Training_Deep_AE.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Training Deep AE.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "metadata": {
        "id": "CF0dq-NUAgUM",
        "colab_type": "code",
        "outputId": "3309746e-c68c-439e-d57c-23c3fa659669",
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 244
        }
      },
      "cell_type": "code",
      "source": [
        "from google.colab import files\n",
        "uploaded = files.upload()"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-7033ef6c-e652-45be-9428-072b2e0db6a4\" name=\"files[]\" multiple disabled />\n",
              "     <output id=\"result-7033ef6c-e652-45be-9428-072b2e0db6a4\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Saving loss.py to loss.py\n",
            "Saving model.py to model.py\n",
            "Saving test.csv to test.csv\n",
            "Saving TestDataset.py to TestDataset.py\n",
            "Saving train.csv to train.csv\n",
            "Saving TrainDataset.py to TrainDataset.py\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "_wckq9iHQclG",
        "colab_type": "code",
        "outputId": "ee909613-08cc-4cf1-fbf6-3999e854cbd2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 751
        }
      },
      "cell_type": "code",
      "source": [
        "!pip install livelossplot"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting livelossplot\n",
            "  Downloading https://files.pythonhosted.org/packages/07/9d/54f8a93d65eece0bcd475b191c4c9a3bff9dbf993db8d5e2d02b76c2d2c3/livelossplot-0.3.3-py3-none-any.whl\n",
            "Requirement already satisfied: notebook in /usr/local/lib/python3.6/dist-packages (from livelossplot) (5.2.2)\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.6/dist-packages (from livelossplot) (3.0.3)\n",
            "Requirement already satisfied: tornado>=4 in /usr/local/lib/python3.6/dist-packages (from notebook->livelossplot) (4.5.3)\n",
            "Requirement already satisfied: traitlets>=4.2.1 in /usr/local/lib/python3.6/dist-packages (from notebook->livelossplot) (4.3.2)\n",
            "Requirement already satisfied: ipython-genutils in /usr/local/lib/python3.6/dist-packages (from notebook->livelossplot) (0.2.0)\n",
            "Requirement already satisfied: jupyter-client in /usr/local/lib/python3.6/dist-packages (from notebook->livelossplot) (5.2.4)\n",
            "Requirement already satisfied: nbconvert in /usr/local/lib/python3.6/dist-packages (from notebook->livelossplot) (5.4.1)\n",
            "Requirement already satisfied: jupyter-core in /usr/local/lib/python3.6/dist-packages (from notebook->livelossplot) (4.4.0)\n",
            "Requirement already satisfied: nbformat in /usr/local/lib/python3.6/dist-packages (from notebook->livelossplot) (4.4.0)\n",
            "Requirement already satisfied: terminado>=0.3.3; sys_platform != \"win32\" in /usr/local/lib/python3.6/dist-packages (from notebook->livelossplot) (0.8.2)\n",
            "Requirement already satisfied: ipykernel in /usr/local/lib/python3.6/dist-packages (from notebook->livelossplot) (4.6.1)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.6/dist-packages (from notebook->livelossplot) (2.10)\n",
            "Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib->livelossplot) (2.5.3)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.6/dist-packages (from matplotlib->livelossplot) (0.10.0)\n",
            "Requirement already satisfied: numpy>=1.10.0 in /usr/local/lib/python3.6/dist-packages (from matplotlib->livelossplot) (1.14.6)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib->livelossplot) (2.3.1)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib->livelossplot) (1.0.1)\n",
            "Requirement already satisfied: decorator in /usr/local/lib/python3.6/dist-packages (from traitlets>=4.2.1->notebook->livelossplot) (4.4.0)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from traitlets>=4.2.1->notebook->livelossplot) (1.11.0)\n",
            "Requirement already satisfied: pyzmq>=13 in /usr/local/lib/python3.6/dist-packages (from jupyter-client->notebook->livelossplot) (17.0.0)\n",
            "Requirement already satisfied: testpath in /usr/local/lib/python3.6/dist-packages (from nbconvert->notebook->livelossplot) (0.4.2)\n",
            "Requirement already satisfied: entrypoints>=0.2.2 in /usr/local/lib/python3.6/dist-packages (from nbconvert->notebook->livelossplot) (0.3)\n",
            "Requirement already satisfied: defusedxml in /usr/local/lib/python3.6/dist-packages (from nbconvert->notebook->livelossplot) (0.5.0)\n",
            "Requirement already satisfied: pandocfilters>=1.4.1 in /usr/local/lib/python3.6/dist-packages (from nbconvert->notebook->livelossplot) (1.4.2)\n",
            "Requirement already satisfied: mistune>=0.8.1 in /usr/local/lib/python3.6/dist-packages (from nbconvert->notebook->livelossplot) (0.8.4)\n",
            "Requirement already satisfied: pygments in /usr/local/lib/python3.6/dist-packages (from nbconvert->notebook->livelossplot) (2.1.3)\n",
