movie_recommendation_using_NN.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Neural Collaborative Filtering (NCF)\n",
    "The goal is to create a NCF deep learning model that will provide the user movie recommendations based off of the users reviews as well as similar movies to create detailed movie recommendations. The goal of this model is to build a collaborative filtering model using deep neural learning to offer a top 10 list of movie recommendations from a selected user."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import dependencies\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sqlalchemy import create_engine\n",
    "from sqlalchemy import inspect\n",
    "import config\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Log into the SQL database to retrieve the data\n",
    "protocol = 'postgresql'\n",
    "username = config.Username\n",
    "password = config.Password\n",
    "host = 'localhost'\n",
    "port = 5432\n",
    "database_name = 'movies_db'\n",
    "rds_connection_string = f'{protocol}://{username}:{password}@{host}:{port}/{database_name}'\n",
    "engine = create_engine(rds_connection_string)\n",
    "insp = inspect(engine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Perform an SQL search query to select all of the data within the data set and create a pandas data frame from it\n",
    "sql_join = r\"\"\"SELECT movies.movie_id, movies.title, movies.genres, ratings.user_id, ratings.rating, ratings.timestamps\n",
    "FROM movies\n",
    "INNER JOIN ratings\n",
    "ON movies.movie_id=ratings.movie_id;\"\"\"\n",
    "joined_movies_df=pd.read_sql_query(sql_join, con=engine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movie_id</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "      <th>user_id</th>\n",
       "      <th>rating</th>\n",
       "      <th>timestamps</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>Adventure|Animation|Children|Comedy|Fantasy</td>\n",
       "      <td>1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>964982703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>Grumpier Old Men (1995)</td>\n",
       "      <td>Comedy|Romance</td>\n",
       "      <td>1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>964981247</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6</td>\n",
       "      <td>Heat (1995)</td>\n",
       "      <td>Action|Crime|Thriller</td>\n",
       "      <td>1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>964982224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>47</td>\n",
       "      <td>Seven (a.k.a. Se7en) (1995)</td>\n",
       "      <td>Mystery|Thriller</td>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>964983815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>50</td>\n",
       "      <td>Usual Suspects, The (1995)</td>\n",
       "      <td>Crime|Mystery|Thriller</td>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>964982931</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100831</th>\n",
       "      <td>166534</td>\n",
       "      <td>Split (2017)</td>\n",
       "      <td>Drama|Horror|Thriller</td>\n",
       "      <td>610</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1493848402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100832</th>\n",
       "      <td>168248</td>\n",
       "      <td>John Wick: Chapter Two (2017)</td>\n",
       "      <td>Action|Crime|Thriller</td>\n",
       "      <td>610</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1493850091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100833</th>\n",
       "      <td>168250</td>\n",
       "      <td>Get Out (2017)</td>\n",
       "      <td>Horror</td>\n",
       "      <td>610</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1494273047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100834</th>\n",
       "      <td>168252</td>\n",
       "      <td>Logan (2017)</td>\n",
       "      <td>Action|Sci-Fi</td>\n",
       "      <td>610</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1493846352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100835</th>\n",
       "      <td>170875</td>\n",
       "      <td>The Fate of the Furious (2017)</td>\n",
       "      <td>Action|Crime|Drama|Thriller</td>\n",
       "      <td>610</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1493846415</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100836 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        movie_id                           title  \\\n",
       "0              1                Toy Story (1995)   \n",
       "1              3         Grumpier Old Men (1995)   \n",
       "2              6                     Heat (1995)   \n",
       "3             47     Seven (a.k.a. Se7en) (1995)   \n",
       "4             50      Usual Suspects, The (1995)   \n",
       "...          ...                             ...   \n",
       "100831    166534                    Split (2017)   \n",
       "100832    168248   John Wick: Chapter Two (2017)   \n",
       "100833    168250                  Get Out (2017)   \n",
       "100834    168252                    Logan (2017)   \n",
       "100835    170875  The Fate of the Furious (2017)   \n",
       "\n",
       "                                             genres  user_id  rating  \\\n",
       "0       Adventure|Animation|Children|Comedy|Fantasy        1     4.0   \n",
       "1                                    Comedy|Romance        1     4.0   \n",
       "2                             Action|Crime|Thriller        1     4.0   \n",
       "3                                  Mystery|Thriller        1     5.0   \n",
       "4                            Crime|Mystery|Thriller        1     5.0   \n",
       "...                                             ...      ...     ...   \n",
