  \\\n",
       "0     0.0   0.0          0.0  0.0      0.0      0.0        0.0   \n",
       "1     0.0   0.0          0.0  0.0      0.0      0.0        0.0   \n",
       "2     0.0   0.0          0.0  0.0      0.0      0.0        0.0   \n",
       "3     0.0   0.0          0.0  0.0      0.0      0.0        0.0   \n",
       "4     0.0   0.0          0.0  0.0      0.0      0.0        0.0   \n",
       "\n",
       "   (no genres listed)  \n",
       "0                 0.0  \n",
       "1                 0.0  \n",
       "2                 0.0  \n",
       "3                 0.0  \n",
       "4                 0.0  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Import movies data\n",
    "movies_data = pd.read_csv(\"./ml-25m/movies.csv\")\n",
    "genres_split = movies_data.genres.apply(lambda x: x.split(sep='|')).apply(pd.value_counts, 1).fillna(0)\n",
    "movies_data = pd.concat([movies_data.iloc[:,:-1], genres_split], axis=1)\n",
    "movies_data.shape\n",
    "movies_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:03.684262Z",
     "iopub.status.busy": "2023-03-18T06:53:03.683873Z",
     "iopub.status.idle": "2023-03-18T06:53:11.229039Z",
     "shell.execute_reply": "2023-03-18T06:53:11.227707Z",
     "shell.execute_reply.started": "2023-03-18T06:53:03.684214Z"
    },
    "id": "enDM1SAkoH06",
    "outputId": "90cd21e3-bb3d-4855-b355-03876d7fd92b"
   },
   "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>0</th>\n",
       "      <td>1</td>\n",
       "      <td>296</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2006-05-17 15:34:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>306</td>\n",
       "      <td>3.5</td>\n",
       "      <td>2006-05-17 12:26:57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>307</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2006-05-17 12:27:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>665</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2006-05-17 15:13:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>899</td>\n",
       "      <td>3.5</td>\n",
       "      <td>2006-05-17 12:21:50</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   userId  movieId  rating           timestamp\n",
       "0       1      296     5.0 2006-05-17 15:34:04\n",
       "1       1      306     3.5 2006-05-17 12:26:57\n",
       "2       1      307     5.0 2006-05-17 12:27:08\n",
       "3       1      665     5.0 2006-05-17 15:13:40\n",
       "4       1      899     3.5 2006-05-17 12:21:50"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Import ratings data\n",
    "ratings_data = pd.read_csv(\"./ml-25m/ratings.csv\")\n",
    "ratings_data.timestamp = pd.to_datetime(ratings_data.timestamp, unit='s')\n",
    "ratings_data.shape\n",
    "ratings_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:11.231293Z",
     "iopub.status.busy": "2023-03-18T06:53:11.230861Z",
     "iopub.status.idle": "2023-03-18T06:53:11.931474Z",
     "shell.execute_reply": "2023-03-18T06:53:11.930338Z",
     "shell.execute_reply.started": "2023-03-18T06:53:11.231255Z"
    },
    "id": "ZPvBnDhuoH06",
    "outputId": "622fbeb9-02b8-4b87-b916-cb65c71bfb9e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 25000095 rows in the original data from 162541 users\n",
      "There are 2503205 rows in the sampled data from 16254 users\n"
     ]
    }
   ],
   "source": [
    "# User Sampling\n",
    "# Select 1/10 unique users from the original user pool due to memory limitations\n",
    "\n",
    "print(f'There are {len(ratings_data)} rows in the original data from {len(ratings_data.userId.unique())} users')\n",
    "\n",
    "rand_userIds = np.random.choice(ratings_data['userId'].unique(), size=int(len(ratings_data['userId'].unique())*0.1), replace=False)\n",
    "ratings_data = ratings_data.loc[ratings_data['userId'].isin(rand_userIds)]\n",
    "\n",
    "print(f'There are {len(ratings_data)} rows in the sampled data from {len(rand_userIds)} users')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ER1J58LQU5Gs",
    "outputId": "99f5f7f0-5ecf-4130-e38c-43f9a9187a34"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(31646, 22)\n"
     ]
    }
   ],
   "source": [
    "# Drop Movies without any user ratings\n",
    "movies_data = movies_data[movies_data.movieId.isin(ratings_data.movieId)].reset_index(drop=True)\n",
    "print(movies_data.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "m-6jt0wZoH07"
   },
   "source": [
    "### Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 299
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:11.934916Z",
     "iopub.status.busy": "2023-03-18T06:53:11.934610Z",
     "iopub.status.idle": "2023-03-18T06:53:12.306461Z",
     "shell.execute_reply": "2023-03-18T06:53:12.305366Z",
     "shell.execute_reply.started": "2023-03-18T06:53:11.934889Z"
    },
    "id": "2Yu9e12GoH07",
    "outputId": "8b261d1a-24be-48d1-ac4b-bdbef74233c0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 1077.0 movies with invalid year data\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,4))\n",
    "movies_data.iloc[:, 2:].sum().plot.barh()\n",
    "_=plt.title('Count of Records per Genre')\n",
    "\n",
    "print(f\"There are {movies_data['(no genres listed)'].sum()} movies with invalid year data\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "xhgZg0ud02GX"
   },
   "outputs": [],
   "source": [
    "# Drop ratings with movies without genre listed\n",
    "ratings_data = ratings_data[ratings_data.movieId.isin(movies_data[movies_data['(no genres listed)'] == 0].movieId)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:12.308276Z",
     "iopub.status.busy": "2023-03-18T06:53:12.307929Z",
     "iopub.status.idle": "2023-03-18T06:53:12.470126Z",
     "shell.execute_reply": "2023-03-18T06:53:12.469178Z",
     "shell.execute_reply.started": "2023-03-18T06:53:12.308248Z"
    },
    "id": "_8XmAKRyoH07",
    "outputId": "6d1944d8-b0cd-4fd8-b3c5-de732c362623"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 134 movies with invalid year data\n"
     ]
    }
   ],
   "source": [
    "# Extract year from movie title and add as new feature\n",
    "# Any invalid year will be labelled '9999'\n",
    "\n",
    "movies_data_year = movies_data.title.apply(lambda x: re.findall('\\(([0-9]*)\\)\\s*$', x)[0] if len(re.findall('\\(([0-9]*)\\)\\s*$', x)) > 0 else '9999')\n",
    "movies_data['year'] = movies_data_year.astype('int64')\n",
    "\n",
    "print(f'There are {(movies_data.year == 9999).sum()} movies with invalid year data')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "T55GhYBue-EF"
   },
   "outputs": [],
   "source": [
    "# Drop ratings with movies labelled with year 9999\n",
    "ratings_data = ratings_data[ratings_data.movieId.isin(movies_data[movies_data.year != 9999].movieId)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 281
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:12.504210Z",
     "iopub.status.busy": "2023-03-18T06:53:12.503635Z",
     "iopub.status.idle": "2023-03-18T06:53:12.947476Z",
     "shell.execute_reply": "2023-03-18T06:53:12.946368Z",
     "shell.execute_reply.started": "2023-03-18T06:53:12.504173Z"
    },
    "id": "urV6UgbLoH08",
    "outputId": "e858c757-ae2b-4e97-b06a-dfb6b7ddf2ab"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,4))\n",
    "plt.barh(movies_data.year.value_counts().index, movies_data.year.value_counts().values)\n",
    "plt.ylim([1850,2023])\n",
    "_=plt.title('Movie release year distribution')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:12.472086Z",
     "iopub.status.busy": "2023-03-18T06:53:12.471697Z",
     "iopub.status.idle": "2023-03-18T06:53:12.502022Z",
     "shell.execute_reply": "2023-03-18T06:53:12.500824Z",
     "shell.execute_reply.started": "2023-03-18T06:53:12.472051Z"
    },
    "id": "J4zaBL9roH08"
   },
   "outputs": [],
   "source": [
    "movies_data = movies_data[movies_data.year != 9999][movies_data['(no genres listed)'] == 0].reset_index(drop=True)\n",
    "movies_data.drop(columns={'(no genres listed)'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "T7CZrA9Q1soQ",
    "outputId": "eb839408-8292-4f89-dc3c-97fca9e7c14a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 30516 movies in the ratings dataset\n",
      "There are 30516 movies in the movies dataset\n"
     ]
    }
   ],
   "source": [
    "print(f'There are {len(ratings_data.movieId.unique())} movies in the ratings dataset')\n",
    "print(f'There are {len(movies_data)} movies in the movies dataset')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "i8kxiiZn2XQa"
   },
   "source": [
    "### Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 423
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:12.949686Z",
     "iopub.status.busy": "2023-03-18T06:53:12.948744Z",
     "iopub.status.idle": "2023-03-18T06:53:13.374295Z",
     "shell.execute_reply": "2023-03-18T06:53:13.373234Z",
