SOM_and_Recommendation.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "SOM and Recommendation.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "RTcLD5OMSEfi",
        "outputId": "bfe3003c-7c64-475c-b11c-8206b5df062b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 203
        }
      },
      "source": [
        "!pip install MiniSom"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting MiniSom\n",
            "  Downloading https://files.pythonhosted.org/packages/9d/10/a1c1621000d5ca00c41695689551c1a4d6d245d7bbf099d81e067da3e8f2/MiniSom-2.2.6.tar.gz\n",
            "Building wheels for collected packages: MiniSom\n",
            "  Building wheel for MiniSom (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for MiniSom: filename=MiniSom-2.2.6-cp36-none-any.whl size=8525 sha256=0628d00d7b1a132faf9b97ccbdedbb422340c6184f2d7406a522b3754a4bc4aa\n",
            "  Stored in directory: /root/.cache/pip/wheels/b8/c6/01/330066e36e1f7c826c96f656f9185822cfcdef0591315949ea\n",
            "Successfully built MiniSom\n",
            "Installing collected packages: MiniSom\n",
            "Successfully installed MiniSom-2.2.6\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Jc1NHo6xSJa-"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qc-LYMBWTbRE",
        "outputId": "9e31a854-8cdb-4081-a419-54b0ca49d8bd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 165
        }
      },
      "source": [
        "dataset = pd.read_csv('Credit_Card_Applications.csv')\n",
        "print(dataset.head())\n",
        "X = dataset.iloc[:, :-1].values \n",
        "y = dataset.iloc[:, -1].values"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   CustomerID  A1     A2     A3  A4  A5  ...  A10  A11  A12  A13   A14  Class\n",
            "0    15776156   1  22.08  11.46   2   4  ...    0    1    2  100  1213      0\n",
            "1    15739548   0  22.67   7.00   2   8  ...    0    0    2  160     1      0\n",
            "2    15662854   0  29.58   1.75   1   4  ...    0    1    2  280     1      0\n",
            "3    15687688   0  21.67  11.50   1   5  ...   11    1    2    0     1      1\n",
            "4    15715750   1  20.17   8.17   2   6  ...   14    0    2   60   159      1\n",
            "\n",
            "[5 rows x 16 columns]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QC1tOmWLTv0-"
      },
      "source": [
        "from sklearn.preprocessing import MinMaxScaler\n",
        "sc = MinMaxScaler(feature_range = (0,1))\n",
        "X = sc.fit_transform(X)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cg613BLaT7pi",
        "outputId": "9fbb0c97-0fb6-409c-e2e3-0b7bee345415",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 276
        }
      },
      "source": [
        "print(X.shape)\n",
        "X"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(690, 15)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[8.42681467e-01, 1.00000000e+00, 1.25263158e-01, ...,\n",
              "        5.00000000e-01, 5.00000000e-02, 1.21200000e-02],\n",
              "       [6.96090562e-01, 0.00000000e+00, 1.34135338e-01, ...,\n",
              "        5.00000000e-01, 8.00000000e-02, 0.00000000e+00],\n",
              "       [3.88981656e-01, 0.00000000e+00, 2.38045113e-01, ...,\n",
              "        5.00000000e-01, 1.40000000e-01, 0.00000000e+00],\n",
              "       ...,\n",
              "       [4.39420332e-01, 0.00000000e+00, 7.63909774e-02, ...,\n",
              "        5.00000000e-01, 5.00000000e-02, 0.00000000e+00],\n",
              "       [8.44034934e-01, 0.00000000e+00, 2.05563910e-01, ...,\n",
              "        5.00000000e-01, 6.00000000e-02, 1.10000000e-04],\n",
              "       [1.06907888e-01, 1.00000000e+00, 4.09774436e-01, ...,\n",
              "        0.00000000e+00, 2.80000000e-01, 0.00000000e+00]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z10yLAe6T8sN"
      },
      "source": [
        "from minisom import MiniSom\n",
        "som = MiniSom(x=10, y=10, input_len= 15, sigma= 1.0, learning_rate = 0.05)\n",
        "som.random_weights_init(X)\n",
        "som.train_random(data = X, num_iteration = 200)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MkzoNEZ55Gty",
        "outputId": "c4439e49-ad2d-4454-b31b-e03ae58b96ff",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "som.distance_map()[0][3]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.5773222540305397"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yOwB46EXUBem",
