Neural_Recommendation_System.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Neural Recommendation System.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "mount_file_id": "1V72tTGKWhlnCySrWVPFA9Gihm1piSRqn",
      "authorship_tag": "ABX9TyOZHPg75Oc7ZvB2xuohYVKW",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Tejas-TA/Neural-Network-Movie-Recommendation-/blob/main/Neural_Recommendation_System.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DEga11HkpqJe"
      },
      "source": [
        "# **Neural Collaborative filter based Movie Recommendation System**\n",
        "---\n",
        "This is a movie recommendation project. The dataset is from movielens dataset extract which contains ~100K records. The recommendation of a movie is based on the **Neural Collaborative** filtering technique."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ECKKGeE_dXBc"
      },
      "source": [
        "## Data Preprocessing\n",
        "\n",
        "Including Data Analysis, Visualization and Manipulation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Tsi2ClbCsnbF"
      },
      "source": [
        "#Importing necessary libraries at once\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "from keras.models import Model\n",
        "from keras.layers import Input, Embedding, Flatten, Dense, Concatenate\n",
        "import tensorflow as tf"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 514
        },
        "id": "9g8JuY6ct2B0",
        "outputId": "dd971ac7-3ef1-4703-f393-69e801ceb9b8"
      },
      "source": [
        "#Importing dataset and assigning names to features\n",
        "titles = ['movie_ratings','movie_ids','user_ids','Timestamp']\n",
        "df = pd.read_csv('/content/drive/MyDrive/Recommendation System/data.csv', names = titles)\n",
        "df.head(15)"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "execute_result",
          "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_ratings</th>\n",
              "      <th>movie_ids</th>\n",
              "      <th>user_ids</th>\n",
              "      <th>Timestamp</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>3</td>\n",
              "      <td>242</td>\n",
              "      <td>196</td>\n",
              "      <td>881250949</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3</td>\n",
              "      <td>302</td>\n",
              "      <td>186</td>\n",
              "      <td>891717742</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>377</td>\n",
              "      <td>22</td>\n",
              "      <td>878887116</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2</td>\n",
              "      <td>51</td>\n",
              "      <td>244</td>\n",
              "      <td>880606923</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>346</td>\n",
              "      <td>166</td>\n",
              "      <td>886397596</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>4</td>\n",
              "      <td>474</td>\n",
              "      <td>298</td>\n",
              "      <td>884182806</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>2</td>\n",
              "      <td>265</td>\n",
              "      <td>115</td>\n",
              "      <td>881171488</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>5</td>\n",
              "      <td>465</td>\n",
              "      <td>253</td>\n",
              "      <td>891628467</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>3</td>\n",
              "      <td>451</td>\n",
              "      <td>305</td>\n",
              "      <td>886324817</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>3</td>\n",
              "      <td>86</td>\n",
              "      <td>6</td>\n",
              "      <td>883603013</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>2</td>\n",
              "      <td>257</td>\n",
              "      <td>62</td>\n",
              "      <td>879372434</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>5</td>\n",
              "      <td>1014</td>\n",
              "      <td>286</td>\n",
              "      <td>879781125</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>5</td>\n",
              "      <td>222</td>\n",
              "      <td>200</td>\n",
              "      <td>876042340</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>3</td>\n",
              "      <td>40</td>\n",
              "      <td>210</td>\n",
              "      <td>891035994</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>3</td>\n",
              "      <td>29</td>\n",
              "      <td>224</td>\n",
              "      <td>888104457</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    movie_ratings  movie_ids  user_ids  Timestamp\n",
              "0               3        242       196  881250949\n",
              "1               3        302       186  891717742\n",
              "2               1        377        22  878887116\n",
              "3               2         51       244  880606923\n",
              "4               1        346       166  886397596\n",
              "5               4        474       298  884182806\n",
              "6               2        265       115  881171488\n",
              "7               5        465       253  891628467\n",
              "8               3        451       305  886324817\n",
              "9               3         86         6  883603013\n",
              "10              2        257        62  879372434\n",