            "Requirement already satisfied: bleach in /usr/local/lib/python3.6/dist-packages (from nbconvert->notebook->livelossplot) (3.1.0)\n",
            "Requirement already satisfied: jsonschema!=2.5.0,>=2.4 in /usr/local/lib/python3.6/dist-packages (from nbformat->notebook->livelossplot) (2.6.0)\n",
            "Requirement already satisfied: ptyprocess; os_name != \"nt\" in /usr/local/lib/python3.6/dist-packages (from terminado>=0.3.3; sys_platform != \"win32\"->notebook->livelossplot) (0.6.0)\n",
            "Requirement already satisfied: ipython>=4.0.0 in /usr/local/lib/python3.6/dist-packages (from ipykernel->notebook->livelossplot) (5.5.0)\n",
            "Requirement already satisfied: MarkupSafe>=0.23 in /usr/local/lib/python3.6/dist-packages (from jinja2->notebook->livelossplot) (1.1.1)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from kiwisolver>=1.0.1->matplotlib->livelossplot) (40.9.0)\n",
            "Requirement already satisfied: webencodings in /usr/local/lib/python3.6/dist-packages (from bleach->nbconvert->notebook->livelossplot) (0.5.1)\n",
            "Requirement already satisfied: pexpect; sys_platform != \"win32\" in /usr/local/lib/python3.6/dist-packages (from ipython>=4.0.0->ipykernel->notebook->livelossplot) (4.6.0)\n",
            "Requirement already satisfied: simplegeneric>0.8 in /usr/local/lib/python3.6/dist-packages (from ipython>=4.0.0->ipykernel->notebook->livelossplot) (0.8.1)\n",
            "Requirement already satisfied: pickleshare in /usr/local/lib/python3.6/dist-packages (from ipython>=4.0.0->ipykernel->notebook->livelossplot) (0.7.5)\n",
            "Requirement already satisfied: prompt-toolkit<2.0.0,>=1.0.4 in /usr/local/lib/python3.6/dist-packages (from ipython>=4.0.0->ipykernel->notebook->livelossplot) (1.0.15)\n",
            "Requirement already satisfied: wcwidth in /usr/local/lib/python3.6/dist-packages (from prompt-toolkit<2.0.0,>=1.0.4->ipython>=4.0.0->ipykernel->notebook->livelossplot) (0.1.7)\n",
            "Installing collected packages: livelossplot\n",
            "Successfully installed livelossplot-0.3.3\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "6VLw_wzeAqa9",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torchvision\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from tqdm import tqdm\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim\n",
        "from torch.autograd import Variable\n",
        "from torch.utils.data import Dataset, DataLoader\n",
        "from torchvision import transforms, utils\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.pyplot as pyplot\n",
        "from livelossplot import PlotLosses\n",
        "from loss import MSEloss_with_Mask\n",
        "from model import AutoEncoder\n",
        "from TestDataset import TestDataset\n",
        "from TrainDataset import TrainDataset"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "wVhNWiG8JZvr",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "transformations = transforms.Compose([transforms.ToTensor()])\n",
        "train_dat = TrainDataset('train.csv', transformations)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "_4dX-G1bJkrb",
        "colab_type": "code",
        "outputId": "975a7d5e-8ee6-489b-bb24-4e5db39d1d97",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "print(\"Training Length: \", train_dat.__len__())\n",
        "print(\"6th User Ratings: \", train_dat.__getitem__(7))"
      ],
      "execution_count": 292,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training Length:  595\n",
            "6th User Ratings:  tensor([0., 4., 0.,  ..., 0., 0., 0.], dtype=torch.float64)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "SpHPV7sLKSK5",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "test_dat = TestDataset('test.csv', transformations)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "XEWIGA4fKlaK",
        "colab_type": "code",
        "outputId": "e65195e4-36ff-4483-e384-a506d5ba0b7c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "print(\"Testing Length: \", test_dat.__len__())\n",
        "print(\"6th User Ratings: \", test_dat.__getitem__(4))"
      ],
      "execution_count": 296,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Testing Length:  595\n",
            "6th User Ratings:  tensor([4., 0., 0.,  ..., 0., 0., 0.], dtype=torch.float64)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "QLy46VQzMlhG",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Dataloader"
      ]
    },
    {
      "metadata": {
        "id": "H6JjuurPKy56",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "train_dl = DataLoader(dataset=train_dat, batch_size = 128, shuffle=True, num_workers = 1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "RVpG9OuHMsa6",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "test_dl = DataLoader(dataset=test_dat, batch_size=128, shuffle=True, num_workers=1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "6WhXvSEeMuH4",
        "colab_type": "code",
        "outputId": "4d2058c7-ecb4-4dd4-efa3-09376706cd80",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "# Checking for availability of GPU\n",