       "100831                        Drama|Horror|Thriller      610     4.0   \n",
       "100832                        Action|Crime|Thriller      610     5.0   \n",
       "100833                                       Horror      610     5.0   \n",
       "100834                                Action|Sci-Fi      610     5.0   \n",
       "100835                  Action|Crime|Drama|Thriller      610     3.0   \n",
       "\n",
       "        timestamps  \n",
       "0        964982703  \n",
       "1        964981247  \n",
       "2        964982224  \n",
       "3        964983815  \n",
       "4        964982931  \n",
       "...            ...  \n",
       "100831  1493848402  \n",
       "100832  1493850091  \n",
       "100833  1494273047  \n",
       "100834  1493846352  \n",
       "100835  1493846415  \n",
       "\n",
       "[100836 rows x 6 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Now we have ported in the all the data from our SQL lets visualise it\n",
    "joined_movies_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>movie_id</th>\n",
       "      <th>rating</th>\n",
       "      <th>timestamps</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>964982703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4.0</td>\n",
       "      <td>964981247</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>4.0</td>\n",
       "      <td>964982224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>47</td>\n",
       "      <td>5.0</td>\n",
       "      <td>964983815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>50</td>\n",
       "      <td>5.0</td>\n",
       "      <td>964982931</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>100831</th>\n",
       "      <td>610</td>\n",
       "      <td>166534</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1493848402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100832</th>\n",
       "      <td>610</td>\n",
       "      <td>168248</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1493850091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100833</th>\n",
       "      <td>610</td>\n",
       "      <td>168250</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1494273047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100834</th>\n",
       "      <td>610</td>\n",
       "      <td>168252</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1493846352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100835</th>\n",
       "      <td>610</td>\n",
       "      <td>170875</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1493846415</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100836 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        user_id  movie_id  rating  timestamps\n",
       "0             1         1     4.0   964982703\n",
       "1             1         3     4.0   964981247\n",
       "2             1         6     4.0   964982224\n",
       "3             1        47     5.0   964983815\n",
       "4             1        50     5.0   964982931\n",
       "...         ...       ...     ...         ...\n",
       "100831      610    166534     4.0  1493848402\n",
       "100832      610    168248     5.0  1493850091\n",
       "100833      610    168250     5.0  1494273047\n",
       "100834      610    168252     5.0  1493846352\n",
       "100835      610    170875     3.0  1493846415\n",
       "\n",
       "[100836 rows x 4 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Not all of that data is needed yet, but will be later in the process. We only need the users feedback data\n",
    "user_df = joined_movies_df[['user_id','movie_id','rating', 'timestamps']]\n",
    "user_df"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Processing\n",
    "The data that has now been imported needs to be processed. The 'user_id's and 'movie_id's are being enumerated as the machine learning will be more effective when the data is 0 indexed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\pereg\\AppData\\Local\\Temp\\ipykernel_12708\\2628197176.py:7: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  user_df[\"user_enumerated\"] = user_df[\"user_id\"].map(User_list_enumerated)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0           0\n",
       "1           0\n",
       "2           0\n",
       "3           0\n",
       "4           0\n",
       "         ... \n",
       "100831    609\n",
       "100832    609\n",
       "100833    609\n",
       "100834    609\n",
       "100835    609\n",
       "Name: user_enumerated, Length: 100836, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We are going to create a unique list of every user within our dataset\n",
    "User_list = user_df[\"user_id\"].unique().tolist()\n",
    "#Next we are going to enumerate each user within the list, this will show their user id and the order in which it appears within the data set.\n",
    "#This is important as we may have missing users, it is also data that is used to construct the matrix later on to give recommendations.\n",
    "User_list_enumerated = {x: i for i, x in enumerate(User_list)}\n",
    "#Now we map the enumerated data to the user_id's and add it to the user_enumerated column\n",
    "user_df[\"user_enumerated\"] = user_df[\"user_id\"].map(User_list_enumerated)\n",
    "user_df[\"user_enumerated\"]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although the 'user_id's are not the only data we are using, we are aiming to include movies in the deep learning model as well. This means we need to enumerate the 'movie_id' so we can use them later in the process."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0            0\n",
       "1            1\n",
       "2            2\n",
       "3            3\n",
       "4            4\n",
       "          ... \n",