              "11              5       1014       286  879781125\n",
              "12              5        222       200  876042340\n",
              "13              3         40       210  891035994\n",
              "14              3         29       224  888104457"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 630
        },
        "id": "REcVhW65Y8_Z",
        "outputId": "6989b1b1-847a-41f5-a0ce-202591264a99"
      },
      "source": [
        "#Plotting ratings chart\n",
        "plt.figure(figsize=(10,10)) \n",
        "plt.rc(\"font\", size=11)\n",
        "df.movie_ratings.value_counts(sort=False).plot(kind='bar', color ='maroon')\n",
        "plt.title('Movie Rating Distribution')\n",
        "plt.xlabel('\\nRating')\n",
        "plt.ylabel('No. of movies')\n",
        "plt.show()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6fUijopA0CyS",
        "outputId": "08f0549d-5a3d-4a1e-bbf0-1405cc5d8cac"
      },
      "source": [
        "#Extracting column names\n",
        "column_names = list(df.columns.values)\n",
        "column_names"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['movie_ratings', 'movie_ids', 'user_ids', 'Timestamp']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 514
        },
        "id": "uRmD0zUb0KbV",
        "outputId": "61850ddb-71b7-4a9c-e85e-b3cab2d6c5eb"
      },
      "source": [
        "#Re-ordering columns\n",
        "df = df[['user_ids', 'movie_ids', 'movie_ratings','Timestamp']]\n",
        "df.head(15)"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "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_ids</th>\n",
              "      <th>movie_ids</th>\n",
              "      <th>movie_ratings</th>\n",
              "      <th>Timestamp</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>196</td>\n",
              "      <td>242</td>\n",
              "      <td>3</td>\n",
              "      <td>881250949</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>186</td>\n",
              "      <td>302</td>\n",
              "      <td>3</td>\n",
              "      <td>891717742</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>22</td>\n",
              "      <td>377</td>\n",
              "      <td>1</td>\n",
              "      <td>878887116</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>244</td>\n",
              "      <td>51</td>\n",
              "      <td>2</td>\n",
              "      <td>880606923</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>166</td>\n",
              "      <td>346</td>\n",
              "      <td>1</td>\n",
              "      <td>886397596</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>298</td>\n",
              "      <td>474</td>\n",
              "      <td>4</td>\n",
              "      <td>884182806</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>115</td>\n",
              "      <td>265</td>\n",
              "      <td>2</td>\n",
              "      <td>881171488</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>253</td>\n",
              "      <td>465</td>\n",
              "      <td>5</td>\n",
              "      <td>891628467</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>305</td>\n",
              "      <td>451</td>\n",
              "      <td>3</td>\n",
              "      <td>886324817</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>6</td>\n",
              "      <td>86</td>\n",
              "      <td>3</td>\n",
              "      <td>883603013</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>62</td>\n",
              "      <td>257</td>\n",
              "      <td>2</td>\n",
              "      <td>879372434</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>286</td>\n",
              "      <td>1014</td>\n",
              "      <td>5</td>\n",
              "      <td>879781125</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>200</td>\n",
              "      <td>222</td>\n",
              "      <td>5</td>\n",
              "      <td>876042340</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>210</td>\n",
              "      <td>40</td>\n",
              "      <td>3</td>\n",
              "      <td>891035994</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>224</td>\n",
              "      <td>29</td>\n",
              "      <td>3</td>\n",
              "      <td>888104457</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    user_ids  movie_ids  movie_ratings  Timestamp\n",
              "0        196        242              3  881250949\n",
              "1        186        302              3  891717742\n",
              "2         22        377              1  878887116\n",
              "3        244         51              2  880606923\n",
              "4        166        346              1  886397596\n",
              "5        298        474              4  884182806\n",
              "6        115        265              2  881171488\n",
              "7        253        465              5  891628467\n",
              "8        305        451              3  886324817\n",
              "9          6         86              3  883603013\n",
              "10        62        257              2  879372434\n",
              "11       286       1014              5  879781125\n",
              "12       200        222              5  876042340\n",
              "13       210         40              3  891035994\n",