     "shell.execute_reply.started": "2023-03-18T06:53:12.949648Z"
    },
    "id": "-bI9N4gGoH08",
    "outputId": "44c19277-a061-40f5-a577-1b117002d731"
   },
   "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>mean_rating</th>\n",
       "      <th>title</th>\n",
       "      <th>count_rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>3.906575</td>\n",
       "      <td>Toy Story (1995)</td>\n",
       "      <td>5673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>3.251423</td>\n",
       "      <td>Jumanji (1995)</td>\n",
       "      <td>2460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>3.136477</td>\n",
       "      <td>Grumpier Old Men (1995)</td>\n",
       "      <td>1198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>2.957721</td>\n",
       "      <td>Waiting to Exhale (1995)</td>\n",
       "      <td>272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>3.014856</td>\n",
       "      <td>Father of the Bride Part II (1995)</td>\n",
       "      <td>1178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30511</th>\n",
       "      <td>208765</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>Picture a Perfect Christmas (2019)</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30512</th>\n",
       "      <td>208800</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>Lady and the Tramp (2019)</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30513</th>\n",
       "      <td>208843</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>An Acceptable Loss (2019)</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30514</th>\n",
       "      <td>209085</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>The Mistletoe Secret (2019)</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30515</th>\n",
       "      <td>209155</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>Santosh Subramaniam (2008)</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>30516 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId  mean_rating                               title  count_rating\n",
       "0            1     3.906575                    Toy Story (1995)          5673\n",
       "1            2     3.251423                      Jumanji (1995)          2460\n",
       "2            3     3.136477             Grumpier Old Men (1995)          1198\n",
       "3            4     2.957721            Waiting to Exhale (1995)           272\n",
       "4            5     3.014856  Father of the Bride Part II (1995)          1178\n",
       "...        ...          ...                                 ...           ...\n",
       "30511   208765     5.000000  Picture a Perfect Christmas (2019)             1\n",
       "30512   208800     3.500000           Lady and the Tramp (2019)             1\n",
       "30513   208843     2.500000           An Acceptable Loss (2019)             1\n",
       "30514   209085     5.000000         The Mistletoe Secret (2019)             1\n",
       "30515   209155     5.000000          Santosh Subramaniam (2008)             1\n",
       "\n",
       "[30516 rows x 4 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_ratings = ratings_data.groupby(['movieId'])['rating'].mean().reset_index().merge(movies_data.iloc[:,:2], on='movieId')\n",
    "mean_ratings.columns = ['movieId', 'mean_rating', 'title']\n",
    "\n",
    "count_ratings = ratings_data.movieId.value_counts().reset_index()\n",
    "count_ratings.columns = ['movieId', 'count_rating']\n",
    "\n",
    "ratings_aggr = mean_ratings.merge(count_ratings, on='movieId')\n",
    "ratings_aggr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 423
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:13.377029Z",
     "iopub.status.busy": "2023-03-18T06:53:13.375866Z",
     "iopub.status.idle": "2023-03-18T06:53:13.397343Z",
     "shell.execute_reply": "2023-03-18T06:53:13.396379Z",
     "shell.execute_reply.started": "2023-03-18T06:53:13.376983Z"
    },
    "id": "PUUlSgpEoH08",
    "outputId": "f530a773-0ba9-4023-c93a-70c641611506"
   },
   "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>mean_rating</th>\n",
       "      <th>title</th>\n",
       "      <th>count_rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>22817</th>\n",
       "      <td>147328</td>\n",
       "      <td>4.571429</td>\n",
       "      <td>The Adventures of Sherlock Holmes and Dr. Wats...</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26031</th>\n",
       "      <td>171011</td>\n",
       "      <td>4.556122</td>\n",
       "      <td>Planet Earth II (2016)</td>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24376</th>\n",
       "      <td>159817</td>\n",
       "      <td>4.533133</td>\n",
       "      <td>Planet Earth (2006)</td>\n",
       "      <td>166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9434</th>\n",
       "      <td>32853</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>Sorrow and the Pity, The (Le chagrin et la pit...</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17492</th>\n",
       "      <td>107408</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>Only Old Men Are Going to Battle (V boy idut o...</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6213</th>\n",
       "      <td>6483</td>\n",
       "      <td>1.079545</td>\n",
       "      <td>From Justin to Kelly (2003)</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5506</th>\n",
       "      <td>5738</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Faces of Death 5 (1996)</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4577</th>\n",
       "      <td>4775</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Glitter (2001)</td>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1714</th>\n",
       "      <td>1826</td>\n",
       "      <td>0.978723</td>\n",
       "      <td>Barney's Great Adventure (1998)</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7904</th>\n",
       "      <td>8859</td>\n",
       "      <td>0.812500</td>\n",
       "      <td>SuperBabies: Baby Geniuses 2 (2004)</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10017 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId  mean_rating  \\\n",
       "22817   147328     4.571429   \n",
       "26031   171011     4.556122   \n",
       "24376   159817     4.533133   \n",
       "9434     32853     4.500000   \n",
       "17492   107408     4.500000   \n",
       "...        ...          ...   \n",
       "6213      6483     1.079545   \n",
       "5506      5738     1.000000   \n",
       "4577      4775     1.000000   \n",
       "1714      1826     0.978723   \n",
       "7904      8859     0.812500   \n",
       "\n",
       "                                                   title  count_rating  \n",
       "22817  The Adventures of Sherlock Holmes and Dr. Wats...            14  \n",
       "26031                             Planet Earth II (2016)            98  \n",
       "24376                                Planet Earth (2006)           166  \n",
       "9434   Sorrow and the Pity, The (Le chagrin et la pit...            13  \n",
       "17492  Only Old Men Are Going to Battle (V boy idut o...            11  \n",
       "...                                                  ...           ...  \n",
       "6213                         From Justin to Kelly (2003)            44  \n",
       "5506                             Faces of Death 5 (1996)            15  \n",
       "4577                                      Glitter (2001)            63  \n",
       "1714                     Barney's Great Adventure (1998)            47  \n",
       "7904                 SuperBabies: Baby Geniuses 2 (2004)            24  \n",
       "\n",
       "[10017 rows x 4 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ratings_aggr[ratings_aggr.count_rating > 10].sort_values('mean_rating', ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gc6wNXBBoH09"
   },
   "source": [
    "### Data Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7vMZhM_W2pSE"
   },
   "source": [
    "<img src='https://www.statology.org/wp-content/uploads/2020/11/loocv4-768x834.png' width='200' height='250'>\n",
    "\n",
    "**Leave-One-Out Methodology:** For each user, the model will be tested on the most recently rated movie, and trained on all of the movies rated prior to the most recent. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:13.403004Z",
     "iopub.status.busy": "2023-03-18T06:53:13.402735Z",
     "iopub.status.idle": "2023-03-18T06:53:14.810680Z",
     "shell.execute_reply": "2023-03-18T06:53:14.809452Z",
     "shell.execute_reply.started": "2023-03-18T06:53:13.402979Z"
    },
    "id": "kqjNfV6hoH09",
    "outputId": "6c15eade-3070-4ba3-9c68-b9d9eccd7afe"
   },
   "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",
