S7_Assignment_v1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "S7_Assignment_v1.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyPDpJJX1ZWdYUNXNzNHcZac",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/thamizhannal/eva5/blob/master/S7_Assignment_v1.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "27Vv2nbW1wD2",
        "colab_type": "text"
      },
      "source": [
        "### Import Pytorch Libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Y6prYBZ60vAH",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import torch\n",
        "import torchvision\n",
        "import torchvision.transforms as transforms"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DFyAY4LX10yV",
        "colab_type": "text"
      },
      "source": [
        "### Import EVA5-Jedi reporitory\n",
        "#### https://test.pypi.org/project/tsai-jedi/0.0.1/#files"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6JOs4j4U1vPW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "eb9cf3cc-a9b5-4358-b8e1-830c2c2bb4da"
      },
      "source": [
        "pip install -i https://test.pypi.org/simple/ eva5-jedi==0.0.1"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Looking in indexes: https://test.pypi.org/simple/\n",
            "Requirement already satisfied: eva5-jedi==0.0.1 in /usr/local/lib/python3.6/dist-packages (0.0.1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xBVz_4Jz2Exn",
        "colab_type": "text"
      },
      "source": [
        "#### Import Necessay packages for Model, BatchNormm Train, Test & DataLoader etc"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_lq-OVlD1CW9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from com.tsai.jedi.batchnorm  import BatchNorm\n",
        "from com.tsai.jedi.batchnorm  import GhostBatchNorm\n",
        "from com.tsai.jedi.model import Net\n",
        "from com.tsai.jedi.model import model_summary\n",
        "from com.tsai.jedi.Engine_train_test import train\n",
        "from com.tsai.jedi.Engine_train_test import test\n",
        "from com.tsai.jedi.dataloader import train_loader, test_loader\n",
        "from com.tsai.jedi import config"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "c37iJJbv2RKJ",
        "colab_type": "text"
      },
      "source": [
        "#### Import scheduler"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-rAln_gb06ek",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "61cabb60-053d-45ac-ef08-4750ccde0233"
      },
      "source": [
        "from __future__ import print_function\n",
        "import torch.optim as optim\n",
        "from torch.optim.lr_scheduler import StepLR\n",
        "import sys\n",
        "\n",
        "\n",
        "print(model_summary(Net(1), config.input_size))\n",
        "\n",
        "accu = []\n",
        "loss_test = []\n",
        "\n",
        "\n",
        "model_ = Net(BN_flag=1).to(config.device)\n",
        "optimizer = optim.SGD(model_.parameters(), lr=0.016, momentum=0.9)\n",
        "scheduler = StepLR(optimizer, step_size=2, gamma=0.92)\n",
        "\n",
        "valid_acc = []\n",
        "loss_test_ = []\n",
        "l1_regularization = [0, 0]\n",
        "for epoch in range(1, 31):\n",
        "    train(model_, config.device, train_loader, optimizer, epoch, l1_regularization)\n",
        "    scheduler.step()\n",
        "    valid_a, valid_l = test(model_, config.device, test_loader)\n",
        "    # Appending to loss and accuracy lists\n",
        "    valid_acc.append(valid_a)\n",
        "    loss_test_.append(valid_l)\n",
        "\n",
        "accu.append(valid_acc)\n",
        "loss_test.append(loss_test_)\n",
        "\n",
        "import pickle\n",
        "with open(\"model_op_params.pickle\",\"wb\") as f:\n",
        "    pickle.dump(accu, f)\n",
        "    pickle.dump(loss_test, f)"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/com/tsai/jedi/model.py:149: UserWarning: Implicit dimension choice for log_softmax has been deprecated. Change the call to include dim=X as an argument.\n",
            "  return F.log_softmax(x)\n",
            "\r  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "----------------------------------------------------------------\n",
            "        Layer (type)               Output Shape         Param #\n",
            "================================================================\n",
            "            Conv2d-1           [-1, 32, 32, 32]             896\n",
            "    GhostBatchNorm-2           [-1, 32, 32, 32]              64\n",
            "            Conv2d-3           [-1, 32, 32, 32]           9,248\n",
            "    GhostBatchNorm-4           [-1, 32, 32, 32]              64\n",
            "           Dropout-5           [-1, 32, 32, 32]               0\n",
            "         MaxPool2d-6           [-1, 32, 16, 16]               0\n",
            "            Conv2d-7           [-1, 16, 18, 18]             528\n",
            "    GhostBatchNorm-8           [-1, 16, 18, 18]              32\n",
            "           Dropout-9           [-1, 16, 18, 18]               0\n",
            "           Conv2d-10           [-1, 16, 18, 18]             160\n",
            "           Conv2d-11           [-1, 64, 18, 18]           1,088\n",
            "depthwise_separable_conv-12           [-1, 64, 18, 18]               0\n",
            "   GhostBatchNorm-13           [-1, 64, 18, 18]             128\n",
            "          Dropout-14           [-1, 64, 18, 18]               0\n",
            "        MaxPool2d-15             [-1, 64, 9, 9]               0\n",
            "           Conv2d-16           [-1, 32, 11, 11]           2,080\n",
            "   GhostBatchNorm-17           [-1, 32, 11, 11]              64\n",
            "          Dropout-18           [-1, 32, 11, 11]               0\n",
            "           Conv2d-19          [-1, 128, 11, 11]          36,992\n",