              "14       224         29              3  888104457"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 359
        },
        "id": "R-Sn0ilPeo3y",
        "outputId": "2d825acd-ceea-4664-925d-fe0bb48c3a99"
      },
      "source": [
        "#Dataset has ~100k records\n",
        "df.tail(10)"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "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_ids</th>\n",
              "      <th>movie_ids</th>\n",
              "      <th>movie_ratings</th>\n",
              "      <th>Timestamp</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>99990</th>\n",
              "      <td>806</td>\n",
              "      <td>421</td>\n",
              "      <td>4</td>\n",
              "      <td>882388897</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99991</th>\n",
              "      <td>676</td>\n",
              "      <td>538</td>\n",
              "      <td>4</td>\n",
              "      <td>892685437</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99992</th>\n",
              "      <td>721</td>\n",
              "      <td>262</td>\n",
              "      <td>3</td>\n",
              "      <td>877137285</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99993</th>\n",
              "      <td>913</td>\n",
              "      <td>209</td>\n",
              "      <td>2</td>\n",
              "      <td>881367150</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99994</th>\n",
              "      <td>378</td>\n",
              "      <td>78</td>\n",
              "      <td>3</td>\n",
              "      <td>880056976</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99995</th>\n",
              "      <td>880</td>\n",
              "      <td>476</td>\n",
              "      <td>3</td>\n",
              "      <td>880175444</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99996</th>\n",
              "      <td>716</td>\n",
              "      <td>204</td>\n",
              "      <td>5</td>\n",
              "      <td>879795543</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99997</th>\n",
              "      <td>276</td>\n",
              "      <td>1090</td>\n",
              "      <td>1</td>\n",
              "      <td>874795795</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99998</th>\n",
              "      <td>13</td>\n",
              "      <td>225</td>\n",
              "      <td>2</td>\n",
              "      <td>882399156</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99999</th>\n",
              "      <td>12</td>\n",
              "      <td>203</td>\n",
              "      <td>3</td>\n",
              "      <td>879959583</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       user_ids  movie_ids  movie_ratings  Timestamp\n",
              "99990       806        421              4  882388897\n",
              "99991       676        538              4  892685437\n",
              "99992       721        262              3  877137285\n",
              "99993       913        209              2  881367150\n",
              "99994       378         78              3  880056976\n",
              "99995       880        476              3  880175444\n",
              "99996       716        204              5  879795543\n",
              "99997       276       1090              1  874795795\n",
              "99998        13        225              2  882399156\n",
              "99999        12        203              3  879959583"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "z5r5qjsNdyZP"
      },
      "source": [
        "## Neural Network Building"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wFm-IRn11dVu"
      },
      "source": [
        "#Splitting the dataset into train and test \n",
        "train, test = train_test_split(df, test_size=0.25, random_state=42)"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rWdn9zq-9w_Y",
        "outputId": "b0e0e934-ee78-4644-97ec-4415552afef2"
      },
      "source": [
        "n_users = len(df.user_ids.unique())\n",
        "n_users"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "943"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xhr7hCSo99ga",
        "outputId": "4503eebe-a0f9-4fb3-fdfa-ac69895dc0cc"
      },
      "source": [
        "n_items = len(df.movie_ids.unique())\n",
        "n_items"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1682"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sIeWdc4T1l_G"
      },
      "source": [
        "#Creating movie embedding path\n",
        "movie_input = Input(shape = [1], name = \"Movie-Input\")\n",
        "movie_embedding = Embedding(n_items + 1, 5, name= \"Movie-Embedding\")(movie_input)\n",
        "movie_vec = Flatten(name = \"Flatten-Movies\")(movie_embedding)\n",
        "\n",
        "#Creating user embedding path\n",
        "user_input = Input(shape = [1], name = \"User-Input\")\n",
        "user_embedding = Embedding(n_users + 1, 5, name = \"User-Embedding\")(user_input)\n",
        "user_vec = Flatten(name = \"Flatten-Users\")(user_embedding)\n",
        "\n",
        "#Concatenate features\n",
        "conc = Concatenate()([movie_vec, user_vec])\n",
        "\n",
        "#Weight Initialization\n",
        "initializer = tf.keras.initializers.RandomUniform(minval=0., maxval=1., seed = None)\n",
        "\n",
        "#Add fully-connected-layers\n",
        "fc1 = Dense(128, activation='relu')(conc)\n",