        "outputId": "efd04880-c026-4c8b-a500-d37b93583653",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        }
      },
      "source": [
        "from pylab import bone, pcolor, colorbar, plot, show\n",
        "bone()\n",
        "pcolor(som.distance_map().T) # returns mean of the neuron distance for winning nodes\n",
        "colorbar() # adds legend\n",
        "markers = ['o', 's'] # red o is cust who didn't get approval, s is square, who got approval\n",
        "colors = ['r', 'g']\n",
        "for i, x in enumerate(X):\n",
        "    w = som.winner(x)\n",
        "    plot(w[0] + 0.5,\n",
        "         w[1] + 0.5,\n",
        "         markers[y[i]],\n",
        "         markeredgecolor = colors[y[i]], # if cust didn't get approval y[i] = 0, this marker[0] = red circle \n",
        "         markerfacecolor = 'None',\n",
        "         markersize = 10,\n",
        "         markeredgewidth = 2)\n",
        "show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WrMPTbuQYuyf",
        "outputId": "fd4d34fc-fb9d-40bf-b338-db7f9feacfea",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        }
      },
      "source": [
        "mappings = som.win_map(X)\n",
        "print(len(mappings[(6,1)]))\n",
        "print('and')\n",
        "print(len(mappings[(4,1)]))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "6\n",
            "and\n",
            "5\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fufNLwj7UImU"
      },
      "source": [
        "frauds = np.concatenate((mappings[(6,1)], mappings[(4,1)]), axis = 0)\n",
        "frauds = sc.inverse_transform(frauds)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hApKZtoUUIpz",
        "outputId": "d65dcfe8-f2fb-4aec-8a52-6a99e5529d79",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 239
        }
      },
      "source": [
        "print('Fraud Customer IDs')\n",
        "for i in frauds[:, 0]:\n",
        "  print(int(i))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fraud Customer IDs\n",
            "15815271\n",
            "15744273\n",
            "15570990\n",
            "15789865\n",
            "15618703\n",
            "15810716\n",
            "15684722\n",
            "15623369\n",
            "15650591\n",
            "15633608\n",
            "15675450\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6RkbMdVNoThF"
      },
      "source": [
        "# Recommendation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SBBscV0Coc1d",
        "outputId": "6af48750-92df-4384-839f-dbd1529a28db",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 349
        }
      },
      "source": [
        "!wget \"http://files.grouplens.org/datasets/movielens/ml-1m.zip\"\n",
        "!unzip ml-1m.zip\n",
        "!ls"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-09-15 12:37:46--  http://files.grouplens.org/datasets/movielens/ml-1m.zip\n",
            "Resolving files.grouplens.org (files.grouplens.org)... 128.101.65.152\n",
            "Connecting to files.grouplens.org (files.grouplens.org)|128.101.65.152|:80... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 5917549 (5.6M) [application/zip]\n",
            "Saving to: ‘ml-1m.zip’\n",
            "\n",
            "ml-1m.zip           100%[===================>]   5.64M  18.8MB/s    in 0.3s    \n",
            "\n",
            "2020-09-15 12:37:46 (18.8 MB/s) - ‘ml-1m.zip’ saved [5917549/5917549]\n",
            "\n",
            "Archive:  ml-1m.zip\n",
            "   creating: ml-1m/\n",
            "  inflating: ml-1m/movies.dat        \n",
            "  inflating: ml-1m/ratings.dat       \n",
            "  inflating: ml-1m/README            \n",
            "  inflating: ml-1m/users.dat         \n",
            "ml-1m  ml-1m.zip  sample_data\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ugb3v2SRogAM",
        "outputId": "ddae4c02-afec-4686-8df9-b963b2e64e70",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 700
        }
      },
      "source": [
        "!wget \"http://files.grouplens.org/datasets/movielens/ml-100k.zip\"\n",
        "!unzip ml-100k.zip\n",
        "!ls"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-09-15 12:37:47--  http://files.grouplens.org/datasets/movielens/ml-100k.zip\n",
            "Resolving files.grouplens.org (files.grouplens.org)... 128.101.65.152\n",
            "Connecting to files.grouplens.org (files.grouplens.org)|128.101.65.152|:80... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 4924029 (4.7M) [application/zip]\n",
