FINAL_FASHION_MNIST_WITH_TRAINING_AND_VALIDATION_ACCURACY.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "FINAL_FASHION_MNIST_WITH_TRAINING_AND_VALIDATION_ACCURACY",
      "version": "0.3.2",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "metadata": {
        "id": "rZFoBK4AIAyi",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "1d138705-dfb3-41d1-c291-b8dc725c82a2"
      },
      "cell_type": "code",
      "source": [
        "# http://pytorch.org/\n",
        "from os.path import exists\n",
        "from wheel.pep425tags import get_abbr_impl, get_impl_ver, get_abi_tag\n",
        "platform = '{}{}-{}'.format(get_abbr_impl(), get_impl_ver(), get_abi_tag())\n",
        "cuda_output = !ldconfig -p|grep cudart.so|sed -e 's/.*\\.\\([0-9]*\\)\\.\\([0-9]*\\)$/cu\\1\\2/'\n",
        "accelerator = cuda_output[0] if exists('/dev/nvidia0') else 'cpu'\n",
        "\n",
        "!pip install -q http://download.pytorch.org/whl/{accelerator}/torch-0.4.1-{platform}-linux_x86_64.whl torchvision\n",
        "import torch"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "tcmalloc: large alloc 1073750016 bytes == 0x5881a000 @  0x7f4f5e3cc2a4 0x591a07 0x5b5d56 0x502e9a 0x506859 0x502209 0x502f3d 0x506859 0x504c28 0x502540 0x502f3d 0x506859 0x504c28 0x502540 0x502f3d 0x506859 0x504c28 0x502540 0x502f3d 0x507641 0x502209 0x502f3d 0x506859 0x504c28 0x502540 0x502f3d 0x507641 0x504c28 0x502540 0x502f3d 0x507641\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "HlflMzNEIJQK",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 122
        },
        "outputId": "a28c6b34-3bb6-447c-f0ec-d2d8ae642222"
      },
      "cell_type": "code",
      "source": [
        "import torch\n",
        "from torchvision import datasets, transforms\n",
        "# Define a transform to normalize the data\n",
        "transform = transforms.Compose([transforms.ToTensor(),\n",
        "                                transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])\n",
        "# Download and load the training data\n",
        "trainset = datasets.FashionMNIST('~/.pytorch/F_MNIST_data/', download=True, train=True, transform=transform)\n",
        "trainloader = torch.utils.data.DataLoader(trainset, batch_size=64, shuffle=True)\n",
        "\n",
        "# Download and load the test data\n",
        "testset = datasets.FashionMNIST('~/.pytorch/F_MNIST_data/', download=True, train=False, transform=transform)\n",
        "testloader = torch.utils.data.DataLoader(testset, batch_size=64, shuffle=True)"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz\n",
            "Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz\n",
            "Processing...\n",
            "Done!\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "t6sN5E4eILaU",
        "colab_type": "code",
        "outputId": "b8b9c9ba-61bf-419d-c95a-2c0e1a8ea998",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 210
        }
      },
      "cell_type": "code",
      "source": [