        "fc2 = Dense(100, activation='relu')(fc1)\n",
        "fc3 = Dense(80, activation='relu')(fc2)\n",
        "fc4 = Dense(60, activation='relu')(fc3)\n",
        "fc5 = Dense(32, activation='relu')(fc4)\n",
        "out = Dense(1)(fc5)\n",
        "\n",
        "#Create model and compile it\n",
        "model = Model([user_input, movie_input], out)\n",
        "model.compile('adam', 'mean_squared_error')"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S3IEwT96gs79",
        "outputId": "96ab6f14-875a-497a-8847-e565982569b6"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"model\"\n",
            "__________________________________________________________________________________________________\n",
            "Layer (type)                    Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            "Movie-Input (InputLayer)        [(None, 1)]          0                                            \n",
            "__________________________________________________________________________________________________\n",
            "User-Input (InputLayer)         [(None, 1)]          0                                            \n",
            "__________________________________________________________________________________________________\n",
            "Movie-Embedding (Embedding)     (None, 1, 5)         8415        Movie-Input[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "User-Embedding (Embedding)      (None, 1, 5)         4720        User-Input[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "Flatten-Movies (Flatten)        (None, 5)            0           Movie-Embedding[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "Flatten-Users (Flatten)         (None, 5)            0           User-Embedding[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "concatenate (Concatenate)       (None, 10)           0           Flatten-Movies[0][0]             \n",
            "                                                                 Flatten-Users[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "dense (Dense)                   (None, 128)          1408        concatenate[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "dense_1 (Dense)                 (None, 100)          12900       dense[0][0]                      \n",
            "__________________________________________________________________________________________________\n",
            "dense_2 (Dense)                 (None, 80)           8080        dense_1[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "dense_3 (Dense)                 (None, 60)           4860        dense_2[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "dense_4 (Dense)                 (None, 32)           1952        dense_3[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "dense_5 (Dense)                 (None, 1)            33          dense_4[0][0]                    \n",
            "==================================================================================================\n",
            "Total params: 42,368\n",
            "Trainable params: 42,368\n",
            "Non-trainable params: 0\n",
            "__________________________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Owe3eqE62vPS",
        "outputId": "0ab976f5-1735-4898-97c8-81b17009f907"
      },
      "source": [
        "history = model.fit([train.user_ids, train.movie_ids], train.movie_ratings, epochs = 18, verbose = 1)"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/18\n",
            "2344/2344 [==============================] - 20s 3ms/step - loss: 1.6062\n",
            "Epoch 2/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.8927\n",
            "Epoch 3/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.8538\n",
            "Epoch 4/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.8277\n",
            "Epoch 5/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.7902\n",
            "Epoch 6/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.7529\n",
            "Epoch 7/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.7246\n",
            "Epoch 8/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.7077\n",
            "Epoch 9/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.6888\n",
            "Epoch 10/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.6831\n",
            "Epoch 11/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.6623\n",
            "Epoch 12/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.6589\n",
            "Epoch 13/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.6529\n",
            "Epoch 14/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.6499\n",
            "Epoch 15/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.6433\n",
            "Epoch 16/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.6364\n",
            "Epoch 17/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.6317\n",
            "Epoch 18/18\n",
            "2344/2344 [==============================] - 7s 3ms/step - loss: 0.6149\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fmaVvAgPLwtA",
        "outputId": "40df285f-7efc-486f-c449-36873316d604"
      },
      "source": [
        "model.evaluate([test.user_ids, test.movie_ids], test.movie_ratings)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "782/782 [==============================] - 2s 2ms/step - loss: 0.9274\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9273605346679688"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SDwKFc--d61h"