       "100831    3120\n",
       "100832    2035\n",
       "100833    3121\n",
       "100834    1392\n",
       "100835    2873\n",
       "Name: movies_enumerated, Length: 100836, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We are going to create a unique list of every movie within our dataset\n",
    "movie_list = user_df[\"movie_id\"].unique().tolist()\n",
    "#Next we are going to enumerate each user within the list, this will show their user id and the order in which it appears within the data set.\n",
    "#This is important as we may have missing users, it is also data that is used to construct the matrix later on to give recommendations.\n",
    "movie_list_enumerated = {x: i for i, x in enumerate(movie_list)}\n",
    "#Now we map the enumerated data to the user_id's and add it to the user_enumerated column\n",
    "user_df[\"movies_enumerated\"] = user_df[\"movie_id\"].map(movie_list_enumerated)\n",
    "user_df[\"movies_enumerated\"]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The next step is to split the data and create th training data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[   0,    0],\n",
       "       [   0,    1],\n",
       "       [   0,    2],\n",
       "       ...,\n",
       "       [ 609, 3121],\n",
       "       [ 609, 1392],\n",
       "       [ 609, 2873]], dtype=int64)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating X values as \"user_enumerated\" & \"movies_enumerated\" in an array\n",
    "x = user_df[[\"user_enumerated\", \"movies_enumerated\"]].values\n",
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.77777778, 0.77777778, 0.77777778, ..., 1.        , 1.        ,\n",
       "       0.55555556])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Normalising the rating data between 0 and 1 makes the data easier to train\n",
    "\n",
    "#Creating the min and max saves time when processing the data\n",
    "min_rating = np.min(user_df[\"rating\"])\n",
    "max_rating = np.max(user_df[\"rating\"])\n",
    "\n",
    "#Normalising the rating data\n",
    "y = user_df[\"rating\"].apply(lambda x: (x - min_rating) / (max_rating - min_rating)).values\n",
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "90752"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Splitting the data so 90% is used for training whilst 10% is used for results\n",
    "train_indices = int(0.9 * user_df.shape[0])\n",
    "train_indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Creating the training data.\n",
    "x_train, x_test, y_train, y_test = (\n",
    "    x[:train_indices],\n",
    "    x[train_indices:],\n",
    "    y[:train_indices],\n",
    "    y[train_indices:],\n",
    ")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Building a deep learning model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Building the deep learning model\n",
    "model = Sequential()\n",
    "model.add(Dense(20, activation='relu'))\n",
    "model.add(Dense(10, activation='relu'))\n",
    "model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "2836/2836 [==============================] - 5s 2ms/step - loss: 8.2870 - accuracy: 0.0137\n",
      "Epoch 2/5\n",
      "2836/2836 [==============================] - 4s 2ms/step - loss: 8.1071 - accuracy: 0.0137\n",
      "Epoch 3/5\n",
      "2836/2836 [==============================] - 5s 2ms/step - loss: 7.1500 - accuracy: 0.0137\n",
      "Epoch 4/5\n",
      "2836/2836 [==============================] - 5s 2ms/step - loss: 7.1460 - accuracy: 0.0137\n",
      "Epoch 5/5\n",
      "2836/2836 [==============================] - 4s 1ms/step - loss: 7.1471 - accuracy: 0.0137\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1c9b051ac70>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting the training data to the model\n",
    "model.fit(x_train, y_train, epochs=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "316/316 [==============================] - 1s 2ms/step - loss: 7.3052 - accuracy: 0.0127\n",
      "[7.30519437789917, 0.0126933753490448]\n"
     ]
    }
   ],
   "source": [
    "#Evaluate the keras model, [Loss, Accuracy]\n",
    "accuracy = model.evaluate(x_test, y_test, verbose=1)\n",
    "print(accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Using some of the columns from the imported data\n",
    "movie_df = joined_movies_df[[\"movie_id\",\"title\",\"genres\"]]\n",
    "#Select the user to recommend\n",
    "userId = 315\n",
    "#Find movies the user has seen and movies the user has not seen\n",
    "movies_watched_by_user = user_df[user_df.user_id == userId]\n",
    "#Make movies not watched a pandas series so it can be read into the learning model\n",
    "movies_not_watched = movie_df[~movie_df[\"movie_id\"].isin(movies_watched_by_user.movie_id.values)][\"movie_id\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "303/303 [==============================] - 0s 1ms/step\n"
     ]
    }
   ],
   "source": [
    "#Takes the movies not watched and finds their enumerated keys\n",
    "movies_not_watched = list(\n",
    "    set(movies_not_watched).intersection(set(movie_list_enumerated.keys()))\n",
    ")\n",
    "movies_not_watched = [[movie_list_enumerated.get(x)] for x in movies_not_watched]\n",
    "user_encoder = User_list_enumerated.get(userId)\n",
    "#Formats the users movies they haven't watched for model.predict\n",
    "user_movie_array = np.hstack(\n",
    "    ([[user_encoder]] * len(movies_not_watched), movies_not_watched)\n",
    ")\n",
    "#Predicts the best suited movies for the selected user\n",
    "movie_recs = model.predict(user_movie_array).flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1320, 90540, 92440, 91910, 91720, 89170, 79720, 75420, 74540,\n",
       "       74110], dtype=int64)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Sorts the movies into a top 10\n",