       "      <th>recency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2450</th>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2010-02-03 19:06:10</td>\n",
       "      <td>261.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2451</th>\n",
       "      <td>13</td>\n",
       "      <td>19</td>\n",
       "      <td>3.5</td>\n",
       "      <td>2009-03-26 21:39:22</td>\n",
       "      <td>310.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2452</th>\n",
       "      <td>13</td>\n",
       "      <td>32</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2009-03-25 08:34:05</td>\n",
       "      <td>399.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2453</th>\n",
       "      <td>13</td>\n",
       "      <td>34</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2010-02-03 19:07:23</td>\n",
       "      <td>253.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2454</th>\n",
       "      <td>13</td>\n",
       "      <td>47</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2010-02-03 19:06:46</td>\n",
       "      <td>258.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      userId  movieId  rating           timestamp  recency\n",
       "2450      13        1     4.0 2010-02-03 19:06:10    261.0\n",
       "2451      13       19     3.5 2009-03-26 21:39:22    310.0\n",
       "2452      13       32     4.0 2009-03-25 08:34:05    399.0\n",
       "2453      13       34     4.0 2010-02-03 19:07:23    253.0\n",
       "2454      13       47     4.0 2010-02-03 19:06:46    258.0"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calcuate the recency ranking of each rating data\n",
    "ratings_data['recency'] = ratings_data.groupby(['userId'])['timestamp'].rank(method='first', ascending=False)\n",
    "ratings_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:14.935289Z",
     "iopub.status.busy": "2023-03-18T06:53:14.934844Z",
     "iopub.status.idle": "2023-03-18T06:53:15.317610Z",
     "shell.execute_reply": "2023-03-18T06:53:15.315806Z",
     "shell.execute_reply.started": "2023-03-18T06:53:14.935252Z"
    },
    "id": "LEE3Nus-oH09"
   },
   "outputs": [],
   "source": [
    "# Train-test Split using leave-one-out methodology\n",
    "ratings_data_test = ratings_data[ratings_data.recency == 1][['userId', 'movieId', 'rating', 'recency']]\n",
    "ratings_data_train = ratings_data[ratings_data.recency != 1][['userId', 'movieId', 'rating']]\n",
    "\n",
    "# Merge movies data to add content-based features to the input vectors\n",
    "# ratings_data_train = ratings_data_train.merge(movies_data.loc[:, movies_data.columns != 'title'], on='movieId')\n",
    "# ratings_data_test = ratings_data_test.merge(movies_data.loc[:, movies_data.columns != 'title'], on='movieId')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:15.320954Z",
     "iopub.status.busy": "2023-03-18T06:53:15.319938Z",
     "iopub.status.idle": "2023-03-18T06:53:15.347237Z",
     "shell.execute_reply": "2023-03-18T06:53:15.345160Z",
     "shell.execute_reply.started": "2023-03-18T06:53:15.320882Z"
    },
    "id": "GlaVgwosoH09",
    "outputId": "91e0a0a7-f525-4f5a-9409-81898c753d1e"
   },
   "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>recency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2855</th>\n",
       "      <td>13</td>\n",
       "      <td>87306</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3785</th>\n",
       "      <td>21</td>\n",
       "      <td>51255</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4335</th>\n",
       "      <td>25</td>\n",
       "      <td>509</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6376</th>\n",
       "      <td>45</td>\n",
       "      <td>861</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11803</th>\n",
       "      <td>85</td>\n",
       "      <td>263</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       userId  movieId  rating  recency\n",
       "2855       13    87306     3.5      1.0\n",
       "3785       21    51255     3.5      1.0\n",
       "4335       25      509     5.0      1.0\n",
       "6376       45      861     4.0      1.0\n",
       "11803      85      263     4.0      1.0"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ratings_data_test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZaDU3rpH5Fa_"
   },
   "source": [
    "Convert **Explicit Feedback** ratings data to **Implicit Feedback** by converting the *rating* metric (0 to 5) to *interacted* metric (0 or 1). We will then create a model that predicts the probability that a given user will interact with a movie/item (which can represent clicking on a thumbnail, searching for a movie, watching the movie, etc.) instead of predicting what rating a user will give a movie."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:15.354962Z",
     "iopub.status.busy": "2023-03-18T06:53:15.354108Z",
     "iopub.status.idle": "2023-03-18T06:53:15.530571Z",
     "shell.execute_reply": "2023-03-18T06:53:15.529351Z",
     "shell.execute_reply.started": "2023-03-18T06:53:15.354910Z"
    },
    "id": "CMtjuPhGoH0-"
   },
   "outputs": [],
   "source": [
    "# Move target column to last column index\n",
    "# Convert explicit feedback to implicit feedback\n",
    "\n",
    "col = ratings_data_train.pop('rating')\n",
    "ratings_data_train = pd.concat([ratings_data_train, col], axis=1)\n",
    "\n",
    "col = ratings_data_test.pop('rating')\n",
    "ratings_data_test = pd.concat([ratings_data_test, col], axis=1)\n",
    "\n",
    "ratings_data_train.rename(columns={'rating':'interacted'})\n",
    "ratings_data_test.rename(columns={'rating':'interacted'})\n",
    "\n",
    "ratings_data_train.loc[:, 'interacted'] = 1\n",
    "ratings_data_test.loc[:, 'interacted'] = 1\n",
    "\n",
    "ratings_data_test.drop(columns={'recency'}, inplace=True)\n",
    "ratings_data_test.drop(columns={'rating'}, inplace=True)\n",
    "ratings_data_train.drop(columns={'rating'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 242
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:15.536514Z",
     "iopub.status.busy": "2023-03-18T06:53:15.535771Z",
     "iopub.status.idle": "2023-03-18T06:53:15.554827Z",
     "shell.execute_reply": "2023-03-18T06:53:15.553650Z",
     "shell.execute_reply.started": "2023-03-18T06:53:15.536473Z"
    },
    "id": "dN5VL9K1oH0-",
    "outputId": "8438cfb2-fca9-4c47-da85-d6de8b285a68"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of training data: (2483622, 3) \n",
      "\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>userId</th>\n",
       "      <th>movieId</th>\n",
       "      <th>interacted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2450</th>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2451</th>\n",
       "      <td>13</td>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2452</th>\n",
       "      <td>13</td>\n",
       "      <td>32</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2453</th>\n",
       "      <td>13</td>\n",
       "      <td>34</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2454</th>\n",
       "      <td>13</td>\n",
       "      <td>47</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      userId  movieId  interacted\n",
       "2450      13        1           1\n",
       "2451      13       19           1\n",
       "2452      13       32           1\n",
       "2453      13       34           1\n",
       "2454      13       47           1"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Final Training Data\n",
    "print(f'Shape of training data: {ratings_data_train.shape} \\n')\n",
    "ratings_data_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pY935L5w7Rmg"
   },
   "source": [
    "### Pre-Model Transformations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PfJPrKoB7lpr"
   },
   "source": [
    "**NCF General Framework** [2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NaBbon9KoH0-"
   },
   "source": [
    "<img src=\"https://i.ibb.co/m869ZFR/Recommender-System-Architecture.png\"  width=\"60%\" height=\"30%\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 423
    },
    "id": "0KfLV-FcRNcj",
    "outputId": "b6ca01eb-9ade-4e22-e3c4-1e27b4905b19"
   },
   "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>Adventure</th>\n",
       "      <th>Animation</th>\n",
       "      <th>Children</th>\n",
       "      <th>Comedy</th>\n",
       "      <th>Fantasy</th>\n",
       "      <th>Romance</th>\n",
       "      <th>Drama</th>\n",
       "      <th>Action</th>\n",
       "      <th>Crime</th>\n",
       "      <th>Thriller</th>\n",
       "      <th>Horror</th>\n",
       "      <th>Mystery</th>\n",
       "      <th>Sci-Fi</th>\n",
       "      <th>IMAX</th>\n",
       "      <th>Documentary</th>\n",
       "      <th>War</th>\n",
       "      <th>Musical</th>\n",
       "      <th>Western</th>\n",