        "\n",
        "use_cuda = torch.cuda.is_available()\n",
        "device = torch.device(\"cuda:0\" if use_cuda else \"cpu\")\n",
        "print(device)"
      ],
      "execution_count": 273,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "cuda:0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "CFU2Obd5UuTF",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Training the Model"
      ]
    },
    {
      "metadata": {
        "id": "5POnWp4pU0z4",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Model"
      ]
    },
    {
      "metadata": {
        "id": "wuWfH8uJM3q6",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "layer_sizes = [9559, 512, 512, 1024]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "yQ0yrC6xUgt5",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model = AutoEncoder(layer_sizes=layer_sizes, nl_type='selu', is_constrained=True, dp_drop_prob=0.0, last_layer_activations=False)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Rr_Fnnx1UnB6",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model = model.cuda()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "vRdekXMJU5i_",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Loss and Optimizer"
      ]
    },
    {
      "metadata": {
        "id": "J2TTqBscU47D",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Loss Function\n",
        "\n",
        "criterion = MSEloss_with_Mask()\n",
        "criterion = criterion.cuda()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "PbwPphAzUp2I",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Optimizer\n",
        "\n",
        "# optimizer = optim.SGD(model.parameters(), lr=0.01, weight_decay= 1e-6, momentum = 0.9, nesterov = True)\n",
        "optimizer = optim.Adam(model.parameters(), lr=0.001)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "0fCHMZCtVR1i",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Fitting the Model"
      ]
    },
    {
      "metadata": {
        "id": "Msy_9GiWCKhr",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# hl1_tr_loss = []\n",
        "# hl2_val_loss = []"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "jglVF6-0GF4e",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# hl1_tr_loss = []\n",
        "# hl2_val_loss = []"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "q-MMP7GDVKnY",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def train(model, criterion, optimizer, train_dl, test_dl, num_epochs=40):\n",
        "  # We will run for 40 epochs\n",
        "  liveloss = PlotLosses()\n",
        "  for epoch in range(num_epochs):\n",
        "    train_loss, valid_loss = [], []\n",
        "    logs = {}\n",
        "    prefix = ''\n",
        "\n",
        "    # Training Part\n",
        "    model.train()\n",
        "    for i, data in enumerate(train_dl, 0):\n",
        "      # Get the inputs\n",
        "      inputs = labels = data\n",
        "      inputs = inputs.cuda()\n",
        "      labels = labels.cuda()\n",
        "\n",
        "      inputs = inputs.float()\n",
        "      labels = labels.float()\n",
        "\n",
        "      # zero the parameter gradients\n",
        "      optimizer.zero_grad()\n",
        "\n",
        "      # forward + backward + optimize\n",
        "      outputs = model(inputs)\n",
        "      outputs = outputs.cuda()\n",
        "      loss = criterion(outputs,labels)\n",
        "      loss.backward()\n",
        "      optimizer.step()\n",
        "\n",
        "      ## -> Iterative Dense Output Re-feeding <- ##\n",
        "      \n",
        "      # Add a \"for\" loop to iterate as much you want\n",
        "      \n",
        "      # Zero the gradiants\n",
        "      optimizer.zero_grad()\n",
        "      \n",
        "      # Important -> detach() the output, to avoid unecessary construction of \n",
        "      # the computational graph\n",
        "      outputs = model(outputs.detach())\n",
        "      outputs = outputs.cuda()\n",
        "      loss = criterion(outputs, labels)\n",
        "      loss.backward()\n",
        "      optimizer.step()\n",
        "\n",
        "      train_loss.append(loss.item())\n",
        "      logs[prefix + 'MMSE loss'] = loss.item()\n",
        "\n",
        "    for i, data in enumerate(test_dl, 0):\n",
        "      model.eval()\n",
        "      inputs = labels = data\n",
        "      inputs = inputs.cuda()\n",
        "      labels = labels.cuda()\n",
        "\n",
        "      inputs = inputs.float()\n",
        "      labels = labels.float()\n",
        "\n",
        "      outputs = model(inputs)\n",
        "      outputs = outputs.cuda()\n",
        "      loss = criterion(outputs, labels)\n",
        "\n",
        "      valid_loss.append(loss.item())\n",
        "      prefix = 'val_'\n",
        "      logs[prefix + 'MMSE loss'] = loss.item()\n",
        "\n",
        "    lr2_tr_loss.append(np.mean(train_loss))\n",
        "    lr2_val_loss.append(np.mean(valid_loss))\n",