    "top_10_movie_recs = movie_recs.argsort()[-10:][::-1]\n",
    "top_10_movie_recs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movie_id</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1320</th>\n",
       "      <td>2002</td>\n",
       "      <td>Lethal Weapon 3 (1992)</td>\n",
       "      <td>Action|Comedy|Crime|Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74110</th>\n",
       "      <td>3566</td>\n",
       "      <td>Big Kahuna, The (2000)</td>\n",
       "      <td>Comedy|Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74540</th>\n",
       "      <td>6101</td>\n",
       "      <td>Missing (1982)</td>\n",
       "      <td>Drama|Mystery|Thriller</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75420</th>\n",
       "      <td>595</td>\n",
       "      <td>Beauty and the Beast (1991)</td>\n",
       "      <td>Animation|Children|Fantasy|Musical|Romance|IMAX</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79720</th>\n",
       "      <td>97938</td>\n",
       "      <td>Life of Pi (2012)</td>\n",
       "      <td>Adventure|Drama|IMAX</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>89170</th>\n",
       "      <td>2146</td>\n",
       "      <td>St. Elmo's Fire (1985)</td>\n",
       "      <td>Drama|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90540</th>\n",
       "      <td>193</td>\n",
       "      <td>Showgirls (1995)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91720</th>\n",
       "      <td>8620</td>\n",
       "      <td>Exterminating Angel, The (Ángel exterminador, ...</td>\n",
       "      <td>Comedy|Drama|Fantasy|Mystery</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91910</th>\n",
       "      <td>4979</td>\n",
       "      <td>Royal Tenenbaums, The (2001)</td>\n",
       "      <td>Comedy|Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92440</th>\n",
       "      <td>2109</td>\n",
       "      <td>Jerk, The (1979)</td>\n",
       "      <td>Comedy</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movie_id                                              title  \\\n",
       "1320       2002                             Lethal Weapon 3 (1992)   \n",
       "74110      3566                             Big Kahuna, The (2000)   \n",
       "74540      6101                                     Missing (1982)   \n",
       "75420       595                        Beauty and the Beast (1991)   \n",
       "79720     97938                                  Life of Pi (2012)   \n",
       "89170      2146                             St. Elmo's Fire (1985)   \n",
       "90540       193                                   Showgirls (1995)   \n",
       "91720      8620  Exterminating Angel, The (Ángel exterminador, ...   \n",
       "91910      4979                       Royal Tenenbaums, The (2001)   \n",
       "92440      2109                                   Jerk, The (1979)   \n",
       "\n",
       "                                                genres  \n",
       "1320                         Action|Comedy|Crime|Drama  \n",
       "74110                                     Comedy|Drama  \n",
       "74540                           Drama|Mystery|Thriller  \n",
       "75420  Animation|Children|Fantasy|Musical|Romance|IMAX  \n",
       "79720                             Adventure|Drama|IMAX  \n",
       "89170                                    Drama|Romance  \n",
       "90540                                            Drama  \n",
       "91720                     Comedy|Drama|Fantasy|Mystery  \n",
       "91910                                     Comedy|Drama  \n",
       "92440                                           Comedy  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Provides a list of the movies recommended to the user\n",
    "movies_to_watch = movie_df[movie_df.index.isin(top_10_movie_recs)]\n",
    "movies_to_watch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movie_id</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>48709</th>\n",
       "      <td>154</td>\n",
       "      <td>Beauty of the Day (Belle de jour) (1967)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48710</th>\n",
       "      <td>599</td>\n",
       "      <td>Wild Bunch, The (1969)</td>\n",
       "      <td>Adventure|Western</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48711</th>\n",
       "      <td>902</td>\n",
       "      <td>Breakfast at Tiffany's (1961)</td>\n",
       "      <td>Drama|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48712</th>\n",
       "      <td>909</td>\n",
       "      <td>Apartment, The (1960)</td>\n",
       "      <td>Comedy|Drama|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48713</th>\n",
       "      <td>914</td>\n",
       "      <td>My Fair Lady (1964)</td>\n",
       "      <td>Comedy|Drama|Musical|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48714</th>\n",
       "      <td>924</td>\n",
       "      <td>2001: A Space Odyssey (1968)</td>\n",
       "      <td>Adventure|Drama|Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48715</th>\n",
       "      <td>1028</td>\n",
       "      <td>Mary Poppins (1964)</td>\n",
       "      <td>Children|Comedy|Fantasy|Musical</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48716</th>\n",
       "      <td>1035</td>\n",
       "      <td>Sound of Music, The (1965)</td>\n",
       "      <td>Musical|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48717</th>\n",
       "      <td>1084</td>\n",
       "      <td>Bonnie and Clyde (1967)</td>\n",
       "      <td>Crime|Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48718</th>\n",
       "      <td>1201</td>\n",
       "      <td>Good, the Bad and the Ugly, The (Buono, il bru...</td>\n",