        "!wget -c https://raw.githubusercontent.com/udacity/deep-learning-v2-pytorch/master/intro-to-pytorch/helper.py"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2018-12-26 09:30:12--  https://raw.githubusercontent.com/udacity/deep-learning-v2-pytorch/master/intro-to-pytorch/helper.py\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 151.101.0.133, 151.101.64.133, 151.101.128.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|151.101.0.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 2813 (2.7K) [text/plain]\n",
            "Saving to: ‘helper.py’\n",
            "\n",
            "\rhelper.py             0%[                    ]       0  --.-KB/s               \rhelper.py           100%[===================>]   2.75K  --.-KB/s    in 0s      \n",
            "\n",
            "2018-12-26 09:30:13 (59.8 MB/s) - ‘helper.py’ saved [2813/2813]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "a5FFCtGfIPZS",
        "colab_type": "code",
        "outputId": "9de75b82-f655-4b03-cc29-3828ca923472",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 330
        }
      },
      "cell_type": "code",
      "source": [
        "import helper\n",
        "images, labels = next(iter(trainloader))\n",
        "helper.imshow(images[0,:]);"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAATkAAAE5CAYAAADr4VfxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAADYZJREFUeJzt3U2MXedZB/Az399jT+x43MZ1auN+\nEBo5lYJUJFMJL0pXVSVECWpVqaqEwqLqAoIKqlAXVUEEFqULIqSqEmrFAkVFYgMLDCFeIpJZUJzU\n343tOPH3TOZ77mXRTe6dc06fM74z43n6++3ue955z3nvvfOfc++Z57x97Xa7XQAk1b/bBwCwnYQc\nkJqQA1ITckBqQg5ITcgBqQ3WbTx96mRp+/f/4eXiq1/+nW05oN1iTntHxnmZ08M5c3auctuWzuSO\nHT+x5YN5VJnT3pFxXua0fXxcBVITckBqQg5Ira+udvXSxfOPzOdqgK2oDbmqq6tnzs5VbturzGnv\nyDgvc3r4fVXxcRVITcgBqQk5IDUhB6Qm5IDUhByQmpADUhNyQGpCDkhNyAGpCTkgNSEHpCbkgNSE\nHJCakANSE3JAakIOSE3IAakJOSA1IQekJuSA1IQckJqQA1ITckBqQg5ITcgBqQk5IDUhB6Qm5IDU\nhByQmpADUhNyQGpCDkhNyAGpCTkgNSEHpCbkgNSEHJCakANSE3JAakIOSE3IAakJOSA1IQekJuSA\n1IQckJqQA1ITckBqQg5ITcgBqQk5IDUhB6TW126321UbL108Xxw7fmInjwegp2pD7vSpk6XtZ87O\nVW7bq8xp78g4L3N6+H1V8XEVSE3IAakJOSC1wd0+ANhJHzpyJNz3Yx/9WLjv9RvXw33feOON0vaB\n/s5zjo1WKzwm1ZzJAakJOSA1IQekJuSA1IQckJqQA1ITckBqQg5ITcgBqQk5IDVlXeyovr6+cN+a\nu4BtMjU5Ger36d/8dHjM9fX1cN+xsdFw308+88nS9ud+77mOx6/PvR4e839/8pNw3182zuSA1IQc\nkJqQA1ITckBqQg5ITcgBqQk5IDUhB6Qm5IDUVDywo5pUMTQxvW9fqN+t27fCYy4vLYX7jozEKx7u\n3b8Xan/m5DPhMY80WKDn1rux5+D2ndvhMa9cvVraXlbhsl3vgSrO5IDUhByQmpADUhNyQGpCDkhN\nyAGpCTkgNSEHpCbkgNSEHJCasi5SOPLEE6F+7Va8pKhJqVZff3yBnsdmZkLtBw8eDI/5wQ8cDve9\nfuPtUL8HD+6Hx/zVj3+8tP2zn/nMpra3b94Mjfna6/GFfOo4kwNSE3JAakIOSE3IAakJOSA1IQek\nJuSA1IQckJqQA1ITckBqyroeYWUrHVXZ6RWQdsLQ0FB420c/8pHQmPfula+UVWZxcTHcd3R4LNz3\nzt27ofbLl6+Ex/zwh58M992/P7ay2eTkRHjMqueqrDRtYiI27utzc+H913EmB6Qm5IDUhByQmpAD\nUutr13xjfeni+eLY8RM7eTwAPVUbcqdPnSxtP3N2rnLbXvUozulhr64+inNqourq6r/9x38Xv/1b\nz3a0PfeFL4TG3Larq2Pxq6tl/ujP/qL4m+/8aUfbg/sPwj/f5Orq9PR0qN/6+np4zLLn6it/+CfF\nD/7urza1Ly0thcZ8+cc/Du//31+tvsGmj6tAakIOSE3IAampeHiEZaxiaOLXn302vG0s+J3Y/Px8\neP9NvpNaDn7PVDfuQtexjU+Mh8d869q1cN/pB7Hv+vbv2x8e89Djh8LtN96+ERqzV+9/Z3JAakIO\nSE3IAakJOSA1IQekJuSA1IQckJqQA1ITckBqQg5IrWdlXdHbAm1XqdJu7387NLnVUnSM3Z7/zMxM\nuO8zJ6tvE9W9LfpcjY/HS6UWFt4L951oUIJVZf/+eBlVt8HB+K9y9Dm4ded2eMzVtdXS9us3rm9q\nW1tbC4/bC87