       "      <th>Film-Noir</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1406</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1407</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1408</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1409</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1410</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1411 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Adventure  Animation  Children  Comedy  Fantasy  Romance  Drama  Action  \\\n",
       "0           1.0        1.0       1.0     1.0      1.0      0.0    0.0     0.0   \n",
       "1           1.0        0.0       1.0     0.0      1.0      0.0    0.0     0.0   \n",
       "2           0.0        0.0       0.0     1.0      0.0      1.0    0.0     0.0   \n",
       "3           0.0        0.0       0.0     1.0      0.0      1.0    1.0     0.0   \n",
       "4           0.0        0.0       0.0     1.0      0.0      0.0    0.0     0.0   \n",
       "...         ...        ...       ...     ...      ...      ...    ...     ...   \n",
       "1406        0.0        0.0       0.0     0.0      1.0      1.0    0.0     0.0   \n",
       "1407        0.0        0.0       0.0     0.0      0.0      0.0    0.0     1.0   \n",
       "1408        1.0        0.0       1.0     1.0      0.0      0.0    0.0     0.0   \n",
       "1409        1.0        0.0       0.0     0.0      0.0      0.0    1.0     0.0   \n",
       "1410        0.0        1.0       1.0     1.0      1.0      0.0    0.0     0.0   \n",
       "\n",
       "      Crime  Thriller  Horror  Mystery  Sci-Fi  IMAX  Documentary  War  \\\n",
       "0       0.0       0.0     0.0      0.0     0.0   0.0          0.0  0.0   \n",
       "1       0.0       0.0     0.0      0.0     0.0   0.0          0.0  0.0   \n",
       "2       0.0       0.0     0.0      0.0     0.0   0.0          0.0  0.0   \n",
       "3       0.0       0.0     0.0      0.0     0.0   0.0          0.0  0.0   \n",
       "4       0.0       0.0     0.0      0.0     0.0   0.0          0.0  0.0   \n",
       "...     ...       ...     ...      ...     ...   ...          ...  ...   \n",
       "1406    0.0       0.0     1.0      0.0     1.0   0.0          0.0  0.0   \n",
       "1407    0.0       0.0     1.0      0.0     0.0   0.0          0.0  0.0   \n",
       "1408    0.0       0.0     1.0      0.0     0.0   0.0          0.0  0.0   \n",
       "1409    0.0       1.0     0.0      1.0     1.0   0.0          0.0  0.0   \n",
       "1410    0.0       0.0     1.0      0.0     0.0   0.0          0.0  0.0   \n",
       "\n",
       "      Musical  Western  Film-Noir  \n",
       "0         0.0      0.0        0.0  \n",
       "1         0.0      0.0        0.0  \n",
       "2         0.0      0.0        0.0  \n",
       "3         0.0      0.0        0.0  \n",
       "4         0.0      0.0        0.0  \n",
       "...       ...      ...        ...  \n",
       "1406      0.0      0.0        0.0  \n",
       "1407      0.0      1.0        0.0  \n",
       "1408      0.0      0.0        0.0  \n",
       "1409      0.0      0.0        0.0  \n",
       "1410      0.0      0.0        0.0  \n",
       "\n",
       "[1411 rows x 19 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a lookup table where each column represent a movie genre and the indices represent corresponding feature encoding for each genre combination\n",
    "movie_features_lookup = movies_data.iloc[:, 2:-1].drop_duplicates().reset_index(drop=True)\n",
    "movie_features_lookup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "id": "cRb6O6nZKug6"
   },
   "outputs": [],
   "source": [
    "# Convert to np array for faster execution\n",
    "movie_features_lookup = movie_features_lookup.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "id": "DZ6CWdRvS5z8"
   },
   "outputs": [],
   "source": [
    "def encode_movie_features(movie_feature, movie_features_lookup = movie_features_lookup):\n",
    "    return np.where(np.all(movie_features_lookup == movie_feature, axis=1))[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "s-aXJtb4pBet"
   },
   "outputs": [],
   "source": [
    "# All unique movies in the dataset\n",
    "all_movieIds = movies_data['movieId'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "cq5vhS559foy"
   },
   "source": [
    "**Generate Negative Samples**: All of our ratings data represent positive samples (i.e., movies the user has interacted with), we need to generate negative samples for the model as well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:15.559653Z",
     "iopub.status.busy": "2023-03-18T06:53:15.559268Z",
     "iopub.status.idle": "2023-03-18T06:53:17.636526Z",
     "shell.execute_reply": "2023-03-18T06:53:17.635484Z",
     "shell.execute_reply.started": "2023-03-18T06:53:15.559617Z"
    },
    "id": "OsOm9TZZoH0-",
    "outputId": "53820da5-f010-40a6-dd90-6817d5ed2904"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "userId\n",
       "13        [1, 19, 32, 34, 47, 60, 70, 104, 110, 112, 141...\n",
       "21        [260, 293, 318, 527, 541, 589, 741, 750, 778, ...\n",
       "25        [16, 17, 19, 21, 25, 36, 39, 47, 50, 52, 110, ...\n",
       "45        [82, 83, 125, 140, 141, 150, 296, 380, 562, 59...\n",
       "85        [7, 11, 17, 19, 21, 25, 34, 39, 74, 105, 141, ...\n",
       "                                ...                        \n",
       "162469    [110, 260, 318, 356, 527, 589, 1036, 1196, 119...\n",
       "162511    [1088, 1197, 1680, 2145, 2161, 2193, 2248, 310...\n",
       "162523    [252, 296, 370, 466, 543, 832, 1028, 1080, 137...\n",
       "162532    [260, 296, 541, 1021, 1036, 1136, 1196, 1197, ...\n",
       "162536    [50, 296, 318, 1089, 1208, 1222, 2959, 3535, 3...\n",
       "Name: movieId, Length: 16254, dtype: object"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Series containing all the movies interacted by each user.\n",
    "user_interactions = ratings_data_train.groupby('userId')['movieId'].apply(list)\n",
    "user_interactions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 349,
     "referenced_widgets": [
      "588cef06b04342548874de6175440b49",
      "b8f288a1a2554f2489fe88eba2eeae0e",
      "c9ceedc332444559839c57d672009638",
      "a9f303fbcfc14c50834c553ec0c1bfa0",
      "79dadf5599984f24838169f6dac96e0c",
      "f92f0602e273456d82a3ba977d5ba69c",
      "6a8a6e578a3a4305b7d7b9a98c281fa2",
      "75f916c110ff426a8f497dc1415e57bf",
      "2a5a2b65470b48798298d13175937e80",
      "c8c0d436a5d1476fb947adb96957ad39",
      "bcf1c1036d5a48c6adba17297be2bf2a"
     ]
    },
    "execution": {
     "iopub.execute_input": "2023-03-18T06:53:17.639028Z",
     "iopub.status.busy": "2023-03-18T06:53:17.637925Z",
     "iopub.status.idle": "2023-03-18T06:53:17.870473Z",
     "shell.execute_reply": "2023-03-18T06:53:17.869023Z",
     "shell.execute_reply.started": "2023-03-18T06:53:17.638989Z"
    },
    "id": "PragefhdoH0_",
    "outputId": "9a40b765-3f78-4c83-e2ed-aa243cd01542"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5a84902917bf492681cd0e534a9b1717",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/16254 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Series containing all the movies NOT interacted by each user.\n",
    "neg_user_interactions = []\n",
    "num_negatives = 3\n",
    "\n",
    "for u in tqdm(ratings_data_train.userId.unique()):\n",
    "\n",
    "    # Negative Samples\n",
    "    negative_samples = np.random.choice(list(set(all_movieIds) - set(user_interactions[u])), size=len(user_interactions[u]) * num_negatives)\n",
    "    neg_user_interactions.append(negative_samples)\n",
    "    \n",
    "neg_user_interactions = pd.Series(neg_user_interactions)\n",
    "neg_user_interactions.index = user_interactions.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "id": "YGPO5itDLORs"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "userId\n",
       "13        [72386, 142821, 193761, 116539, 158613, 113705...\n",
       "21        [144786, 74740, 165103, 5051, 79010, 139620, 8...\n",
       "25        [169248, 174703, 44861, 176171, 2724, 203381, ...\n",
       "45        [170887, 8667, 109576, 161153, 74698, 80405, 1...\n",
       "85        [152656, 1289, 175937, 154911, 7241, 66295, 36...\n",
       "                                ...                        \n",
       "162469    [81411, 1504, 69387, 2128, 166906, 60579, 1776...\n",
       "162511    [174607, 36083, 70742, 86773, 5474, 6651, 1392...\n",
       "162523    [112625, 8924, 1422, 174601, 152916, 117434, 9...\n",
       "162532    [104480, 908, 174745, 173941, 121129, 69072, 1...\n",
       "162536    [4468, 37475, 143511, 42002, 88382, 33896, 195...\n",
       "Length: 16254, dtype: object"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neg_user_interactions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "execution": {