            "   GhostBatchNorm-20          [-1, 128, 11, 11]             256\n",
            "          Dropout-21          [-1, 128, 11, 11]               0\n",
            "           Conv2d-22            [-1, 128, 9, 9]         147,584\n",
            "   GhostBatchNorm-23            [-1, 128, 9, 9]             256\n",
            "          Dropout-24            [-1, 128, 9, 9]               0\n",
            "        MaxPool2d-25            [-1, 128, 4, 4]               0\n",
            "           Conv2d-26            [-1, 256, 4, 4]         295,168\n",
            "           Conv2d-27             [-1, 64, 4, 4]         147,520\n",
            "        AvgPool2d-28             [-1, 64, 1, 1]               0\n",
            "           Conv2d-29             [-1, 10, 1, 1]             650\n",
            "================================================================\n",
            "Total params: 642,778\n",
            "Trainable params: 641,914\n",
            "Non-trainable params: 864\n",
            "----------------------------------------------------------------\n",
            "Input size (MB): 0.01\n",
            "Forward/backward pass size (MB): 2.88\n",
            "Params size (MB): 2.45\n",
            "Estimated Total Size (MB): 5.34\n",
            "----------------------------------------------------------------\n",
            "None\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 1,LR: 0.016.\n",
            "Train set: train Average loss: 1.4768, train_Accuracy: 16381/50000 (32.7620%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 1.6392, Accuracy: 4132/10000 (41.3200%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 2,LR: 0.016.\n",
            "Train set: train Average loss: 1.2410, train_Accuracy: 25280/50000 (50.5600%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 1.3805, Accuracy: 4928/10000 (49.2800%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 3,LR: 0.01472.\n",
            "Train set: train Average loss: 1.1832, train_Accuracy: 28855/50000 (57.7100%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 1.0768, Accuracy: 6133/10000 (61.3300%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 4,LR: 0.01472.\n",
            "Train set: train Average loss: 1.0343, train_Accuracy: 30977/50000 (61.9540%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 1.0791, Accuracy: 6174/10000 (61.7400%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 5,LR: 0.013542400000000001.\n",
            "Train set: train Average loss: 0.6647, train_Accuracy: 32474/50000 (64.9480%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.9432, Accuracy: 6754/10000 (67.5400%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 6,LR: 0.013542400000000001.\n",
            "Train set: train Average loss: 0.8063, train_Accuracy: 33581/50000 (67.1620%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.8585, Accuracy: 7050/10000 (70.5000%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 7,LR: 0.012459008000000002.\n",
            "Train set: train Average loss: 0.9312, train_Accuracy: 34561/50000 (69.1220%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.7734, Accuracy: 7311/10000 (73.1100%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 8,LR: 0.012459008000000002.\n",
            "Train set: train Average loss: 0.7448, train_Accuracy: 35345/50000 (70.6900%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.8649, Accuracy: 7007/10000 (70.0700%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 9,LR: 0.011462287360000002.\n",
            "Train set: train Average loss: 0.9555, train_Accuracy: 36187/50000 (72.3740%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.7802, Accuracy: 7330/10000 (73.3000%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 10,LR: 0.011462287360000002.\n",
            "Train set: train Average loss: 0.8755, train_Accuracy: 36717/50000 (73.4340%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.7541, Accuracy: 7440/10000 (74.4000%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 11,LR: 0.010545304371200001.\n",
            "Train set: train Average loss: 0.6386, train_Accuracy: 37322/50000 (74.6440%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.7406, Accuracy: 7507/10000 (75.0700%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 12,LR: 0.010545304371200001.\n",
            "Train set: train Average loss: 0.5609, train_Accuracy: 37634/50000 (75.2680%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.6807, Accuracy: 7715/10000 (77.1500%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 13,LR: 0.009701680021504002.\n",
            "Train set: train Average loss: 0.6824, train_Accuracy: 38087/50000 (76.1740%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.6781, Accuracy: 7739/10000 (77.3900%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 14,LR: 0.009701680021504002.\n",
            "Train set: train Average loss: 0.6505, train_Accuracy: 38308/50000 (76.6160%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.6528, Accuracy: 7840/10000 (78.4000%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 15,LR: 0.008925545619783683.\n",
            "Train set: train Average loss: 0.7042, train_Accuracy: 38646/50000 (77.2920%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.6416, Accuracy: 7897/10000 (78.9700%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 16,LR: 0.008925545619783683.\n",
            "Train set: train Average loss: 0.5693, train_Accuracy: 38891/50000 (77.7820%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.6545, Accuracy: 7847/10000 (78.4700%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 17,LR: 0.008211501970200989.\n",
            "Train set: train Average loss: 0.5097, train_Accuracy: 39256/50000 (78.5120%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.6701, Accuracy: 7838/10000 (78.3800%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 18,LR: 0.008211501970200989.\n",