kgNSEHJCakANSE3JAakIOSE3IAakJOSA1IQekJuSA1IQckFrPyrp2u1xot/cf1aRU\nqxdz2qnnZXZ2NtTv85/7XHjMmzdvhretrpaXFXVbeC9eqjU2GlsBrCiKYnllJdx3vaKsaWFhoePx\nxkYrPGYTrY2NUL8m+3/nnXfC7QcOHAiP2wvO5IDUhByQmpADUhNyQGpCDkhNyAGpCTkgNSEHpCbk\ngNR6VvFAzE4v5NPdPjYW/y/+J48eDfc9fvx4qN+5c+fCY9ZVEbx17VrH41Yr9t/5rQb/xX/37t1w\n31Yr/roODw+Vtt9/8KDjcZOKi8nJiXDfkZGRUL/V1fiCM/PzD0rbFxcXN7U9zII9W+FMDkhNyAGp\nCTkgNSEHpCbkgNSEHJCakANSE3JAakIOSE3IAakp6+qR6AI1Bw8eDI959MiHwn1HRstLdX7jU5/q\neDw8NBwec2Ag/jcwWgK1sBBfSKbdri7B6i4XipZ1jY6MhvffZNGhycnxeN+JydL2x2Ye63g8Ph4v\n6xoejr+u0UV/+vrj85+amg63bwQX0pmZmQnvv44zOSA1IQekJuSA1IQckFpfu+YGZ5cuni+OHT+x\nk8cD0FO1IXf61MnS9jNn5yq37VUPO6dH8erqt198qfjmC893tG3X1dWl5eVQv15cXf3uSz8qvv78\nFzvatuPqaqvm6m63JjcjLbu6+o1v/XXxl9/644623b66ura+Hh5zreQGm2VzKor4vF559dXw/l/+\nl/+s3ObjKpCakANSE3JAakIOSE1ZV41fCa5AVRRFsX9fbAWiqamp8JhLy0vxvkvlfbvbq/qVWW/w\nxXNfX+zv5dpa7EvvoiiK8bHqUqnuCyjR57WuVKzbaIOLCQP9A/G+FRd0RkY657RSs1pZt8XF+Ou6\nsRF7XVsNVpYbGixfgaxMdMW6Dxw+HB6zjjM5IDUhB6Qm5IDUhByQmpADUhNyQGpCDkhNyAGpCTkg\nNSEHpPZLV9Y1UnPfre5ts7Oz4XHv3bsX6tekVKvJalFVpTJNSoO6RVdVKoqiGByMvZUmJibCYw7V\njDk8HC8j6hizwf30mmhSLrayUl5WtbLSWfIWLb/6ed/4/qOv60Yr/vovV5QLzs8/2NQWLS1s8vtX\nx5kckJqQA1ITckBqQg5ITcgBqQk5IDUhB6Qm5IDUhByQWs8qHqanp0P9mixOEf0v+qKIL44xPl69\nOMrTTz/d8bjJoi9jo7FFT8pWuq+ytrZ5VfKmuisMhobilQJ1C8lsVZPKgLrFcSa6VqEfGup98U50\ncZ5e7X9sbLTjcZP3f6sVX3RmfT32vurvj89/vaKK4rEDBza1LczPh8bcF8yUX8SZHJCakANSE3JA\nakIOSE3IAakJOSA1IQekJuSA1IQckJqQA1KrrRvZv29feNtQzQIx73fv/v1Qv6JoVtYyOhIrlxoc\nrC5rWl3tLHdZWVkO7z9aArPw3kLPxyyK6gViusvdmiyOs7i0GO4bVbc4Tbe6179JeVh0zG7Dwfd0\nUTR7Xqt0l9w123+DEqz1+AI58THLS8XKfi9b0YV0GiykVMeZHJCakANSE3JAakIOSE3IAakJOSA1\nIQekJuSA1IQckJqQA1KrrXF56qmnwtui5Tp37t4N9SuKolheXgn3jbp1693wtr4GZVXRUpX+gYHw\nmE1KoKqeq9u373Q8blLW1N8fL1UaaDCvXoz5zrtdr1WwrKnJnJroxcpq/3fuXMfjgYEmJWjxVdii\nJWhNyr/6K57/y5evhMfoNjU1teWffT9nckBqQg5ITcgBqQk5ILW+dvcNx97n5tvXitnDT+zk8QD0\nVG3IffOF50vbv/3iS5u27ZWrq+9V3LTyh//0r8WXfvezHW176epq2dXF7//wn4uvfunzHW0Zrq5+\n96UfFV9//osdbXv96ur3/v4fi6/9we93tO31q6svfu8HxQtf+0p4jG5Nrq7++Xf+tnKbj6tAakIO\nSE3IAanVfuj/n9deC287PDsb2uHjjz8e6lcURXGoQd/x8fFQv5WV1cptn/i1T3Q83tiIfyfRqv5q\ns0N/gwVPerE4yZNHj3aO2eA7qSYLqbRbsfm3WvEFaOoWq2ny3tiqJt+J9cITH9z6Rb4mr+tAf+z7\n060uFvR+hw5tfp2afC/dC87kgNSEHJCakANSE3JAakIOSE3IAakJOSA1IQekJuSA1IQckFpt3cry\n8nJ42+UrsQUrov2ait5CaKDi9knfKIrilf96pbNvg9sSjY+NhfqNjI6Gx2xSAlblwsULHY+jt9kp\nip0va9q8/+q/wVeuXu3qGysV6m9y+6wGJWi9cOXq1n83WsGyup/3jd0WrBfe/OlPN7VF34OLi4s9\nOQZnckBqQg5ITcgBqQk5IDUhB6Qm5IDUhByQmpADUhNyQGpCDkhtd+t2eii62nddr5XVrpW8uh/X\n6FUJSq/97K23dvsQtsX5Cxd+cac95o0339ztQ+i5Cxcv7vYhOJMDchNyQGpCDkhNyAGpCTkgNSEH\npCbkgNSEHJCakANSE3JAakIOSE3IAakJOSA1IQekJuSA1IQckJqQA1ITckBqQg5ITcgBqQk5IDUh\nB6Qm5IDUhByQmpADUutrt9vtqo2XLp4vjh0/sZPHA9BTtSF3+tTJ0vYzZ+cqt+1V5rR3ZJyXOT38\nvqr4uAqkJuSA1IQckJqQA1ITckBqQg5ITcgBqQk5IDUhB6Qm5IDUhByQmpADUhNyQGpCDkhNyAGp\nCTkgNSEHpCbkgNSEHJCakANSE3JAakIOSE3IAakJOSA1IQek1tdut9u7fRAA28WZHJCakANSE3JA\nakIOSE3IAakJOSC1/wfIuCzYcOKXigAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd96ceed668>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "hses2CPXIQ0L",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from torch import nn\n",