     "iopub.status.busy": "2023-03-18T06:53:17.871872Z",
     "iopub.status.idle": "2023-03-18T06:53:17.872358Z",
     "shell.execute_reply": "2023-03-18T06:53:17.872134Z",
     "shell.execute_reply.started": "2023-03-18T06:53:17.872111Z"
    },
    "id": "tPWppKzmoH0_"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0e0c5d17d16b4ee4802a5ea29c57eaaa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Generate Dataset (sample):\n",
    "users - User features (userId)\n",
    "items - Item features\n",
    "      - Selected item feature for this project are feature encodings based on movie genres.\n",
    "label - Whether user interacted with the corresponding user or not\n",
    "\"\"\"\n",
    "\n",
    "users, items, labels = [], [], []\n",
    "\n",
    "for u in tqdm(ratings_data_train.userId.unique()[:1000]):\n",
    "\n",
    "    users.extend([u] * (len(user_interactions[u]) + len(neg_user_interactions[u])))\n",
    "    \n",
    "    # user_interacted_movie_features = movies_data.set_index('movieId').loc[user_interactions[u]].iloc[:, 1:-1].values\n",
    "    # neg_user_interacted_movie_features = movies_data.set_index('movieId').loc[neg_user_interactions[u]].iloc[:, 1:-1].values\n",
    "\n",
    "    # items.extend(np.apply_along_axis(encode_movie_features, 1, user_interacted_movie_features).tolist())\n",
    "    # items.extend(np.apply_along_axis(encode_movie_features, 1, neg_user_interacted_movie_features).tolist())\n",
    "\n",
    "    items.extend(user_interactions[u])\n",
    "    items.extend(neg_user_interactions[u])\n",
    "    \n",
    "    labels.extend([1] * len(user_interactions[u]))\n",
    "    labels.extend([0] * len(neg_user_interactions[u]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "id": "K6_uuco0HIZH"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Length of user list (sample): 600632\n",
      "Length of item list (sample): 600632\n",
      "Length of label list (sample): 600632\n"
     ]
    }
   ],
   "source": [
    "print(f\"Length of user list (sample): {len(users)}\")\n",
    "print(f\"Length of item list (sample): {len(items)}\")\n",
    "print(f\"Length of label list (sample): {len(labels)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "id": "3i40hgsb6m8T"
   },
   "outputs": [],
   "source": [
    "# def createMovieEmbedding(item, movies_data = movies_data, num_embeddings = 19, embedding_dim = 8, padding_idx = 0):\n",
    "\n",
    "#   item_embedding = nn.Embedding(num_embeddings=num_embeddings, embedding_dim=embedding_dim, padding_idx=padding_idx)\n",
    "#   input = torch.LongTensor([np.squeeze(np.argwhere(movies_data.set_index('movieId').iloc[item, 1:-1].values == 1)).tolist()])\n",
    "\n",
    "#   return item_embedding(input).mean(dim=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "id": "LXbqSAOVniml"
   },
   "outputs": [],
   "source": [
    "sample_items = [     1,     19,     32,     34,     47,     60,     70,    104,    110,\n",
    "           112,    141,    150,    153,    173,    208,    216,    231,    256,\n",
    "           260,    277,    296,    303,    318,    344,    356,    364,    367,\n",
    "           377,    432,    441,    455,    480,    520,    541,    551,    553,\n",
    "           589,    592,    593,    595,    648,    736,    745,    778,    780,\n",
    "           924,   1035,   1049,   1073,   1079,   1089,   1097,   1127,   1136,\n",
    "          1196,   1198,   1200,   1201,   1206,   1210,   1214,   1240,   1265,\n",
    "          1266,   1270,   1274,   1278,   1291,   1359,   1377,   1378,   1441,\n",
    "          1500,   1517,   1527,   1562,   1580,   1584,   1616,   1653,   1704,\n",
    "          1721,   1732,   1748,   1921,   1923,   2011,   2012,   2028,   2046,\n",
    "          2060,   2115,   2141,   2142,   2232,   2288,   2296,   2329,   2355,\n",
    "          2366,   2501,   2502,   2542,   2571,   2628,   2683,   2701,   2710,\n",
    "          2716,   2746,   2761,   2762,   2826,   2858,   2916,   2918,   2959,\n",
    "          3000,   3033,   3074,   3100,   3113,   3147,   3175,   3210,   3275,\n",
    "          3300,   3354,   3418,   3471,   3508,   3527,   3578,   3624,   3671,\n",
    "          3681,   3698,   3703,   3745,   3752,   3793,   3916,   3948,   3972,\n",
    "          3981,   3986,   3994,   3996,   4011,   4027,   4161,   4226,   4247,\n",
    "          4306,   4321,   4343,   4370,   4558,   4643,   4721,   4816,   4878,\n",
    "          4886,   4887,   4896,   4963,   4993,   5010,   5072,   5110,   5219,\n",
    "          5244,   5349,   5378,   5418,   5444,   5445,   5459,   5463,   5481,\n",
    "          5502,   5528,   5816,   5881,   5882,   5903,   5952,   5956,   5989,\n",
    "          6188,   6264,   6283,   6294,   6333,   6365,   6377,   6502,   6537,\n",
    "          6539,   6664,   6753,   6857,   6874,   6934,   6936,   6950,   6951,\n",
    "          7004,   7090,   7143,   7147,   7153,   7254,   7360,   7361,   7373,\n",
    "          7438,   7445,   7458,   7482,   8360,   8361,   8366,   8368,   8371,\n",
    "          8533,   8622,   8636,   8641,   8644,   8665,   8784,   8861,   8865,\n",
    "          8874,   8914,   8950,   8957,   8961,   8984,  26925,  27660,  27831,\n",
    "         27904,  30810,  31660,  31878,  32031,  32587,  33004,  33158,  33493,\n",
    "         33495,  33794,  33836,  34048,  34111,  34150,  34162,  34319,  34405,\n",
    "         35836,  37380,  37830,  39381,  40815,  41569,  42718,  43396,  44191,\n",
    "         44195,  44199,  44665,  45499,  45722,  46322,  46578,  47997,  48043,\n",
    "         48394,  48516,  48774,  48780,  49272,  50872,  51255,  51662,  51698,\n",
    "         52281,  52321,  52328,  52722,  52973,  53000,  53121,  53125,  53322,\n",
    "         53519,  53953,  53996,  54001,  54256,  54286,  54503,  54997,  55232,\n",
    "         55247,  55363,  55620,  55820,  55908,  56145,  56174,  56757,  57368,\n",
    "         57640,  57669,  58559,  59103,  59126,  59295,  59315,  59369,  59615,\n",
    "         59784,  60037,  60040,  60069,  60647,  60684,  60756,  61024,  61160,\n",
    "         61289,  61986,  62434,  63072,  63113,  63131,  63853,  65514,  66097,\n",
    "         66171,  66335,  67087,  67408,  68157,  68237,  68319,  68358,  68554,\n",
    "         68791,  68952,  68954,  69122,  69526,  69784,  69844,  69951,  70286,\n",
    "         70336,  71135,  71156,  71379,  71520,  71535,  72226,  72378,  72393,\n",
    "         72489,  72998,  73017,  73321,  73870,  74458,  74688,  74789,  74851,\n",
    "         76077,  76093,  76175,  77364,  77561,  77866,  78105,  78266,  78469,\n",
    "         78499,  78637,  78893,  79091,  79132,  79139,  79185,  79293,  80939,\n",
    "         81537,  81834,  81847,  82459,  82461,  82852,  83349,  84944,  86332,\n",
    "         87869,  88125,  88129,  88163,  88785,  89840,  72386, 142821, 193761,\n",
    "        116539, 158613, 113705,  86539,  26085,  26237,   1293,  47501, 174613,\n",
    "        141978, 142254,   5255,  79627, 192717,  63033,   2362,  45991, 150048,\n",
    "        189149,  32646, 167094,  82276,  59727, 135797,   3094, 182993, 105492,\n",
    "        156244, 129229, 194270, 105597, 155288,   6909,  84807, 132906,   3509,\n",
    "        196957,   2868, 148164,  45852,     11, 140525,   4442, 167018, 106906,\n",
    "         70889, 175885,   4944, 180201, 135508,   8273, 179685,  56474, 139729,\n",
    "         47455,   4755,  73713,  81156,  37824, 115828,  94410, 112743, 145094,\n",
    "         62245,  45732,   1872, 154939,  91485,  26755,  70181,  49929, 102245,\n",
    "         51077, 164536,   3083,  46848,  64234,   8574, 113855,  58367,   8236,\n",
    "           133,  81214,   4977,  59604,  82328, 191987,   6423, 173829, 176085,\n",
    "         74750, 200824, 144464,   3474, 162596, 170669,   5913, 104135]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "id": "kI0nx8N_PsmA"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature Embeddings: torch.Size([512, 8])\n"
     ]
    }
   ],
   "source": [
    "item_input = torch.tensor(sample_items)\n",
    "item_embedding = nn.Embedding(num_embeddings=19, embedding_dim=8, padding_idx=0)\n",
    "\n",
    "item_input = item_input.tolist()\n",
    "feature_vector = movies_data.set_index('movieId').loc[item_input].iloc[:, 1:-1].values.tolist()\n",
    "\n",
    "feature_vector_positive_class = [torch.tensor(np.squeeze(np.argwhere(i).tolist())) for i in feature_vector]\n",
    "\n",
    "#feature_vectors = [torch.LongTensor(np.squeeze(np.argwhere(i).tolist())) for i in movies_data.set_index('movieId').loc[sample_item].iloc[:, 1:-1].values.tolist()]\n",
    "feature_embeddings = torch.stack([torch.reshape(item_embedding(i), (-1, 8)).mean(dim=0) for i in feature_vector_positive_class])\n",
    "\n",