            "Train set: train Average loss: 0.5020, train_Accuracy: 39299/50000 (78.5980%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.6026, Accuracy: 8074/10000 (80.7400%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 19,LR: 0.007554581812584909.\n",
            "Train set: train Average loss: 0.6203, train_Accuracy: 39484/50000 (78.9680%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.6049, Accuracy: 8017/10000 (80.1700%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 20,LR: 0.007554581812584909.\n",
            "Train set: train Average loss: 0.6132, train_Accuracy: 39932/50000 (79.8640%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.6003, Accuracy: 8014/10000 (80.1400%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 21,LR: 0.006950215267578117.\n",
            "Train set: train Average loss: 0.5388, train_Accuracy: 40124/50000 (80.2480%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.5830, Accuracy: 8102/10000 (81.0200%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 22,LR: 0.006950215267578117.\n",
            "Train set: train Average loss: 0.8209, train_Accuracy: 40212/50000 (80.4240%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.6340, Accuracy: 7982/10000 (79.8200%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 23,LR: 0.006394198046171868.\n",
            "Train set: train Average loss: 0.7129, train_Accuracy: 40434/50000 (80.8680%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.5795, Accuracy: 8157/10000 (81.5700%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 24,LR: 0.006394198046171868.\n",
            "Train set: train Average loss: 0.5801, train_Accuracy: 40466/50000 (80.9320%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.5614, Accuracy: 8179/10000 (81.7900%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 25,LR: 0.005882662202478119.\n",
            "Train set: train Average loss: 0.5203, train_Accuracy: 40750/50000 (81.5000%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.5627, Accuracy: 8196/10000 (81.9600%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 26,LR: 0.005882662202478119.\n",
            "Train set: train Average loss: 0.5939, train_Accuracy: 40952/50000 (81.9040%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.5995, Accuracy: 8096/10000 (80.9600%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                       "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 27,LR: 0.005412049226279869.\n",
            "Train set: train Average loss: 0.4978, train_Accuracy: 40930/50000 (81.8600%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.5742, Accuracy: 8235/10000 (82.3500%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                        "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 28,LR: 0.005412049226279869.\n",
            "Train set: train Average loss: 0.4701, train_Accuracy: 41006/50000 (82.0120%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.5745, Accuracy: 8164/10000 (81.6400%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                        "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 29,LR: 0.00497908528817748.\n",
            "Train set: train Average loss: 0.3059, train_Accuracy: 41266/50000 (82.5320%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 0/196 [00:00<?, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.5604, Accuracy: 8225/10000 (82.2500%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "                                                                                        "
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 30,LR: 0.00497908528817748.\n",
            "Train set: train Average loss: 0.4463, train_Accuracy: 41338/50000 (82.6760%)\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\r"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Test set: Average loss: 0.5979, Accuracy: 8123/10000 (81.2300%)\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YoO45H1Q2dRh",
        "colab_type": "text"
      },
      "source": [
        "### Validation Loss Plot"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dnJ6NKS-1IIJ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "d8bbbe2d-5590-4660-d50e-8275e30df456"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "x = [i for i in range(1,31)]\n",
        "y = loss_test\n",
        "Modelgraph = ['GBN']\n",
        "\n",
        "plt.title(\"Validation loss graph\")\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Validation loss\")\n",
        "#plt.figure(figsize=(10,10))\n",
        "\n",
        "for i in range(len(y)):\n",
        "    plt.plot(x,y[i],label = 'id %s'%Modelgraph[i])\n",
        "\n",
        "plt.legend()\n",
        "#plt.show()\n",
        "plt.savefig('Validation loss graph')"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "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": "kblahg7f2muo",
        "colab_type": "text"
      },
      "source": [
        "### Validation Accuracy Plot"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "O_yD0tEh2piA",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "outputId": "2c2b7b96-be5a-4621-aba3-48702522d440"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "x = [i for i in range(1,31)]\n",
        "y = accu\n",
        "Modelgraph = ['GBN']\n",
        "\n",
        "\n",
        "plt.title(\"Validation Accuracy graph\")\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Validation Accuracy\")\n",
        "\n",
        "\n",
        "for i in range(len(y)):\n",
        "    plt.plot(x,y[i],label = 'id %s'%Modelgraph[i])\n",
        "\n",
        "plt.legend()\n",
        "plt.show()\n",
        "plt.savefig('Validation Accuracy graph_')"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "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": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}