MNIST_classisfication_CNN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MNIST_classisfication_CNN.ipynb",
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "7219de9d04b14e8ea232f4a4a80d362a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_155679261ed54ab5a5e9926126b6ae22",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_022d1850af8a43ce9033be750ad1a622",
              "IPY_MODEL_3642c741fcc249878e2bccf32006a72d"
            ]
          }
        },
        "155679261ed54ab5a5e9926126b6ae22": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "022d1850af8a43ce9033be750ad1a622": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_821dd19bc2144731aaa9d1547fffd427",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 50,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 50,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_430660b122504b6c8cc73316be2cfaab"
          }
        },
        "3642c741fcc249878e2bccf32006a72d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_8f31677d7f11402cb0f0801bee4f5cac",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 50/50 [11:50&lt;00:00, 14.21s/epochs]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_99522bffb8db43288a053536a6d2ba27"
          }
        },
        "821dd19bc2144731aaa9d1547fffd427": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "430660b122504b6c8cc73316be2cfaab": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "8f31677d7f11402cb0f0801bee4f5cac": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "99522bffb8db43288a053536a6d2ba27": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gJp-ZtlraarW"
      },
      "source": [
        "#Below, we have used a Le-net architecture CNN Network on MNIST Dataset."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QxlMJyJbaopt"
      },
      "source": [
        "Importing the required Libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uhHr9lrUGbfb"
      },
      "source": [
        "import numpy as np\r\n",
        "import torch\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "import torch.nn as nn\r\n",
        "import torch.nn.functional as F\r\n",
        "import torch.optim as opt\r\n",
        "from tqdm.notebook import tqdm_notebook\r\n",
        "from sklearn.model_selection import train_test_split"
      ],
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YzTHBhC0asCv"
      },
      "source": [
        "Importing data."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iPWyUBaLGevd"
      },
      "source": [
        "import torchvision\r\n",
        "import torchvision.transforms as transforms\r\n",
        "trainset=torchvision.datasets.MNIST(root='./data',train=True,transform=transforms.ToTensor(),download=True)\r\n",
        "testset=torchvision.datasets.MNIST(root='./data',train=False,transform=transforms.ToTensor(),download=True)"
      ],
      "execution_count": 59,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QXE6--4gGmBY"
      },
      "source": [
        "classes_tr=trainset.classes\r\n",
        "train=trainset\r\n",
        "trainloader = torch.utils.data.DataLoader(trainset, batch_size=128, shuffle=True)\r\n",
        "classes_test=testset.classes\r\n",
        "test=testset\r\n",
        "testloader = torch.utils.data.DataLoader(testset, batch_size=128, shuffle=True)"
      ],
      "execution_count": 61,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lGe8TXshazaU"
      },
      "source": [
        "Visualising the dataset, by visualising images."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5vSzyyshG_VH",
        "outputId": "14daf894-17b3-4d76-ac48-75b9f4a4742f"
      },
      "source": [
        "dataiter=iter(trainloader)\r\n",
        "\r\n",
        "images, labels= dataiter.next()\r\n",
        "print(images.shape,labels.shape)"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "torch.Size([128, 1, 28, 28]) torch.Size([128])\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MeSinEUiHKO-"
      },
      "source": [
        "npimg=images.detach().numpy()\r\n",
        "npimg2=np.transpose(npimg[3],(1,2,0))\r\n",
        "npimg2=npimg2.reshape(28,28)                #When printing a white_black image, we have a 2d array rather a 3d. But, for a RBG, we use 3d array."