    "#print(f'Feature Vectors: {feature_vectors}\\n')\n",
    "print(f'Feature Embeddings: {feature_embeddings.shape}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "id": "z2fKV8tQLniE"
   },
   "outputs": [],
   "source": [
    "# # Free up memory\n",
    "# del users, items, labels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "BYgE9WoZB2dm"
   },
   "source": [
    "### Build Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {
    "execution": {
     "iopub.status.busy": "2023-03-18T06:53:17.876029Z",
     "iopub.status.idle": "2023-03-18T06:53:17.876542Z",
     "shell.execute_reply": "2023-03-18T06:53:17.876298Z",
     "shell.execute_reply.started": "2023-03-18T06:53:17.876274Z"
    },
    "id": "T256PMNFoH0_"
   },
   "outputs": [],
   "source": [
    "class MovieLensTrainDataset(Dataset):\n",
    "    \n",
    "    \"\"\"MovieLens PyTorch Dataset for Training\n",
    "    \n",
    "    Args:\n",
    "        ratings (pd.DataFrame): Dataframe containing the movie ratings\n",
    "        all_movieIds (list): List containing all movieIds\n",
    "    \n",
    "    \"\"\"\n",
    "\n",
    "    def __init__(self, ratings_data, user_interactions, neg_user_interactions):\n",
    "        self.users, self.items, self.labels = self.get_dataset(ratings_data, user_interactions, neg_user_interactions)\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.users)\n",
    "  \n",
    "    def __getitem__(self, idx):\n",
    "        return self.users[idx], self.items[idx], self.labels[idx]\n",
    "\n",
    "    def get_dataset(self, ratings_data, user_interactions, neg_user_interactions):\n",
    "        users, items, labels = [], [], []\n",
    "\n",
    "        for u in tqdm(ratings_data.userId.unique()):\n",
    "\n",
    "            users.extend([u] * (len(user_interactions[u]) + len(neg_user_interactions[u])))\n",
    "            \n",
    "            # user_interacted_movie_features = movies_data.set_index('movieId').loc[user_interactions[u]].iloc[:, 1:-1].values\n",
    "            # neg_user_interacted_movie_features = movies_data.set_index('movieId').loc[neg_user_interactions[u]].iloc[:, 1:-1].values\n",
    "\n",
    "            # items.extend(np.apply_along_axis(encode_movie_features, 1, user_interacted_movie_features).tolist())\n",
    "            # items.extend(np.apply_along_axis(encode_movie_features, 1, neg_user_interacted_movie_features).tolist())\n",
    "\n",
    "            items.extend(user_interactions[u])\n",
    "            items.extend(neg_user_interactions[u])\n",
    "            \n",
    "            labels.extend([1] * len(user_interactions[u]))\n",
    "            labels.extend([0] * len(neg_user_interactions[u]))\n",
    "\n",
    "        return torch.tensor(users), torch.tensor(items, device=torch.device('cuda:0')), torch.tensor(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {
    "execution": {
     "iopub.status.busy": "2023-03-18T06:53:17.877986Z",
     "iopub.status.idle": "2023-03-18T06:53:17.878910Z",
     "shell.execute_reply": "2023-03-18T06:53:17.878679Z",
     "shell.execute_reply.started": "2023-03-18T06:53:17.878651Z"
    },
    "id": "PeiKbQE4oH0_"
   },
   "outputs": [],
   "source": [
    "## NEXT TASK: Training too slow! how to set all tensors to use cuda by default\n",
    "\n",
    "class NCF(pl.LightningModule):\n",
    "    \n",
    "    def __init__(self, num_users, num_items, ratings_data, user_interactions, neg_user_interactions, movies_data):\n",
    "        super().__init__()\n",
    "        self.user_embedding = nn.Embedding(num_embeddings=num_users, embedding_dim=8)\n",
    "        self.item_embedding = nn.Embedding(num_embeddings=19, embedding_dim=8, padding_idx=0)\n",
    "        self.fc1 = nn.Linear(in_features=16, out_features=64)\n",
    "        self.fc2 = nn.Linear(in_features=64, out_features=32)\n",
    "        self.output = nn.Linear(in_features=32, out_features=1)\n",
    "        self.ratings_data = ratings_data\n",
    "        self.movies_data = movies_data\n",
    "        self.user_interactions = user_interactions\n",
    "        self.neg_user_interactions = neg_user_interactions\n",
    "        \n",
    "    def forward(self, user_input, item_input):\n",
    "\n",
    "        # Convert item tensor to list\n",
    "        item_input = item_input.tolist()\n",
    "\n",
    "        # Query feature vector from movies data df using item_input\n",
    "        feature_vector = self.movies_data.set_index('movieId').loc[item_input].iloc[:, 1:-1].values.tolist()\n",
    "\n",
    "        # For each feature vector query, get the location of each positive class to use as input for the embeddings\n",
    "        # ex. feature vector: [1, 0, 1, 0, 1, 1, 0, 0, 0] --> positive class locations: [[0, 2, 4, 5]] --> input to embedding layer\n",
    "        feature_vector_positive_class = [torch.tensor(np.squeeze(np.argwhere(i).tolist()), device=self.device) for i in feature_vector]\n",
    "\n",
    "        # Pass user input through embedding layers\n",
    "        user_embedded = self.user_embedding(user_input)\n",
    "        item_embedded = torch.stack([torch.reshape(self.item_embedding(i), (-1, 8)).mean(dim=0) for i in feature_vector_positive_class])\n",
    "\n",
    "        # Concat the two embedding layers\n",
    "        vector = torch.cat([user_embedded, item_embedded], dim=-1)\n",
    "\n",
    "        # Pass through dense layer\n",
    "        vector = nn.ReLU()(self.fc1(vector))\n",
    "        vector = nn.ReLU()(self.fc2(vector))\n",
    "\n",
    "        # Output layer\n",
    "        #pred = nn.ReLU()(self.output(vector)) ##\n",
    "        pred = nn.Sigmoid()(self.output(vector))\n",
    "\n",
    "        return pred\n",
    "    \n",
    "    def training_step(self, batch, batch_idx):\n",
    "        user_input, item_input, labels = batch\n",
    "        predicted_labels = self(user_input, item_input)\n",
    "        #loss = nn.BCEWithLogitsLoss()(predicted_labels, labels.view(-1, 1).float()) ##\n",
    "        loss = nn.BCELoss()(predicted_labels, labels.view(-1, 1).float())\n",
    "        return loss\n",
    "\n",
    "    def configure_optimizers(self):\n",
    "        return torch.optim.Adam(self.parameters())\n",
    "\n",
    "    def train_dataloader(self):\n",
    "        return DataLoader(MovieLensTrainDataset(self.ratings_data, self.user_interactions, self.neg_user_interactions),\n",
    "                          batch_size=512) #, num_workers=4 -->results to pid error in windows"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HY8bVpcTB6bn"
   },
   "source": [
    "### Train Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {
    "execution": {
     "iopub.status.busy": "2023-03-18T06:53:17.880786Z",
     "iopub.status.idle": "2023-03-18T06:53:17.881269Z",
     "shell.execute_reply": "2023-03-18T06:53:17.881037Z",
     "shell.execute_reply.started": "2023-03-18T06:53:17.881014Z"
    },
    "id": "60zrViKToH0_"
   },
   "outputs": [],
   "source": [
    "num_users = ratings_data['userId'].max()+1\n",
    "num_items = ratings_data['movieId'].max()+1 #movie_features_lookup.shape[0]\n",
    "\n",
    "# Instantiate model object / Load Data\n",
    "model = NCF(num_users, num_items, ratings_data_train, user_interactions, neg_user_interactions, movies_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The flag below controls whether to allow TF32 on matmul. This flag defaults to False\n",
    "# in PyTorch 1.12 and later.\n",
    "torch.backends.cuda.matmul.allow_tf32 = True\n",
    "\n",
    "# The flag below controls whether to allow TF32 on cuDNN. This flag defaults to True.\n",
    "torch.backends.cudnn.allow_tf32 = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 410,
     "referenced_widgets": [
      "fba3e9929706429eb158cc958f58162f",
      "a090a2a04e014f2fae2f20a74427ef6a",
      "00efd94dc2ac49daaf4cd85b9ac7ac1f",
      "63200b6f061f40f8af1b6e9d41ca89cc",
      "cfa71eb75820485e8f3075d16d0b7469",
      "f011ece7224e4a2ebda5edc79f767cd0",
      "c79da2cab7f34eaeb7214a5c95d34f47",
      "7e3691ef12fc4e63bc310ff52da1b229",
      "69e82c2ac3af46ac94995b76acf19198",
      "6b3c1d0bf84a48ce84825b743f9634fc",
      "870696dfddde4ac3be3f669833a07985",
      "7f50eeee47fc4d619f45c6b0f6e4d6b7",
      "ed67c458a0e4488487bf0b5d89eca782",
      "e71988ca0d7647d59a1fd4727a85d30c",
      "06f13e915f154de9971aea7d82b1804d",
      "b4418019df534648adfd4e400ae14ddc",
      "302de01dd2d8456e93e627e40f885681",
      "2bfc2ce544c740b999dd3ae2995b0ddf",
      "39d2dee7b6854ffa946874398c9280be",
      "9846ff1153fc45c2bffd05feb944a15c",
      "2990d8a913824059bf777a374aa476d3",
      "e26f9fc2825b486e932cf7d6e73748b8"
     ]
    },
    "id": "dmE0LwbbFOo5",
    "outputId": "85a9b027-e770-4e97-9785-f1a9addc2f58"
   },
   "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",