      },
      "source": [
        "## Movie Rating Prediction"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ym9tCC0TCD43",
        "outputId": "0aa477ee-523c-45ca-f338-03ed8e5923c5"
      },
      "source": [
        "#Predicting ratings\n",
        "predictions = model.predict([test.user_ids.head(10), test.movie_ids.head(10)])\n",
        "[print(predictions[i], test.movie_ratings.iloc[i]) for i in range(0,10)]"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[3.7645354] 4\n",
            "[3.7596421] 3\n",
            "[3.5160546] 4\n",
            "[2.0176213] 2\n",
            "[4.0441256] 2\n",
            "[3.4688585] 3\n",
            "[4.1284647] 5\n",
            "[4.216727] 4\n",
            "[3.4606106] 3\n",
            "[2.9891217] 4\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[None, None, None, None, None, None, None, None, None, None]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 388
        },
        "id": "k31pCiQiKe4X",
        "outputId": "e428f362-08eb-41f9-e13b-2f1cdd98b49f"
      },
      "source": [
        "#Plotting loss function against epochs, as epochs increased error decreases\n",
        "plt.figure(figsize=(10,6)) \n",
        "plt.plot(history.history['loss'])\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Training Error\")\n",
        "plt.legend([\"Epoch vs Training Error\"])\n",
        "plt.show()"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Baa0t5H4pMap",
        "outputId": "edb35e3b-07db-46c9-934d-fa9529bf206d"
      },
      "source": [
        "#Show the best validation RMSE\n",
        "import math\n",
        "min_val_loss, idx = min((val, idx) for (idx, val) in enumerate(history.history['loss']))\n",
        "print ('Minimum RMSE at epoch', '{:d}'.format(idx+1), '=', '{:.4f}'.format(math.sqrt(min_val_loss)))"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Minimum RMSE at epoch 18 = 0.8002\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "37Y5zXqxLpy7"
      },
      "source": [
        "#Run this to save the model. \n",
        "#Note: Consumes lot of drive space\n",
        "#model2.save('Recommendation_Model_Neural_Network.h5')"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "X8BvnIbFfv0_"
      },
      "source": [
        "## Embeddings Visualization"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fL3-DbO-MO5Z"
      },
      "source": [
        "item_em = model.get_layer('Movie-Embedding')\n",
        "item_em_weights = item_em.get_weights()[0]"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Hk5cOhkRNc8P",
        "outputId": "33f4fc0a-9d19-4d86-f8d4-e1453ad4ed97"
      },
      "source": [
        "item_em_weights[:5]"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[ 0.02110325, -0.00528955, -0.02910987, -0.01247922,  0.02127017],\n",
              "       [-0.01775989, -0.359468  , -0.16436082, -0.2660597 , -0.15996885],\n",
              "       [-0.12550692, -0.21725251, -0.04070576, -0.16161682,  0.08622047],\n",
              "       [ 0.02602149,  0.14989322,  0.12241415,  0.10182509, -0.03654853],\n",
              "       [ 0.07379519,  0.0327619 , -0.06232721, -0.07644718, -0.13666295]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "id": "S-xYpnfxNkJM",
        "outputId": "d98110e3-b5a3-45df-9bfc-7e383bab222d"
      },
      "source": [
        "from sklearn.decomposition import PCA\n",
        "import seaborn as sns\n",
        "\n",
        "plt.figure(figsize=(10,6)) \n",
        "\n",
        "pca = PCA(n_components=2)\n",
        "pca_result = pca.fit_transform(item_em_weights)\n",
        "sns.scatterplot(x=pca_result[:,0], y=pca_result[:,1])"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f2ff44d3f10>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "40vRAKONNrjv",
        "outputId": "631981e9-3faf-49eb-afc3-585ff2f8bd55"
      },
      "source": [
        "item_em_weights = item_em_weights / np.linalg.norm(item_em_weights, axis = 1).reshape((-1, 1))\n",
        "item_em_weights[0][:10]\n",
        "np.sum(np.square(item_em_weights[0]))"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.0000001"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "id": "OY5-m6HVN4yG",
        "outputId": "24aec813-e7d5-4bc0-95a6-188890fa5416"
      },
      "source": [
        "plt.figure(figsize=(10,6)) \n",
        "\n",
        "pca = PCA(n_components=2)\n",
        "pca_result = pca.fit_transform(item_em_weights)\n",
        "sns.scatterplot(x=pca_result[:,0], y=pca_result[:,1])"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f2ff47a8e90>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fq6oqz6If2pt"
      },
      "source": [
        "## Suggesting recommendations"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FHzcBMTVOFfM",
        "outputId": "d5cea583-260e-4821-e86b-5cd2863e6fff"
      },
      "source": [
        "# Creating dataset for making recommendations for the first user\n",
        "item_data = np.array(list(set(df.movie_ids)))\n",
        "item_data[:10]"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 1,  2,  3,  4,  5,  6,  7,  8,  9, 10])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "o3N-kp2kOQRE",
        "outputId": "87a1c9ad-eb06-416b-81a2-0d840a67b971"
      },
      "source": [
        "user = np.array([1 for i in range(len(item_data))])\n",
        "user[:10]"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aFCyeUKjOWQ0",
        "outputId": "0c10c336-2070-496b-a655-4a2a920be802"