            "Saving to: ‘ml-100k.zip’\n",
            "\n",
            "ml-100k.zip         100%[===================>]   4.70M  16.3MB/s    in 0.3s    \n",
            "\n",
            "2020-09-15 12:37:47 (16.3 MB/s) - ‘ml-100k.zip’ saved [4924029/4924029]\n",
            "\n",
            "Archive:  ml-100k.zip\n",
            "   creating: ml-100k/\n",
            "  inflating: ml-100k/allbut.pl       \n",
            "  inflating: ml-100k/mku.sh          \n",
            "  inflating: ml-100k/README          \n",
            "  inflating: ml-100k/u.data          \n",
            "  inflating: ml-100k/u.genre         \n",
            "  inflating: ml-100k/u.info          \n",
            "  inflating: ml-100k/u.item          \n",
            "  inflating: ml-100k/u.occupation    \n",
            "  inflating: ml-100k/u.user          \n",
            "  inflating: ml-100k/u1.base         \n",
            "  inflating: ml-100k/u1.test         \n",
            "  inflating: ml-100k/u2.base         \n",
            "  inflating: ml-100k/u2.test         \n",
            "  inflating: ml-100k/u3.base         \n",
            "  inflating: ml-100k/u3.test         \n",
            "  inflating: ml-100k/u4.base         \n",
            "  inflating: ml-100k/u4.test         \n",
            "  inflating: ml-100k/u5.base         \n",
            "  inflating: ml-100k/u5.test         \n",
            "  inflating: ml-100k/ua.base         \n",
            "  inflating: ml-100k/ua.test         \n",
            "  inflating: ml-100k/ub.base         \n",
            "  inflating: ml-100k/ub.test         \n",
            "ml-100k  ml-100k.zip  ml-1m  ml-1m.zip\tsample_data\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wuRpY8--UIk8",
        "outputId": "d0b35721-7f34-4d2f-988c-17aa0ea0031f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        }
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.parallel\n",
        "import torch.optim as optim\n",
        "import torch.utils.data\n",
        "from torch.autograd import Variable\n",
        "\n",
        "# We won't be using this dataset.\n",
        "movies = pd.read_csv('ml-1m/movies.dat', sep = '::', header = None, engine = 'python', encoding = 'latin-1')\n",
        "users = pd.read_csv('ml-1m/users.dat', sep = '::', header = None, engine = 'python', encoding = 'latin-1')\n",
        "ratings = pd.read_csv('ml-1m/ratings.dat', sep = '::', header = None, engine = 'python', encoding = 'latin-1')\n",
        "\n",
        "print(movies.head())\n",
        "print('Users \\n')\n",
        "print(users.head())\n",
        "print('Ratings \\n')\n",
        "print(ratings.head())"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   0                                   1                             2\n",
            "0  1                    Toy Story (1995)   Animation|Children's|Comedy\n",
            "1  2                      Jumanji (1995)  Adventure|Children's|Fantasy\n",
            "2  3             Grumpier Old Men (1995)                Comedy|Romance\n",
            "3  4            Waiting to Exhale (1995)                  Comedy|Drama\n",
            "4  5  Father of the Bride Part II (1995)                        Comedy\n",
            "Users \n",
            "\n",
            "   0  1   2   3      4\n",
            "0  1  F   1  10  48067\n",
            "1  2  M  56  16  70072\n",
            "2  3  M  25  15  55117\n",
            "3  4  M  45   7  02460\n",
            "4  5  M  25  20  55455\n",
            "Ratings \n",
            "\n",
            "   0     1  2          3\n",
            "0  1  1193  5  978300760\n",
            "1  1   661  3  978302109\n",
            "2  1   914  3  978301968\n",
            "3  1  3408  4  978300275\n",
            "4  1  2355  5  978824291\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NKTlJ_zcxV7r",
        "outputId": "c849c82d-2633-4772-a893-14912b0ff63b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        }
      },
      "source": [
        "ratings"
      ],
      "execution_count": null,
      "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>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>1193</td>\n",
              "      <td>5</td>\n",
              "      <td>978300760</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>661</td>\n",
              "      <td>3</td>\n",
              "      <td>978302109</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>914</td>\n",