      ],
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "oYzNRFAXHs3M",
        "outputId": "9f040b61-95aa-4f68-8a5a-7ad931e85a97"
      },
      "source": [
        "plt.imshow(npimg2)"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fea06bfd7b8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 50
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8QWzCbIQJjpF"
      },
      "source": [
        "#Convolutional Neural Network.\r\n",
        "Below, we have have implemented a Le-net Architecture.\r\n",
        "But, as **MNIST Dataset** is a black-white dataset, I had made some changes accordingly.\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pOZr5MsBIdbw"
      },
      "source": [
        "class cov_net(nn.Module):\r\n",
        "  def __init__(self):\r\n",
        "    super(cov_net,self).__init__()\r\n",
        "    self.cnn_net=nn.Sequential(\r\n",
        "        nn.Conv2d(1,6,5),                           #(N,1,28,28)-->(N,6,24,24)      \r\n",
        "        nn.Tanh(),\r\n",
        "        nn.MaxPool2d(2,stride=2),                   #(N,6,24,24)-->(N,6,12,12)              \r\n",
        "        nn.Conv2d(6,16,5),                          #(N,6,12,12)-->(N,16,8,8)\r\n",
        "        nn.Tanh(),\r\n",
        "        nn.MaxPool2d(2,2)                           #(N,16,8,8)-->(N,16,4,4)-->(after flattening)-->(N,256)\r\n",
        "    )\r\n",
        "    self.fcc_net=nn.Sequential(\r\n",
        "        nn.Linear(256,120),\r\n",
        "        nn.Tanh(),\r\n",
        "        nn.Linear(120,84),\r\n",
        "        nn.Tanh(),\r\n",
        "        nn.Linear(84,10),\r\n",
        "        nn.Tanh()\r\n",
        "    )\r\n",
        "\r\n",
        "  def feed_forward(self,x):\r\n",
        "    x = self.cnn_net(x)\r\n",
        "    x=x.view(x.size(0),-1)                         #preserve the first shape output, and colapse the rest into a single column vector.\r\n",
        "    x = self.fcc_net(x)\r\n",
        "\r\n",
        "    return x"
      ],
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vKhHNiDaOJRq"
      },
      "source": [
        "def evaluation_gpu(dataloader):\r\n",
        "  total=0\r\n",
        "  correct=0\r\n",
        "  for data in dataloader:\r\n",
        "    inputs,labels=data\r\n",
        "    inputs,labels=inputs.to(device), labels.to(device)\r\n",
        "    output=net.feed_forward(inputs)\r\n",
        "    a , pred=torch.max(output.data,1)\r\n",
        "    total+=labels.size(0)\r\n",
        "    correct+=(pred==labels).sum().item()\r\n",
        "\r\n",
        "  return (correct/total)*100"
      ],
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fx0K1WTEOxmK",
        "outputId": "0f726892-e29e-4db9-e70e-58e03feb77e7"
      },
      "source": [
        "%%time\r\n",
        "def fit_gpu(trainloader,epochs,model,op,loss_fn):\r\n",
        "  loss_tr=[]\r\n",
        "  loss_val=[]\r\n",
        "  acc=[]\r\n",
        "  for i in tqdm_notebook(range(epochs), total=epochs, unit='epochs'):\r\n",
        "    for i,data in enumerate(trainloader,0):\r\n",
        "      inputs, labels=data\r\n",
        "      inputs , labels= inputs.to(device) , labels.to(device)\r\n",
        "\r\n",
        "      inpt_train,inpt_val,lab_train,lab_val = train_test_split(inputs,labels)\r\n",
        "      op.zero_grad()\r\n",
        "      \r\n",
        "      out_tr=net.feed_forward(inpt_train)\r\n",
        "      \r\n",
        "\r\n",
        "      l=loss_fn(out_tr,lab_train)\r\n",
        "      l.backward()\r\n",
        "\r\n",
        "      op.step()\r\n",
        "\r\n",
        "    loss_tr.append(l.item())\r\n",
        "    \r\n",
        "    out_val=net.feed_forward(inpt_val)\r\n",
        "    l_val=loss_fn(out_val,lab_val)\r\n",
        "    loss_val.append(l_val)\r\n",
        "\r\n",
        "    acc.append(evaluation_gpu(trainloader))\r\n",
        "  \r\n",
        "  fig,ax = plt.subplots()\r\n",
        "  color='tab:red'\r\n",
        "  ax.set_ylabel('loss',color=color)\r\n",
        "  ax.set_xlabel('epochs')\r\n",
        "  ax.plot(loss_tr,color=color)\r\n",
        "  ax.plot(loss_val,color='b')\r\n",
        "  plt.show()\r\n",
        "\r\n",
        "  return acc\r\n",
        "    "
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "CPU times: user 4 µs, sys: 0 ns, total: 4 µs\n",
            "Wall time: 6.2 µs\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "10gGmV0AOzHU"
      },
      "source": [
        "device=torch.device('cuda:0')\r\n",
        "net=cov_net().to(device)\r\n",
        "loss_fn=nn.CrossEntropyLoss()\r\n",
        "op=opt.Adam(net.parameters(),lr=0.00025)\r\n"
      ],
      "execution_count": 149,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 362,
          "referenced_widgets": [
            "7219de9d04b14e8ea232f4a4a80d362a",
            "155679261ed54ab5a5e9926126b6ae22",
            "022d1850af8a43ce9033be750ad1a622",
            "3642c741fcc249878e2bccf32006a72d",
            "821dd19bc2144731aaa9d1547fffd427",
            "430660b122504b6c8cc73316be2cfaab",
            "8f31677d7f11402cb0f0801bee4f5cac",
            "99522bffb8db43288a053536a6d2ba27"
          ]
        },
        "id": "V1rCiZtAO6Sc",
        "outputId": "be2ed983-ff4e-4a79-bb41-362bc17cf1c1"
      },
      "source": [