        "    liveloss.update(logs)\n",
        "    liveloss.draw()\n",
        "    print (\"Epoch:\", epoch+1, \" Training Loss: \", np.mean(train_loss), \" Valid Loss: \", np.mean(valid_loss))\n",
        "    if epoch == num_epochs -1:\n",
        "      return outputs"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "xXcdegFsuu_V",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 390
        },
        "outputId": "997b51cd-ded5-4aca-c4e3-aafdbd484596"
      },
      "cell_type": "code",
      "source": [
        "out = train(model, criterion, optimizer, train_dl, test_dl, 40)"
      ],
      "execution_count": 282,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbEAAAExCAYAAADyYJxpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "MMSE loss:\n",
            "training   (min:    0.310, max:    4.696, cur:    0.348)\n",
            "validation (min:    0.149, max:    1.702, cur:    0.149)\n",
            "Epoch: 40  Training Loss:  0.355342960357666  Valid Loss:  0.15297851860523223\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "wmFzlXslDBBF",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 393
        },
        "outputId": "650d1caf-172b-4372-eb3d-50963865faf5"
      },
      "cell_type": "code",
      "source": [
        "ep = np.arange(1,41)\n",
        "plt.plot(ep, hl1_tr_loss)\n",
        "plt.plot(ep, hl2_tr_loss)\n",
        "plt.ylim(bottom=0, top=2)\n",
        "plt.title(\"Training Loss\")\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"MMSE loss\")\n",
        "plt.legend([\"Hidden layer 1024\", \"Hidden Layer 256\"])"
      ],
      "execution_count": 264,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f9767755dd8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 264
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFnCAYAAAC/5tBZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhHAQ/wPlEPiDTZKkRQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 576x396 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "R2CF8ngiIUSQ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 393
        },
        "outputId": "2498f7e3-0ef9-43ab-f8bf-3135561ddb5d"
      },
      "cell_type": "code",
      "source": [
        "ep = np.arange(1,41)\n",
        "plt.plot(ep, hl1_val_loss)\n",
        "plt.plot(ep, hl2_val_loss)\n",
        "plt.title(\"Validation Loss\")\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"MMSE loss\")\n",
        "plt.legend([\"Hidden layer 1024\", \"Hidden Layer 256\"])"
      ],
      "execution_count": 265,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f97675faf60>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 265
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQZZZRgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 576x396 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "0Nv9myz2G2Fc",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 153
        },
        "outputId": "2cff91cd-58dc-4d7b-a4f9-e47f768e05eb"
      },
      "cell_type": "code",
      "source": [
        "# Input Batch\n",
        "\n",
        "next(iter(train_dl))"
      ],
      "execution_count": 299,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tensor([[0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "        [3.5000, 3.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "        ...,\n",
              "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000]],\n",
              "       dtype=torch.float64)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 299
        }
      ]
    },
    {
      "metadata": {
        "id": "PTTWKPUtaMsG",
        "colab_type": "code",
        "outputId": "1debe24c-8336-43b2-ca56-db719201e58c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 153
        }
      },
      "cell_type": "code",
      "source": [
        "# Output of a batch after training\n",
        "out"
      ],
      "execution_count": 144,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tensor([[2.6717, 0.8065, 0.5438,  ..., 1.9969, 2.6463, 0.9386],\n",
              "        [3.2351, 3.0017, 2.7164,  ..., 0.3129, 0.6977, 0.8530],\n",
              "        [3.2825, 2.9849, 2.8165,  ..., 2.4957, 3.5273, 1.3070],\n",
              "        ...,\n",
              "        [4.1001, 3.3715, 2.9336,  ..., 1.4340, 1.3578, 1.9669],\n",
              "        [3.4013, 3.0175, 3.4592,  ..., 1.3811, 3.0438, 1.5754],\n",
              "        [3.9306, 4.1596, 3.6475,  ..., 0.2179, 1.2647, 0.7416]],\n",
              "       device='cuda:0', grad_fn=<AddmmBackward>)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 144
        }
      ]
    },
    {
      "metadata": {
        "id": "Vmd03RosHC7s",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "So, we can conclude that the **output is Dense**, whereas the **input given to the model is sparse**. Thus we have succeeded in developing state-of-the-art technique in order to predict user ratings."
      ]
    },
    {
      "metadata": {
        "id": "vMyCWuHrIERB",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}