              "      <td>3</td>\n",
              "      <td>978301968</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>3408</td>\n",
              "      <td>4</td>\n",
              "      <td>978300275</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>2355</td>\n",
              "      <td>5</td>\n",
              "      <td>978824291</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1000204</th>\n",
              "      <td>6040</td>\n",
              "      <td>1091</td>\n",
              "      <td>1</td>\n",
              "      <td>956716541</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1000205</th>\n",
              "      <td>6040</td>\n",
              "      <td>1094</td>\n",
              "      <td>5</td>\n",
              "      <td>956704887</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1000206</th>\n",
              "      <td>6040</td>\n",
              "      <td>562</td>\n",
              "      <td>5</td>\n",
              "      <td>956704746</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1000207</th>\n",
              "      <td>6040</td>\n",
              "      <td>1096</td>\n",
              "      <td>4</td>\n",
              "      <td>956715648</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1000208</th>\n",
              "      <td>6040</td>\n",
              "      <td>1097</td>\n",
              "      <td>4</td>\n",
              "      <td>956715569</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1000209 rows × 4 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "            0     1  2          3\n",
              "0           1  1193  5  978300760\n",
              "1           1   661  3  978302109\n",
              "2           1   914  3  978301968\n",
              "3           1  3408  4  978300275\n",
              "4           1  2355  5  978824291\n",
              "...       ...   ... ..        ...\n",
              "1000204  6040  1091  1  956716541\n",
              "1000205  6040  1094  5  956704887\n",
              "1000206  6040   562  5  956704746\n",
              "1000207  6040  1096  4  956715648\n",
              "1000208  6040  1097  4  956715569\n",
              "\n",
              "[1000209 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9dwP-CDKUIjL"
      },
      "source": [
        "training_set = pd.read_csv('ml-100k/u1.base', delimiter = '\\t')\n",
        "training_set = np.array(training_set, dtype = 'int')\n",
        "test_set = pd.read_csv('ml-100k/u1.test', delimiter = '\\t')\n",
        "test_set = np.array(test_set, dtype = 'int')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EbK3yyI1dCbh",
        "outputId": "99e195af-1012-4f93-c76a-3b44db0906d7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 202
        }
      },
      "source": [
        "training_set[:10]\n",
        "#user movie rating timestamp"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[        1,         2,         3, 876893171],\n",
              "       [        1,         3,         4, 878542960],\n",
              "       [        1,         4,         3, 876893119],\n",
              "       [        1,         5,         3, 889751712],\n",
              "       [        1,         7,         4, 875071561],\n",
              "       [        1,         8,         1, 875072484],\n",
              "       [        1,         9,         5, 878543541],\n",
              "       [        1,        11,         2, 875072262],\n",
              "       [        1,        13,         5, 875071805],\n",
              "       [        1,        15,         5, 875071608]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "I5HHP3OgU4FB",
        "outputId": "98b29b3f-fab7-4d5b-f989-f1bf632d0989",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "nb_users = int(max(max(training_set[:, 0], ), max(test_set[:, 0])))\n",
        "nb_movies = int(max(max(training_set[:, 1], ), max(test_set[:, 1])))\n",
        "nb_users,nb_movies"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(943, 1682)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ARomKbmDySYP",
        "outputId": "cf4b006c-e351-47ea-a847-9ceba0899d41",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        }
      },
      "source": [
        "ratings = np.zeros(nb_movies)\n",