        "import torch.nn.functional as F\n",
        "from torch import optim\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "luq5s2rjITql",
        "colab_type": "code",
        "outputId": "b969f0fe-bb3e-467e-cbe6-77a86b4bc2b7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        }
      },
      "cell_type": "code",
      "source": [
        "model = nn.Sequential(nn.Linear(784,256),nn.ReLU(),nn.Linear(256,10),nn.LogSoftmax(dim=1))\n",
        "\n",
        "criterion = nn.NLLLoss()\n",
        "\n",
        "optimizer = optim.SGD(model.parameters(),lr=0.003)\n",
        "\n",
        "epochs = 30\n",
        "steps = 0\n",
        "train_losses, test_losses = [], []\n",
        "\n",
        "\n",
        "for e in range(epochs):\n",
        "  running_loss = 0\n",
        "  for images , labels in trainloader:\n",
        "    optimizer.zero_grad()\n",
        "  \n",
        "    images = images.flatten(start_dim=1)\n",
        "  \n",
        "    output = model(images)\n",
        "  \n",
        "    loss = criterion(output,labels)\n",
        "  \n",
        "    loss.backward()\n",
        "  \n",
        "    optimizer.step()\n",
        "    \n",
        "    running_loss += loss.item()\n",
        "  else:\n",
        "    test_loss = 0\n",
        "    accuracy = 0\n",
        "    with torch.no_grad():\n",
        "      for images , labels in testloader:\n",
        "        images = images.flatten(start_dim=1)\n",
        "        log_ps = model(images)\n",
        "        test_loss += criterion(log_ps,labels)\n",
        "        \n",
        "        \n",
        "        ps = torch.exp(log_ps)\n",
        "        top_p , top_class = ps.topk(1,dim=1)\n",
        "        equals = top_class == labels.view(*top_class.shape)\n",
        "        accuracy += torch.mean(equals.type(torch.FloatTensor))\n",
        "    train_losses.append(running_loss/len(trainloader))\n",
        "    test_losses.append(test_loss/len(testloader))    \n",
        "    print(\"Epoch: {}/{}.. \".format(e+1, epochs),\"Training Loss: {:.3f}.. \".format(running_loss/len(trainloader)),\"Test Loss: {:.3f}.. \".format(test_loss/len(testloader)),\"Test Accuracy: {:.3f}\".format(accuracy/len(testloader)))\n"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch: 1/30..  Training Loss: 1.125..  Test Loss: 0.767..  Test Accuracy: 0.743\n",
            "Epoch: 2/30..  Training Loss: 0.677..  Test Loss: 0.643..  Test Accuracy: 0.778\n",
            "Epoch: 3/30..  Training Loss: 0.593..  Test Loss: 0.587..  Test Accuracy: 0.794\n",
            "Epoch: 4/30..  Training Loss: 0.548..  Test Loss: 0.555..  Test Accuracy: 0.803\n",
            "Epoch: 5/30..  Training Loss: 0.519..  Test Loss: 0.535..  Test Accuracy: 0.809\n",
            "Epoch: 6/30..  Training Loss: 0.499..  Test Loss: 0.518..  Test Accuracy: 0.814\n",
            "Epoch: 7/30..  Training Loss: 0.483..  Test Loss: 0.503..  Test Accuracy: 0.820\n",
            "Epoch: 8/30..  Training Loss: 0.470..  Test Loss: 0.495..  Test Accuracy: 0.824\n",
            "Epoch: 9/30..  Training Loss: 0.459..  Test Loss: 0.487..  Test Accuracy: 0.828\n",