       "      <td>Action|Adventure|Western</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movie_id                                              title  \\\n",
       "48709       154           Beauty of the Day (Belle de jour) (1967)   \n",
       "48710       599                             Wild Bunch, The (1969)   \n",
       "48711       902                      Breakfast at Tiffany's (1961)   \n",
       "48712       909                              Apartment, The (1960)   \n",
       "48713       914                                My Fair Lady (1964)   \n",
       "48714       924                       2001: A Space Odyssey (1968)   \n",
       "48715      1028                                Mary Poppins (1964)   \n",
       "48716      1035                         Sound of Music, The (1965)   \n",
       "48717      1084                            Bonnie and Clyde (1967)   \n",
       "48718      1201  Good, the Bad and the Ugly, The (Buono, il bru...   \n",
       "\n",
       "                                genres  \n",
       "48709                            Drama  \n",
       "48710                Adventure|Western  \n",
       "48711                    Drama|Romance  \n",
       "48712             Comedy|Drama|Romance  \n",
       "48713     Comedy|Drama|Musical|Romance  \n",
       "48714           Adventure|Drama|Sci-Fi  \n",
       "48715  Children|Comedy|Fantasy|Musical  \n",
       "48716                  Musical|Romance  \n",
       "48717                      Crime|Drama  \n",
       "48718         Action|Adventure|Western  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#View users top 10 movie reviews for comparison\n",
    "user_top_movies = joined_movies_df[[\"movie_id\",\"title\",\"genres\",\"user_id\",\"rating\"]]\n",
    "user_top_movies = user_top_movies[user_top_movies.user_id == userId]\n",
    "user_top_movies.sort_values(\"rating\", ascending=False)\n",
    "user_top_movies = user_top_movies[[\"movie_id\",\"title\",\"genres\"]]\n",
    "user_top_movies.head(10)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While the deep learning had worked, the results were very inaccurate and suffered a lot of lost data. Meaning the results from the learning model are probably not very useful. The next part will be to optimise the deep learning model to see if the accuracy increases."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Optimisation 1\n",
    "Adding more epochs and batch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Building the deep learning model\n",
    "model_opt1 = Sequential()\n",
    "model_opt1.add(Dense(20, activation='relu'))\n",
    "model_opt1.add(Dense(10, activation='relu'))\n",
    "model_opt1.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "9076/9076 [==============================] - 15s 2ms/step - loss: 7.7808 - accuracy: 1.2121e-04\n",
      "Epoch 2/10\n",
      "9076/9076 [==============================] - 14s 2ms/step - loss: 7.6846 - accuracy: 7.7133e-05\n",
      "Epoch 3/10\n",
      "9076/9076 [==============================] - 16s 2ms/step - loss: 7.6887 - accuracy: 0.0000e+00\n",
      "Epoch 4/10\n",
      "9076/9076 [==============================] - 24s 3ms/step - loss: 7.6836 - accuracy: 0.0000e+00\n",
      "Epoch 5/10\n",
      "9076/9076 [==============================] - 31s 3ms/step - loss: 7.6840 - accuracy: 0.0000e+00\n",
      "Epoch 6/10\n",
      "9076/9076 [==============================] - 21s 2ms/step - loss: 7.7085 - accuracy: 0.0000e+00\n",
      "Epoch 7/10\n",
      "9076/9076 [==============================] - 20s 2ms/step - loss: 7.3806 - accuracy: 1.2121e-04\n",
      "Epoch 8/10\n",
      "9076/9076 [==============================] - 26s 3ms/step - loss: 7.1461 - accuracy: 0.0000e+00\n",
      "Epoch 9/10\n",
      "9076/9076 [==============================] - 26s 3ms/step - loss: 7.1956 - accuracy: 0.0068\n",
      "Epoch 10/10\n",
      "9076/9076 [==============================] - 26s 3ms/step - loss: 7.2169 - accuracy: 0.0194\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1c9b184aac0>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting the training data to the model\n",
    "model_opt1.fit(x_train, y_train, epochs=10, batch_size=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "316/316 [==============================] - 1s 2ms/step - loss: 7.5686 - accuracy: 0.0125\n",
      "[7.568582534790039, 0.012495041824877262]\n"
     ]
    }
   ],
   "source": [
    "#Evaluate the keras model, [Loss, Accuracy]\n",
    "accuracy = model_opt1.evaluate(x_test, y_test, verbose=1)\n",
    "print(accuracy)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Adding more batch layers and increasing the epochs had increased the accuracy to 0.1357 from the initial 0.0331 accuracy. The model also had less losses going from 5.5 to 4.9, but the model still loses a lot of data meaning the results might not be very useful."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Optimise 2\n",
    "Adding more layers and increasing the density"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Building the deep learning model\n",
    "model_opt2 = Sequential()\n",
    "model_opt2.add(Dense(80, activation='relu'))\n",
    "model_opt2.add(Dense(40, activation='relu'))\n",
    "model_opt2.add(Dense(20, activation='relu'))\n",
    "model_opt2.add(Dense(10, activation='relu'))\n",
    "model_opt2.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "9076/9076 [==============================] - 25s 3ms/step - loss: 6.2176 - accuracy: 0.1005\n",
      "Epoch 2/10\n",
      "9076/9076 [==============================] - 24s 3ms/step - loss: 6.0713 - accuracy: 0.0610\n",
      "Epoch 3/10\n",
      "9076/9076 [==============================] - 25s 3ms/step - loss: 6.2437 - accuracy: 0.0912\n",
      "Epoch 4/10\n",
      "9076/9076 [==============================] - 25s 3ms/step - loss: 6.6088 - accuracy: 0.1318\n",