      },
      "source": [
        "predictions = model.predict([user, item_data])\n",
        "\n",
        "predictions = np.array([a[0] for a in predictions])\n",
        "\n",
        "recommended_movie_ids = (-predictions).argsort()[:10]\n",
        "\n",
        "recommended_movie_ids"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1174,  542, 1466, 1462, 1404, 1387, 1499, 1448, 1628, 1343])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "r5Ks27D_OqRe",
        "outputId": "33660df7-7ca7-4bbb-cb9b-53a939249ea3"
      },
      "source": [
        "movies = pd.read_csv('/content/drive/MyDrive/Recommendation System/movies.csv', encoding='ISO-8859-1')\n",
        "movies.head()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>movie_title</th>\n",
              "      <th>year_of_release</th>\n",
              "      <th>genre</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>Toy Story</td>\n",
              "      <td>1995.0</td>\n",
              "      <td>Animation|Children's|Comedy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>Jumanji</td>\n",
              "      <td>1995.0</td>\n",
              "      <td>Adventure|Children's|Fantasy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>Grumpier Old Men</td>\n",
              "      <td>1995.0</td>\n",
              "      <td>Comedy|Romance</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>Waiting to Exhale</td>\n",
              "      <td>1995.0</td>\n",
              "      <td>Comedy|Drama</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>Father of the Bride Part II</td>\n",
              "      <td>1995.0</td>\n",
              "      <td>Comedy</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   movie_id  ...                         genre\n",
              "0         1  ...   Animation|Children's|Comedy\n",
              "1         2  ...  Adventure|Children's|Fantasy\n",
              "2         3  ...                Comedy|Romance\n",
              "3         4  ...                  Comedy|Drama\n",
              "4         5  ...                        Comedy\n",
              "\n",
              "[5 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 359
        },
        "id": "64inhK9EXz4E",
        "outputId": "36a9255c-9deb-4c8b-a330-621382fad33a"
      },
      "source": [
        "movies[movies['movie_id'].isin(recommended_movie_ids)]"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>movie_title</th>\n",
              "      <th>year_of_release</th>\n",
              "      <th>genre</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>537</th>\n",
              "      <td>542</td>\n",
              "      <td>Son in Law</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Comedy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1157</th>\n",
              "      <td>1174</td>\n",
              "      <td>Grosse Fatigue</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Comedy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1321</th>\n",
              "      <td>1343</td>\n",
              "      <td>Cape Fear</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Thriller</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1365</th>\n",
              "      <td>1387</td>\n",
              "      <td>Jaws</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Action|Horror</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1380</th>\n",
              "      <td>1404</td>\n",
              "      <td>Night Falls on Manhattan</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Crime|Drama</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1422</th>\n",
              "      <td>1448</td>\n",
              "      <td>Fire on the Mountain</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Documentary</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1434</th>\n",
              "      <td>1462</td>\n",
              "      <td>Unforgotten: Twenty-Five Years After Willowbrook</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Documentary</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1438</th>\n",
              "      <td>1466</td>\n",
              "      <td>Donnie Brasco</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Crime|Drama</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1466</th>\n",
              "      <td>1499</td>\n",
              "      <td>Anaconda</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Action|Adventure|Thriller</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1584</th>\n",
              "      <td>1628</td>\n",
              "      <td>Locusts</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "      movie_id  ...                      genre\n",
              "537        542  ...                     Comedy\n",
              "1157      1174  ...                     Comedy\n",
              "1321      1343  ...                   Thriller\n",
              "1365      1387  ...              Action|Horror\n",
              "1380      1404  ...                Crime|Drama\n",
              "1422      1448  ...                Documentary\n",
              "1434      1462  ...                Documentary\n",
              "1438      1466  ...                Crime|Drama\n",
              "1466      1499  ...  Action|Adventure|Thriller\n",
              "1584      1628  ...                        NaN\n",
              "\n",
              "[10 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    }
  ]
}