            "Epoch: 10/30..  Training Loss: 0.450..  Test Loss: 0.479..  Test Accuracy: 0.828\n",
            "Epoch: 11/30..  Training Loss: 0.443..  Test Loss: 0.473..  Test Accuracy: 0.832\n",
            "Epoch: 12/30..  Training Loss: 0.436..  Test Loss: 0.465..  Test Accuracy: 0.835\n",
            "Epoch: 13/30..  Training Loss: 0.429..  Test Loss: 0.460..  Test Accuracy: 0.837\n",
            "Epoch: 14/30..  Training Loss: 0.424..  Test Loss: 0.457..  Test Accuracy: 0.839\n",
            "Epoch: 15/30..  Training Loss: 0.418..  Test Loss: 0.452..  Test Accuracy: 0.837\n",
            "Epoch: 16/30..  Training Loss: 0.413..  Test Loss: 0.446..  Test Accuracy: 0.843\n",
            "Epoch: 17/30..  Training Loss: 0.409..  Test Loss: 0.442..  Test Accuracy: 0.844\n",
            "Epoch: 18/30..  Training Loss: 0.404..  Test Loss: 0.439..  Test Accuracy: 0.842\n",
            "Epoch: 19/30..  Training Loss: 0.400..  Test Loss: 0.438..  Test Accuracy: 0.842\n",
            "Epoch: 20/30..  Training Loss: 0.396..  Test Loss: 0.432..  Test Accuracy: 0.847\n",
            "Epoch: 21/30..  Training Loss: 0.392..  Test Loss: 0.430..  Test Accuracy: 0.847\n",
            "Epoch: 22/30..  Training Loss: 0.389..  Test Loss: 0.425..  Test Accuracy: 0.849\n",
            "Epoch: 23/30..  Training Loss: 0.385..  Test Loss: 0.425..  Test Accuracy: 0.850\n",
            "Epoch: 24/30..  Training Loss: 0.382..  Test Loss: 0.422..  Test Accuracy: 0.850\n",
            "Epoch: 25/30..  Training Loss: 0.379..  Test Loss: 0.418..  Test Accuracy: 0.848\n",
            "Epoch: 26/30..  Training Loss: 0.376..  Test Loss: 0.416..  Test Accuracy: 0.851\n",
            "Epoch: 27/30..  Training Loss: 0.373..  Test Loss: 0.412..  Test Accuracy: 0.853\n",
            "Epoch: 28/30..  Training Loss: 0.371..  Test Loss: 0.418..  Test Accuracy: 0.850\n",
            "Epoch: 29/30..  Training Loss: 0.368..  Test Loss: 0.410..  Test Accuracy: 0.854\n",
            "Epoch: 30/30..  Training Loss: 0.365..  Test Loss: 0.407..  Test Accuracy: 0.855\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "JM33_3p58ZhP",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "%matplotlib inline\n",
        "%config InlineBackend.figure_format = 'retina'\n",
        "\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "b0ImkWz3-N2u",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "outputId": "baba193c-4dcc-4a72-ec66-1137db1b4dc4"
      },
      "cell_type": "code",
      "source": [
        "plt.plot(train_losses, label='Training loss')\n",
        "plt.plot(test_losses, label='Validation loss')\n",
        "plt.legend(frameon=False)"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fd911deb978>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAuUAAAHwCAYAAAAByRFLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIiLSYQrlIiIiIiIdplAuIiIiItJh/z+LZiCXqfrrjQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd911de1a90>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 370,
              "height": 248
            }
          }
        }
      ]
    }
  ]
}