      "Epoch 5/10\n",
      "9076/9076 [==============================] - 24s 3ms/step - loss: 6.6088 - accuracy: 0.1357\n",
      "Epoch 6/10\n",
      "9076/9076 [==============================] - 24s 3ms/step - loss: 6.6089 - accuracy: 0.1269\n",
      "Epoch 7/10\n",
      "9076/9076 [==============================] - 24s 3ms/step - loss: 6.5317 - accuracy: 0.0431\n",
      "Epoch 8/10\n",
      "9076/9076 [==============================] - 27s 3ms/step - loss: 6.6087 - accuracy: 0.0856\n",
      "Epoch 9/10\n",
      "9076/9076 [==============================] - 27s 3ms/step - loss: 6.6088 - accuracy: 0.1130\n",
      "Epoch 10/10\n",
      "9076/9076 [==============================] - 31s 3ms/step - loss: 6.6087 - accuracy: 0.1357\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1c9b1ed0a90>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting the training data to the model\n",
    "model_opt2.fit(x_train, y_train, epochs=10, batch_size=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "316/316 [==============================] - 1s 2ms/step - loss: 6.9314 - accuracy: 0.0891\n",
      "[6.931399822235107, 0.08905196189880371]\n"
     ]
    }
   ],
   "source": [
    "#Evaluate the keras model, [Loss, Accuracy]\n",
    "accuracy = model_opt2.evaluate(x_test, y_test, verbose=1)\n",
    "print(accuracy)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Seemingly this was a massive step back, the increase in density for each layer had increased the lost data to almost 7 and the accuracy is 0 which was very disappointing, potentially adding another layer that with another activation would help."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Optimise 3\n",
    "adding a sigmoid layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Building the deep learning model\n",
    "model_opt3 = Sequential()\n",
    "model_opt3.add(Dense(80, activation='relu'))\n",
    "model_opt3.add(Dense(40, activation='relu'))\n",
    "model_opt3.add(Dense(20, activation='relu'))\n",
    "model_opt3.add(Dense(10, activation='relu'))\n",
    "model_opt3.add(Dense(1, activation='sigmoid'))\n",
    "model_opt3.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "9076/9076 [==============================] - 36s 4ms/step - loss: 0.7482 - accuracy: 0.1297\n",
      "Epoch 2/10\n",
      "9076/9076 [==============================] - 31s 3ms/step - loss: 0.6327 - accuracy: 0.1357\n",
      "Epoch 3/10\n",
      "9076/9076 [==============================] - 38s 4ms/step - loss: 0.6328 - accuracy: 0.1357\n",
      "Epoch 4/10\n",
      "9076/9076 [==============================] - 29s 3ms/step - loss: 0.6326 - accuracy: 0.1357\n",
      "Epoch 5/10\n",
      "9076/9076 [==============================] - 29s 3ms/step - loss: 0.6326 - accuracy: 0.1357\n",
      "Epoch 6/10\n",
      "9076/9076 [==============================] - 38s 4ms/step - loss: 0.6326 - accuracy: 0.1357\n",
      "Epoch 7/10\n",
      "9076/9076 [==============================] - 35s 4ms/step - loss: 0.6326 - accuracy: 0.1357\n",
      "Epoch 8/10\n",
      "9076/9076 [==============================] - 28s 3ms/step - loss: 0.6326 - accuracy: 0.1357\n",
      "Epoch 9/10\n",
      "9076/9076 [==============================] - 32s 4ms/step - loss: 0.6326 - accuracy: 0.1357\n",
      "Epoch 10/10\n",
      "9076/9076 [==============================] - 30s 3ms/step - loss: 0.6326 - accuracy: 0.1357\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1c9b17ea130>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting the training data to the model\n",
    "model_opt3.fit(x_train, y_train, epochs=10, batch_size=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "316/316 [==============================] - 1s 3ms/step - loss: 0.6715 - accuracy: 0.0891\n",
      "[0.6715357303619385, 0.08905196189880371]\n"
     ]
    }
   ],
   "source": [
    "#Evaluate the keras model, [Loss, Accuracy]\n",
    "accuracy = model_opt3.evaluate(x_test, y_test, verbose=1)\n",
    "print(accuracy)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The sigmoid layer was a great improvement since it means the model has a more accurate final layer and better produces results. The accuracy is now 0.1357 and the loss is only 0.6310. While these are not particularly high results by themselves, but in comparison from previous results this is a good improvement."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Optimisation 4\n",
    "adding more relu layers as it seem more effective and lowering the density."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Building the deep learning model\n",
    "model_opt4 = Sequential()\n",
    "model_opt4.add(Dense(55, activation='relu'))\n",
    "model_opt4.add(Dense(50, activation='relu'))\n",
    "model_opt4.add(Dense(45, activation='relu'))\n",
    "model_opt4.add(Dense(40, activation='relu'))\n",
    "model_opt4.add(Dense(20, activation='relu'))\n",
    "model_opt4.add(Dense(10, activation='relu'))\n",
    "model_opt4.add(Dense(1, activation='sigmoid'))\n",
    "model_opt4.compile(loss='binary_crossentropy', optimizer=\"adam\", metrics=['accuracy'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "9076/9076 [==============================] - 28s 3ms/step - loss: 0.6858 - accuracy: 0.1343\n",
      "Epoch 2/10\n",
      "9076/9076 [==============================] - 25s 3ms/step - loss: 0.6326 - accuracy: 0.1357\n",
      "Epoch 3/10\n",
      "9076/9076 [==============================] - 27s 3ms/step - loss: 0.6326 - accuracy: 0.1357\n",
      "Epoch 4/10\n",
      "9076/9076 [==============================] - 26s 3ms/step - loss: 0.6323 - accuracy: 0.1357\n",
      "Epoch 5/10\n",
      "9076/9076 [==============================] - 26s 3ms/step - loss: 0.6310 - accuracy: 0.1357\n",
      "Epoch 6/10\n",