      "Restoring states from the checkpoint path at lightning_logs/version_4/checkpoints/epoch=2-step=58212.ckpt\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
      "\n",
      "  | Name           | Type      | Params\n",
      "---------------------------------------------\n",
      "0 | user_embedding | Embedding | 1.3 M \n",
      "1 | item_embedding | Embedding | 152   \n",
      "2 | fc1            | Linear    | 1.1 K \n",
      "3 | fc2            | Linear    | 2.1 K \n",
      "4 | output         | Linear    | 33    \n",
      "---------------------------------------------\n",
      "1.3 M     Trainable params\n",
      "0         Non-trainable params\n",
      "1.3 M     Total params\n",
      "5.215     Total estimated model params size (MB)\n",
      "Restored all states from the checkpoint at lightning_logs/version_4/checkpoints/epoch=2-step=58212.ckpt\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6f803087ffc04f5fbe7dbb59735b4ab8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/16254 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3e28854a4ad24d88808c0cd40117498c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "`Trainer.fit` stopped: `max_epochs=6` reached.\n"
     ]
    }
   ],
   "source": [
    "# Train model\n",
    "trainer = pl.Trainer(max_epochs=6, devices=1, accelerator=\"gpu\") #, precision=16\n",
    "trainer.fit(model, ckpt_path='lightning_logs/version_4/checkpoints/epoch=2-step=58212.ckpt') #"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0ap6Q-dwB89-"
   },
   "source": [
    "### Evaluate Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NCF(\n",
       "  (user_embedding): Embedding(162537, 8)\n",
       "  (item_embedding): Embedding(19, 8, padding_idx=0)\n",
       "  (fc1): Linear(in_features=16, out_features=64, bias=True)\n",
       "  (fc2): Linear(in_features=64, out_features=32, bias=True)\n",
       "  (output): Linear(in_features=32, out_features=1, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.eval()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {
    "execution": {
     "iopub.status.busy": "2023-03-18T06:53:17.883754Z",
     "iopub.status.idle": "2023-03-18T06:53:17.884233Z",
     "shell.execute_reply": "2023-03-18T06:53:17.884005Z",
     "shell.execute_reply.started": "2023-03-18T06:53:17.883980Z"
    },
    "id": "VojT_u2xoH1A"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3e4bcd1a649146b3ab22ab1f7e5497bc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generate Interaction Data for Testing\n",
    "user_interactions_test = ratings_data.groupby('userId')['movieId'].apply(list) # list of all movies the user has interacted with (both train and test)\n",
    "\n",
    "neg_user_interactions_test = []\n",
    "\n",
    "user_sample_test = sample(list(ratings_data_test.userId.unique()), 1000)\n",
    "\n",
    "for u in tqdm(user_sample_test):\n",
    "\n",
    "    # Negative Samples\n",
    "    negative_samples = sample(list(set(all_movieIds) - set(user_interactions_test[u])), 99)\n",
    "    neg_user_interactions_test.append(negative_samples)\n",
    "    \n",
    "neg_user_interactions_test = pd.Series(neg_user_interactions_test)\n",
    "neg_user_interactions_test.index = user_sample_test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "f5d8RHXdCFG-"
   },
   "source": [
    "Testing methodology: **HR@10** (Hit Ratio at 10) [1] <br>\n",
    "1. For each user, randomly select 99 items that the user has not interacted with.\n",
    "2. Combine these 99 items with the test item (the actual item that the user last interacted with). We now have 100 items. \n",
    "3. Run the model on these 100 items, and rank them according to their predicted probabilities.\n",
    "4. Select the top 10 items from the list of 100 items. If the test item is present within the top 10 items, then we say that this is a hit.\n",
    "5. Repeat the process for all users. The Hit Ratio is then the average hits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {
    "execution": {
     "iopub.status.busy": "2023-03-18T06:53:17.888146Z",
     "iopub.status.idle": "2023-03-18T06:53:17.888661Z",
     "shell.execute_reply": "2023-03-18T06:53:17.888409Z",
     "shell.execute_reply.started": "2023-03-18T06:53:17.888386Z"
    },
    "id": "BghqIKktoH1A"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cfa84f8f1e4841feb823c064ac1f3d8a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Hit Ratio @ 10 is 0.39\n"
     ]
    }
   ],
   "source": [
    "# Implement HR@10\n",
    "hits = []\n",
    "for (u, neg_interacted_items) in tqdm(zip(neg_user_interactions_test.index, neg_user_interactions_test)):\n",
    "    \n",
    "    interacted_item = ratings_data_test[ratings_data_test.userId == u].movieId.values[0]\n",
    "    \n",
    "    test_items = [interacted_item] + list(neg_interacted_items)\n",
    "\n",
    "    predicted_labels = np.squeeze(model(torch.tensor([u]*100), torch.tensor(test_items)).detach().numpy())\n",
    "\n",
    "    top10_items = [test_items[i] for i in np.argsort(predicted_labels)[-10:]]\n",
    "\n",
    "    if interacted_item in top10_items:\n",
    "        hits.append(1)\n",
    "    else:\n",
    "        hits.append(0)\n",
    "        \n",
    "print(\"The Hit Ratio @ 10 is {:.2f}\".format(np.average(hits)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The Hit Ratio @ 10 is 0.39 --> BCELoss @ 2 epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {
    "id": "X6UqEv8eQlwK"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Item interacted by the user: Dances with Wolves (1990)\n"
     ]
    }
   ],
   "source": [
    "print(f'Item interacted by the user: {movies_data[movies_data.movieId == interacted_item].title.values[0]}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zeT4GLlFSPFm"
   },
   "source": [
    "### Check test movies and results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {
    "id": "XXJpw1IERu2T"
   },
   "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>Adventure</th>\n",
       "      <th>Drama</th>\n",
       "      <th>Western</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>576</th>\n",
       "      <td>590</td>\n",
       "      <td>Dances with Wolves (1990)</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1990</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     movieId                      title  Adventure  Drama  Western  year\n",
       "576      590  Dances with Wolves (1990)        1.0    1.0      1.0  1990"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "movies_data[movies_data.movieId == interacted_item].loc[:, (movies_data[movies_data.movieId == interacted_item] != 0).any(axis=0)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {
    "id": "cs3Nk__qQXVQ"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top 10 recommended items: \n",
      "\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>movieId</th>\n",
       "      <th>title</th>\n",
       "      <th>Adventure</th>\n",
       "      <th>Animation</th>\n",
       "      <th>Comedy</th>\n",
       "      <th>Drama</th>\n",
       "      <th>Action</th>\n",
       "      <th>Crime</th>\n",
       "      <th>Thriller</th>\n",
       "      <th>Sci-Fi</th>\n",
       "      <th>IMAX</th>\n",
       "      <th>Western</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>576</th>\n",
       "      <td>590</td>\n",
       "      <td>Dances with Wolves (1990)</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3573</th>\n",
       "      <td>3744</td>\n",
       "      <td>Shaft (2000)</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7659</th>\n",
       "      <td>8597</td>\n",
       "      <td>Riders (2002)</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17809</th>\n",
       "      <td>110102</td>\n",
       "      <td>Captain America: The Winter Soldier (2014)</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2014</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18116</th>\n",
       "      <td>112807</td>\n",
       "      <td>Gunbuster (Top wo Narae) (1988)</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18247</th>\n",
       "      <td>113775</td>\n",
       "      <td>Life of Crime (2013)</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22894</th>\n",
       "      <td>148080</td>\n",
       "      <td>Hard to Be a God (1989)</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25822</th>\n",
       "      <td>169906</td>\n",
       "      <td>The Night Of (2016)</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26814</th>\n",
       "      <td>175673</td>\n",
       "      <td>The Life and Death of John Gotti (2017)</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30038</th>\n",
       "      <td>200514</td>\n",
       "      <td>Full Contact (1992)</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1992</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId                                       title  Adventure  \\\n",