        "%%time\r\n",
        "acc=fit_gpu(trainloader,50,net,op,loss_fn)"
      ],
      "execution_count": 150,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "7219de9d04b14e8ea232f4a4a80d362a",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, max=50.0), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "CPU times: user 8min 5s, sys: 3.14 s, total: 8min 8s\n",
            "Wall time: 8min 9s\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "g53w35gNgzaS"
      },
      "source": [
        "The red is the **training loss**, while the blue is the **validation loss**."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UQcRZVy5PHJJ",
        "outputId": "9b6a2eb6-ffd8-4392-a06f-305231b35829"
      },
      "source": [
        "print(('Train accuracy is : {:.2f}').format(evaluation_gpu(trainloader)))\r\n",
        "print(('Test accuracy is :{:.2f}').format(evaluation_gpu(testloader)))"
      ],
      "execution_count": 151,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train accuracy is : 99.78\n",
            "Test accuracy is :99.00\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OVGaCTn2bzNL"
      },
      "source": [
        "# Visualisation of the output images."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k0rOtlAWb3k_"
      },
      "source": [
        "net=net.to('cpu')\r\n",
        "out=net.feed_forward(images)"
      ],
      "execution_count": 152,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m501EmPsefsg",
        "outputId": "d58d1cfa-a654-4006-faaa-7041cb234204"
      },
      "source": [
        "net.cnn_net"
      ],
      "execution_count": 153,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Sequential(\n",
              "  (0): Conv2d(1, 6, kernel_size=(5, 5), stride=(1, 1))\n",
              "  (1): Tanh()\n",
              "  (2): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "  (3): Conv2d(6, 16, kernel_size=(5, 5), stride=(1, 1))\n",
              "  (4): Tanh()\n",
              "  (5): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              ")"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 153
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LfBEvAhbb6wW",
        "outputId": "a693d8b3-9660-4fc1-d45a-6c0f1cecefb3"
      },
      "source": [
        "out=net.cnn_net[1]((net.cnn_net[0](images)))\r\n",
        "out.shape"
      ],
      "execution_count": 154,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([128, 6, 24, 24])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 154
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 176
        },
        "id": "cRlyfC3wcGyM",
        "outputId": "df81f359-6456-4067-c31e-c47776793372"
      },
      "source": [
        "image_id = 3\r\n",
        "plt.figure(figsize = (2,2))\r\n",
        "plt.imshow(images[image_id,].reshape(28,28))"
      ],
      "execution_count": 155,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fea06dadf60>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 155
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 144x144 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "id": "QJf0qi7yc29Q",
        "outputId": "bdadd80f-93c7-4546-d214-02a144315e06"
      },
      "source": [
        "plt.figure(figsize = (6,6))\r\n",
        "plt.subplot(321)\r\n",
        "for i in range(6):\r\n",
        "    ax1 = plt.subplot(3, 2, i+1,)\r\n",
        "    plt.imshow(out[image_id, i, :, :].detach().numpy(), cmap=\"binary\")                #This is another way of pritning grey-scale images.\r\n",
        "plt.show()"
      ],
      "execution_count": 156,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:4: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
            "  after removing the cwd from sys.path.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x432 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OhxpgCEZfq0o"
      },
      "source": [
        "The above outputs are pretty amazing , right? \r\n",
        "\r\n",
        "\r\n",
        "Each i_th plot is the i_th channel output of the first Convolutional layer, after applying the activation function. **(i.e. Tanh())**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GXd8jvjXdIiu",
        "outputId": "fb3a014f-88c3-44be-98be-e63535abcfbf"
      },
      "source": [
        "out=net.cnn_net[3](net.cnn_net[0](images))\r\n",
        "out.shape"
      ],
      "execution_count": 157,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([128, 16, 20, 20])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 157
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 376
        },
        "id": "X_PdMYD1fdn7",
        "outputId": "f9292bda-0221-4a5d-a157-97fb9a30a87f"
      },
      "source": [
        "plt.figure(figsize = (6,6))\r\n",
        "plt.subplot(321)\r\n",
        "for i in range(16):\r\n",
        "    ax1 = plt.subplot(4, 4, i+1,)\r\n",
        "    plt.imshow(out[image_id, i, :, :].detach().numpy(), cmap=\"binary\")                #This is another way of pritning grey-scale images.\r\n",
        "plt.show()"
      ],
      "execution_count": 158,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x432 with 16 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vBwFhN0VgcNj"
      },
      "source": [
        "Again, These above plots are ouputs of the 2nd convolutional layer.\r\n"
      ]
    }
  ]
}