      "9076/9076 [==============================] - 26s 3ms/step - loss: 0.6305 - accuracy: 0.1357\n",
      "Epoch 7/10\n",
      "9076/9076 [==============================] - 26s 3ms/step - loss: 0.6300 - accuracy: 0.1357\n",
      "Epoch 8/10\n",
      "9076/9076 [==============================] - 28s 3ms/step - loss: 0.6292 - accuracy: 0.1356\n",
      "Epoch 9/10\n",
      "9076/9076 [==============================] - 30s 3ms/step - loss: 0.6290 - accuracy: 0.1357\n",
      "Epoch 10/10\n",
      "9076/9076 [==============================] - 33s 4ms/step - loss: 0.6290 - accuracy: 0.1356\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1c9b2f7c130>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting the training data to the model\n",
    "model_opt4.fit(x_train, y_train, epochs=10, batch_size=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.6636435389518738, 0.08905196189880371]\n"
     ]
    }
   ],
   "source": [
    "#Evaluate the keras model, [Loss, Accuracy]\n",
    "accuracy = model_opt4.evaluate(x_test, y_test, verbose=0)\n",
    "print(accuracy)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While the accuracy stayed at 0.1357 there was a small amount of data that was better retained but overall the final optimisation has had minimal effect on the results."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Post optimisation results\n",
    "After optimising the data the recommendation results are as follows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Using some of the columns from the imported data\n",
    "movie_df = joined_movies_df[[\"movie_id\",\"title\",\"genres\"]]\n",
    "#Select the user to recommend\n",
    "userId = 315\n",
    "#Find movies the user has seen and movies the user has not seen\n",
    "movies_watched_by_user = user_df[user_df.user_id == userId]\n",
    "#Make movies not watched a pandas series so it can be read into the learning model\n",
    "movies_not_watched = movie_df[~movie_df[\"movie_id\"].isin(movies_watched_by_user.movie_id.values)][\"movie_id\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "303/303 [==============================] - 1s 2ms/step\n"
     ]
    }
   ],
   "source": [
    "#Takes the movies not watched and finds their enumerated keys\n",
    "movies_not_watched = list(\n",
    "    set(movies_not_watched).intersection(set(movie_list_enumerated.keys()))\n",
    ")\n",
    "movies_not_watched = [[movie_list_enumerated.get(x)] for x in movies_not_watched]\n",
    "user_encoder = User_list_enumerated.get(userId)\n",
    "#Formats the users movies they haven't watched for model.predict\n",
    "user_movie_array = np.hstack(\n",
    "    ([[user_encoder]] * len(movies_not_watched), movies_not_watched)\n",
    ")\n",
    "#Predicts the best suited movies for the selected user\n",
    "movie_recs = model_opt4.predict(user_movie_array).flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1035, 1030, 1029, 1026, 1036,  940,  931, 1024, 1098,  941],\n",
       "      dtype=int64)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Sorts the movies into a top 10\n",
    "top_10_movie_recs = movie_recs.argsort()[-10:][::-1]\n",
    "top_10_movie_recs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movie_id</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>931</th>\n",
       "      <td>3623</td>\n",
       "      <td>Mission: Impossible II (2000)</td>\n",
       "      <td>Action|Adventure|Thriller</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>940</th>\n",
       "      <td>4306</td>\n",
       "      <td>Shrek (2001)</td>\n",
       "      <td>Adventure|Animation|Children|Comedy|Fantasy|Ro...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>941</th>\n",
       "      <td>4310</td>\n",
       "      <td>Pearl Harbor (2001)</td>\n",
       "      <td>Action|Drama|Romance|War</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1024</th>\n",
       "      <td>49286</td>\n",
       "      <td>Holiday, The (2006)</td>\n",
       "      <td>Comedy|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1026</th>\n",
       "      <td>2</td>\n",
       "      <td>Jumanji (1995)</td>\n",
       "      <td>Adventure|Children|Fantasy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1029</th>\n",
       "      <td>21</td>\n",
       "      <td>Get Shorty (1995)</td>\n",
       "      <td>Comedy|Crime|Thriller</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1030</th>\n",
       "      <td>32</td>\n",
       "      <td>Twelve Monkeys (a.k.a. 12 Monkeys) (1995)</td>\n",
       "      <td>Mystery|Sci-Fi|Thriller</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1035</th>\n",
       "      <td>110</td>\n",
       "      <td>Braveheart (1995)</td>\n",
       "      <td>Action|Drama|War</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1036</th>\n",
       "      <td>141</td>\n",
       "      <td>Birdcage, The (1996)</td>\n",
       "      <td>Comedy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1098</th>\n",
       "      <td>5218</td>\n",
       "      <td>Ice Age (2002)</td>\n",
       "      <td>Adventure|Animation|Children|Comedy</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      movie_id                                      title  \\\n",
       "931       3623              Mission: Impossible II (2000)   \n",
       "940       4306                               Shrek (2001)   \n",
       "941       4310                        Pearl Harbor (2001)   \n",
       "1024     49286                        Holiday, The (2006)   \n",
       "1026         2                             Jumanji (1995)   \n",
       "1029        21                          Get Shorty (1995)   \n",
       "1030        32  Twelve Monkeys (a.k.a. 12 Monkeys) (1995)   \n",
       "1035       110                          Braveheart (1995)   \n",