        "ratings[training_set[:,1][training_set[:,0]==1]] = training_set[:,2][training_set[:,0]==1]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "SyntaxError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;36m  File \u001b[0;32m\"<ipython-input-17-94660fc33204>\"\u001b[0;36m, line \u001b[0;32m2\u001b[0m\n\u001b[0;31m    ratings[[0,1,2,3]]] = training_set[:,2][training_set[:,0]==1]\u001b[0m\n\u001b[0m                      ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m invalid syntax\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MbO1sxAcy3-7",
        "outputId": "ca2235f2-3297-44d0-9c01-ee5fa3ba43a4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "ratings"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0., 0., 3., ..., 0., 0., 0.])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nZwXa_upWpRU",
        "outputId": "bdd7b4e1-f83d-4e05-fe4d-c05a2d6e9fb8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 55
        }
      },
      "source": [
        "# we want users in row, and movies in columns\n",
        "def convert(data):\n",
        "  new_data = []\n",
        "  for i in range(1,nb_users+1):\n",
        "    id_movies = data[:,1][data[:,0]==i]\n",
        "    id_ratings = data[:,2][data[:,0]==i]\n",
        "    ratings = np.zeros(nb_movies)\n",
        "    ratings[id_movies-1] = id_ratings\n",
        "    new_data.append([ratings])\n",
        "  return new_data\n",
        "  \n",
        "training_set = convert(training_set)\n",
        "test_set = convert(test_set)\n",
        "print(training_set[:10])\n",
        "training_set = torch.FloatTensor(training_set)\n",
        "test_set = torch.FloatTensor(test_set)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[array([0., 3., 4., ..., 0., 0., 0.])], [array([4., 0., 0., ..., 0., 0., 0.])], [array([0., 0., 0., ..., 0., 0., 0.])], [array([0., 0., 0., ..., 0., 0., 0.])], [array([0., 0., 0., ..., 0., 0., 0.])], [array([4., 0., 0., ..., 0., 0., 0.])], [array([0., 0., 0., ..., 0., 0., 0.])], [array([0., 0., 0., ..., 0., 0., 0.])], [array([0., 0., 0., ..., 0., 0., 0.])], [array([0., 0., 0., ..., 0., 0., 0.])]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Q5i9szIdb_v5",
        "outputId": "eaf3abe5-88e3-4c18-c44a-855e35d47884",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 257
        }
      },
      "source": [
        "print(training_set[:10])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "tensor([[[0., 3., 4.,  ..., 0., 0., 0.]],\n",
            "\n",
            "        [[4., 0., 0.,  ..., 0., 0., 0.]],\n",
            "\n",
            "        [[0., 0., 0.,  ..., 0., 0., 0.]],\n",
            "\n",
            "        ...,\n",
            "\n",
            "        [[0., 0., 0.,  ..., 0., 0., 0.]],\n",
            "\n",
            "        [[0., 0., 0.,  ..., 0., 0., 0.]],\n",
            "\n",
            "        [[0., 0., 0.,  ..., 0., 0., 0.]]])\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iAAfCF4OhVek"
      },
      "source": [
        "training_set[training_set == 0] = -1\n",
        "training_set[training_set == 1] = 0\n",
        "training_set[training_set == 2] = 0\n",
        "training_set[training_set >= 3] = 1\n",
        "\n",
        "test_set[test_set == 0] = -1\n",
        "test_set[test_set == 1] = 0\n",
        "test_set[test_set == 2] = 0\n",
        "test_set[test_set >= 3] = 1"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oJ0Zk__mhWDL",
        "outputId": "bb548d1b-b981-46cd-d347-b75362417bbe",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 257
        }
      },
      "source": [
        "print(training_set[:10])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "tensor([[[-1.,  1.,  1.,  ..., -1., -1., -1.]],\n",
            "\n",
            "        [[ 1., -1., -1.,  ..., -1., -1., -1.]],\n",
            "\n",
            "        [[-1., -1., -1.,  ..., -1., -1., -1.]],\n",
            "\n",
            "        ...,\n",
            "\n",
            "        [[-1., -1., -1.,  ..., -1., -1., -1.]],\n",
            "\n",
            "        [[-1., -1., -1.,  ..., -1., -1., -1.]],\n",
            "\n",
            "        [[-1., -1., -1.,  ..., -1., -1., -1.]]])\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3rj2aWjGhWG3"
      },
      "source": [
        "class RBM():\n",
        "  def __init__(self, nv, nh):\n",
        "    self.W = torch.tensor((nh, nv))\n",
        "    self.a = torch.tensor((1, nh))\n",
        "    self.b = torch.tensor((1, nv))\n",
        "  def sample_h(self, x):\n",
        "    wx = torch.mm(x, self.W.t())\n",
        "    activation = wx + self.a.expand_as(wx)\n",