       "576        590                   Dances with Wolves (1990)        1.0   \n",
       "3573      3744                                Shaft (2000)        0.0   \n",
       "7659      8597                               Riders (2002)        0.0   \n",
       "17809   110102  Captain America: The Winter Soldier (2014)        1.0   \n",
       "18116   112807             Gunbuster (Top wo Narae) (1988)        0.0   \n",
       "18247   113775                        Life of Crime (2013)        0.0   \n",
       "22894   148080                     Hard to Be a God (1989)        1.0   \n",
       "25822   169906                         The Night Of (2016)        0.0   \n",
       "26814   175673     The Life and Death of John Gotti (2017)        0.0   \n",
       "30038   200514                         Full Contact (1992)        0.0   \n",
       "\n",
       "       Animation  Comedy  Drama  Action  Crime  Thriller  Sci-Fi  IMAX  \\\n",
       "576          0.0     0.0    1.0     0.0    0.0       0.0     0.0   0.0   \n",
       "3573         0.0     0.0    0.0     1.0    1.0       1.0     0.0   0.0   \n",
       "7659         0.0     0.0    0.0     1.0    1.0       1.0     0.0   0.0   \n",
       "17809        0.0     0.0    0.0     1.0    0.0       0.0     1.0   1.0   \n",
       "18116        1.0     0.0    1.0     1.0    0.0       0.0     1.0   0.0   \n",
       "18247        0.0     1.0    0.0     0.0    1.0       0.0     0.0   0.0   \n",
       "22894        0.0     0.0    1.0     0.0    0.0       0.0     1.0   0.0   \n",
       "25822        0.0     0.0    1.0     0.0    1.0       0.0     0.0   0.0   \n",
       "26814        0.0     0.0    1.0     0.0    1.0       0.0     0.0   0.0   \n",
       "30038        0.0     0.0    0.0     1.0    1.0       1.0     0.0   0.0   \n",
       "\n",
       "       Western  year  \n",
       "576        1.0  1990  \n",
       "3573       0.0  2000  \n",
       "7659       0.0  2002  \n",
       "17809      0.0  2014  \n",
       "18116      0.0  1988  \n",
       "18247      0.0  2013  \n",
       "22894      0.0  1989  \n",
       "25822      0.0  2016  \n",
       "26814      0.0  2017  \n",
       "30038      0.0  1992  "
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('Top 10 recommended items: \\n')\n",
    "movies_data[movies_data.movieId.isin(top10_items)].loc[:, (movies_data[movies_data.movieId.isin(top10_items)] != 0).any(axis=0)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "FV82IgIaPFVx"
   },
   "source": [
    "### References:\n",
    "\n",
    "1. [Loy, James, \"Deep Learning based Recommender Systems\"](https://towardsdatascience.com/deep-learning-based-recommender-systems-3d120201db7e)\n",
    "\n",
    "2. [He, et. al, \"Neural Collaborative Filtering](https://arxiv.org/abs/1708.05031)\n",
    "\n",
    "3. [Embedding a sparse vector of multiple items](https://developers.google.com/machine-learning/crash-course/embeddings/translating-to-a-lower-dimensional-space)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "acY9mCh4sNkj"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "provenance": []
  },
  "gpuClass": "standard",
  "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": "3.9.7"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "00efd94dc2ac49daaf4cd85b9ac7ac1f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_7e3691ef12fc4e63bc310ff52da1b229",
      "max": 16254,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_69e82c2ac3af46ac94995b76acf19198",
      "value": 16254
     }
    },
    "06f13e915f154de9971aea7d82b1804d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_2990d8a913824059bf777a374aa476d3",
      "placeholder": "​",
      "style": "IPY_MODEL_e26f9fc2825b486e932cf7d6e73748b8",
      "value": " 4300/19404 [06:28&lt;22:45, 11.06it/s]"
     }
    },
    "2990d8a913824059bf777a374aa476d3": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2a5a2b65470b48798298d13175937e80": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "2bfc2ce544c740b999dd3ae2995b0ddf": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "302de01dd2d8456e93e627e40f885681": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "39d2dee7b6854ffa946874398c9280be": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": "2",
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "588cef06b04342548874de6175440b49": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_b8f288a1a2554f2489fe88eba2eeae0e",
       "IPY_MODEL_c9ceedc332444559839c57d672009638",
       "IPY_MODEL_a9f303fbcfc14c50834c553ec0c1bfa0"
      ],
      "layout": "IPY_MODEL_79dadf5599984f24838169f6dac96e0c"
     }
    },
    "63200b6f061f40f8af1b6e9d41ca89cc": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_6b3c1d0bf84a48ce84825b743f9634fc",
      "placeholder": "​",
      "style": "IPY_MODEL_870696dfddde4ac3be3f669833a07985",
      "value": " 16254/16254 [00:01&lt;00:00, 10404.45it/s]"
     }
    },
    "69e82c2ac3af46ac94995b76acf19198": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "6a8a6e578a3a4305b7d7b9a98c281fa2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "6b3c1d0bf84a48ce84825b743f9634fc": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "75f916c110ff426a8f497dc1415e57bf": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "79dadf5599984f24838169f6dac96e0c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "7e3691ef12fc4e63bc310ff52da1b229": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "7f50eeee47fc4d619f45c6b0f6e4d6b7": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_ed67c458a0e4488487bf0b5d89eca782",
       "IPY_MODEL_e71988ca0d7647d59a1fd4727a85d30c",
       "IPY_MODEL_06f13e915f154de9971aea7d82b1804d"
      ],
      "layout": "IPY_MODEL_b4418019df534648adfd4e400ae14ddc"
     }
    },
    "870696dfddde4ac3be3f669833a07985": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "9846ff1153fc45c2bffd05feb944a15c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "a090a2a04e014f2fae2f20a74427ef6a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f011ece7224e4a2ebda5edc79f767cd0",
      "placeholder": "​",
      "style": "IPY_MODEL_c79da2cab7f34eaeb7214a5c95d34f47",
      "value": "100%"
     }
    },
    "a9f303fbcfc14c50834c553ec0c1bfa0": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_c8c0d436a5d1476fb947adb96957ad39",
      "placeholder": "​",
      "style": "IPY_MODEL_bcf1c1036d5a48c6adba17297be2bf2a",
      "value": " 7604/16254 [01:03&lt;00:56, 153.73it/s]"
     }
    },
    "b4418019df534648adfd4e400ae14ddc": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": "inline-flex",
      "flex": null,
      "flex_flow": "row wrap",
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "100%"
     }
    },
    "b8f288a1a2554f2489fe88eba2eeae0e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f92f0602e273456d82a3ba977d5ba69c",
      "placeholder": "​",
      "style": "IPY_MODEL_6a8a6e578a3a4305b7d7b9a98c281fa2",
      "value": " 47%"
     }
    },
    "bcf1c1036d5a48c6adba17297be2bf2a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "c79da2cab7f34eaeb7214a5c95d34f47": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "c8c0d436a5d1476fb947adb96957ad39": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "c9ceedc332444559839c57d672009638": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "danger",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_75f916c110ff426a8f497dc1415e57bf",
      "max": 16254,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_2a5a2b65470b48798298d13175937e80",
      "value": 7604
     }
    },
    "cfa71eb75820485e8f3075d16d0b7469": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e26f9fc2825b486e932cf7d6e73748b8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "e71988ca0d7647d59a1fd4727a85d30c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_39d2dee7b6854ffa946874398c9280be",
      "max": 19404,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_9846ff1153fc45c2bffd05feb944a15c",
      "value": 4320
     }
    },
    "ed67c458a0e4488487bf0b5d89eca782": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_302de01dd2d8456e93e627e40f885681",
      "placeholder": "​",
      "style": "IPY_MODEL_2bfc2ce544c740b999dd3ae2995b0ddf",
      "value": "Epoch 2:  22%"
     }
    },
    "f011ece7224e4a2ebda5edc79f767cd0": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "f92f0602e273456d82a3ba977d5ba69c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "fba3e9929706429eb158cc958f58162f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_a090a2a04e014f2fae2f20a74427ef6a",
       "IPY_MODEL_00efd94dc2ac49daaf4cd85b9ac7ac1f",
       "IPY_MODEL_63200b6f061f40f8af1b6e9d41ca89cc"
      ],
      "layout": "IPY_MODEL_cfa71eb75820485e8f3075d16d0b7469"
     }
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}