       "1036       141                       Birdcage, The (1996)   \n",
       "1098      5218                             Ice Age (2002)   \n",
       "\n",
       "                                                 genres  \n",
       "931                           Action|Adventure|Thriller  \n",
       "940   Adventure|Animation|Children|Comedy|Fantasy|Ro...  \n",
       "941                            Action|Drama|Romance|War  \n",
       "1024                                     Comedy|Romance  \n",
       "1026                         Adventure|Children|Fantasy  \n",
       "1029                              Comedy|Crime|Thriller  \n",
       "1030                            Mystery|Sci-Fi|Thriller  \n",
       "1035                                   Action|Drama|War  \n",
       "1036                                             Comedy  \n",
       "1098                Adventure|Animation|Children|Comedy  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Provides a list of the movies recommended to the user\n",
    "movies_to_watch = movie_df[movie_df.index.isin(top_10_movie_recs)]\n",
    "movies_to_watch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movie_id</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>48709</th>\n",
       "      <td>154</td>\n",
       "      <td>Beauty of the Day (Belle de jour) (1967)</td>\n",
       "      <td>Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48710</th>\n",
       "      <td>599</td>\n",
       "      <td>Wild Bunch, The (1969)</td>\n",
       "      <td>Adventure|Western</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48711</th>\n",
       "      <td>902</td>\n",
       "      <td>Breakfast at Tiffany's (1961)</td>\n",
       "      <td>Drama|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48712</th>\n",
       "      <td>909</td>\n",
       "      <td>Apartment, The (1960)</td>\n",
       "      <td>Comedy|Drama|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48713</th>\n",
       "      <td>914</td>\n",
       "      <td>My Fair Lady (1964)</td>\n",
       "      <td>Comedy|Drama|Musical|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48714</th>\n",
       "      <td>924</td>\n",
       "      <td>2001: A Space Odyssey (1968)</td>\n",
       "      <td>Adventure|Drama|Sci-Fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48715</th>\n",
       "      <td>1028</td>\n",
       "      <td>Mary Poppins (1964)</td>\n",
       "      <td>Children|Comedy|Fantasy|Musical</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48716</th>\n",
       "      <td>1035</td>\n",
       "      <td>Sound of Music, The (1965)</td>\n",
       "      <td>Musical|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48717</th>\n",
       "      <td>1084</td>\n",
       "      <td>Bonnie and Clyde (1967)</td>\n",
       "      <td>Crime|Drama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48718</th>\n",
       "      <td>1201</td>\n",
       "      <td>Good, the Bad and the Ugly, The (Buono, il bru...</td>\n",
       "      <td>Action|Adventure|Western</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movie_id                                              title  \\\n",
       "48709       154           Beauty of the Day (Belle de jour) (1967)   \n",
       "48710       599                             Wild Bunch, The (1969)   \n",
       "48711       902                      Breakfast at Tiffany's (1961)   \n",
       "48712       909                              Apartment, The (1960)   \n",
       "48713       914                                My Fair Lady (1964)   \n",
       "48714       924                       2001: A Space Odyssey (1968)   \n",
       "48715      1028                                Mary Poppins (1964)   \n",
       "48716      1035                         Sound of Music, The (1965)   \n",
       "48717      1084                            Bonnie and Clyde (1967)   \n",
       "48718      1201  Good, the Bad and the Ugly, The (Buono, il bru...   \n",
       "\n",
       "                                genres  \n",
       "48709                            Drama  \n",
       "48710                Adventure|Western  \n",
       "48711                    Drama|Romance  \n",
       "48712             Comedy|Drama|Romance  \n",
       "48713     Comedy|Drama|Musical|Romance  \n",
       "48714           Adventure|Drama|Sci-Fi  \n",
       "48715  Children|Comedy|Fantasy|Musical  \n",
       "48716                  Musical|Romance  \n",
       "48717                      Crime|Drama  \n",
       "48718         Action|Adventure|Western  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#View users top 10 movie reviews for comparison\n",
    "user_top_movies = joined_movies_df[[\"movie_id\",\"title\",\"genres\",\"user_id\",\"rating\"]]\n",
    "user_top_movies = user_top_movies[user_top_movies.user_id == userId]\n",
    "user_top_movies.sort_values(\"rating\", ascending=False)\n",
    "user_top_movies = user_top_movies[[\"movie_id\",\"title\",\"genres\"]]\n",
    "user_top_movies.head(10)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Final statistical analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1 Score = 0.6636435389518738, Precision = 0.08905196189880371\n"
     ]
    }
   ],
   "source": [
    "f1_score, precision = model_opt4.evaluate(x_test, y_test, verbose=0)\n",
    "print(f\"F1 Score = {f1_score}, Precision = {precision}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall = [0.6636435389518738, 0.08905196189880371]\n"
     ]
    }
   ],
   "source": [
    "recall = model_opt4.evaluate(x_test, y_test, verbose=0)\n",
    "print(f\"Recall = {recall}\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "b98826369006c2dd1ccf8d106235eaa643bd89f632ee848b92ea26653baccf0c"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}