        "    p_h_given_v = torch.sigmoid(activation)\n",
        "    return p_h_given_v, torch.bernoulli(p_h_given_v)\n",
        "  def sample_v(self, y):\n",
        "    wy = torch.mm(y, self.W)\n",
        "    activation = wy + self.b.expand_as(wy)\n",
        "    p_v_given_h = torch.sigmoid(activation)\n",
        "    return p_v_given_h, torch.bernoulli(p_v_given_h)\n",
        "  def train(self, v0, vk, ph0, phk):\n",
        "    self.W += (torch.mm(v0.t(), ph0) - torch.mm(vk.t(), phk)).t()\n",
        "    self.b += torch.sum((v0 - vk), 0)\n",
        "    self.a += torch.sum((ph0 - phk), 0)\n",
        "nv = len(training_set[0])\n",
        "nh = 100\n",
        "batch_size = 100\n",
        "rbm = RBM(nv, nh)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jhcY16KvhWB9",
        "outputId": "0a6506ba-a967-48cf-bad5-ee893d80cfb6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 376
        }
      },
      "source": [
        "nb_epoch = 10\n",
        "for epoch in range(1, nb_epoch + 1):\n",
        "  train_loss = 0\n",
        "  s = 0.\n",
        "  for id_user in range(0, nb_users - batch_size, batch_size):\n",
        "    vk = training_set[id_user : id_user + batch_size]\n",
        "    v0 = training_set[id_user : id_user + batch_size]\n",
        "    ph0,_ = rbm.sample_h(v0)\n",
        "    for k in range(10):\n",
        "      _,hk = rbm.sample_h(vk)\n",
        "      _,vk = rbm.sample_v(hk)\n",
        "      vk[v0<0] = v0[v0<0]\n",
        "    phk,_ = rbm.sample_h(vk)\n",
        "    rbm.train(v0, vk, ph0, phk)\n",
        "    train_loss += torch.mean(torch.abs(v0[v0 >= 0] - vk[v0 >= 0]))\n",
        "    s += 1.\n",
        "  print('epoch: '+str(epoch)+' loss: '+str(train_loss/s))\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "IndexError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-22-c25d531868c4>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      6\u001b[0m     \u001b[0mvk\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtraining_set\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mid_user\u001b[0m \u001b[0;34m:\u001b[0m \u001b[0mid_user\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m     \u001b[0mv0\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtraining_set\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mid_user\u001b[0m \u001b[0;34m:\u001b[0m \u001b[0mid_user\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m     \u001b[0mph0\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0m_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrbm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msample_h\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mv0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      9\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mk\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m10\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     10\u001b[0m       \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mhk\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrbm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msample_h\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvk\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<ipython-input-21-d746b383f756>\u001b[0m in \u001b[0;36msample_h\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m      5\u001b[0m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnv\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0msample_h\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 7\u001b[0;31m     \u001b[0mwx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mW\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mt\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      8\u001b[0m     \u001b[0mactivation\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mwx\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0ma\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexpand_as\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mwx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      9\u001b[0m     \u001b[0mp_h_given_v\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msigmoid\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mactivation\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mIndexError\u001b[0m: Dimension out of range (expected to be in range of [-1, 0], but got 1)"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LYmUkE0O5CeY"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}