ANN_Task_1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ANN Task-1.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyNUwplsdfajdtVslLBV3y03",
      "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/darshank528/Artificial-Neural-Network/blob/master/ANN_Task_1.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5gtOiGTJ4yMi",
        "colab_type": "code",
        "outputId": "359d03b8-978a-40c6-d9a5-72e06b9e1118",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 267
        }
      },
      "source": [
        "import torch\n",
        "import torchvision\n",
        "import torchvision.datasets as datasets\n",
        "from torch.utils import data\n",
        "import torchvision.transforms as transforms\n",
        "import torch.nn as nn\n",
        "import matplotlib.pyplot as plt\n",
        "import torch.optim as optim\n",
        "import time\n",
        "\n",
        "start_time=time.time()\n",
        "\n",
        "#device configuration \n",
        "device=torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "\n",
        "#loading and preparing data\n",
        "train_data = datasets.MNIST(r\"C:\\Users\\Darshan Khandelwal\\OneDrive\\Desktop\\MNIST\",download=True,transform=transforms.ToTensor(),train=True)\n",
        "test_data = datasets.MNIST(r\"C:\\Users\\Darshan Khandelwal\\OneDrive\\Desktop\\MNIST\",download=False,transform=transforms.ToTensor(),train=False)\n",
        "train_loader = data.DataLoader(dataset=train_data,batch_size=100,shuffle=True)\n",
        "test_loader = data.DataLoader(dataset=test_data,batch_size=100)\n",
        "\n",
        "#plotting some examples\n",
        "example = iter(train_loader)\n",
        "samples,labels = next(example)\n",
        "for i in range(6):\n",
        "  plt.subplot(2,3,i+1)\n",
        "  plt.imshow(samples[i,0])\n",
        "plt.show()\n",
        "\n"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1bjkYAr9_TR_",
        "colab_type": "code",
        "outputId": "c09d0284-8473-42f8-890c-3384788c8f1e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 463
        }
      },
      "source": [
        "#defining the proposed model\n",
        "class classifier(nn.Module):\n",
        "  def __init__(self):\n",
        "    super(classifier,self).__init__()\n",
        "    self.conv1 = nn.Conv2d(1,32,3)\n",
        "    self.conv2 = nn.Conv2d(32,64,3)\n",
        "    self.pool = nn.MaxPool2d(2,2)\n",
        "    self.drop1 = nn.Dropout(p=0.25)\n",
        "    self.fc1 = nn.Linear(12*12*64,128)\n",
        "    self.drop2 = nn.Dropout(p=0.5)\n",
        "    self.fc2 = nn.Linear(128,10)\n",
        "\n",
        "  def forward(self,X):\n",
        "    X = torch.relu(self.conv1(X))\n",
        "    X = torch.relu(self.conv2(X))\n",
        "    X = self.pool(X)\n",
        "    X = X.view(-1,12*12*64)\n",
        "    X = self.drop1(X)\n",
        "    X = torch.relu(self.fc1(X))\n",
        "    X = self.drop2(X)\n",
        "    X = self.fc2(X)\n",
        "    return X\n",
        "\n",
        "#initializing model\n",
        "model = classifier().to(device)\n",
        "\n",
        "#hyperparameters\n",
        "num_epochs=20\n",
        "learn_rate=0.001\n",
        "num_iterations=len(train_loader)\n",
        "batch_size=100\n",
        "\n",
        "#defining the loss criteria to be used\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "\n",
        "#defining the optimizer\n",
        "optimizer = optim.Adam(model.parameters(),lr=learn_rate)\n",
        "\n",
        "train_loss=[]\n",
        "#training our model\n",
        "for e in range(num_epochs):\n",
        "  tr_loss=0\n",
        "  model.train()\n",
        "\n",
        "  for i,(samples,labels) in enumerate(train_loader):\n",
        "    samples = samples.to(device)\n",
        "    labels = labels.to(device)\n",
        "    #model predictions\n",
        "    y_predicted = model(samples)\n",
        "    #cost\n",
        "    cost = criterion(y_predicted,labels)\n",
        "    tr_loss+=cost.item()\n",
        "    #calculate gradients\n",
        "    cost.backward()\n",
        "    #update parameters\n",
        "    optimizer.step()\n",
        "    #avoid accumulation of gradients\n",
        "    optimizer.zero_grad()\n",
        "  #printing loss per 2 epochs\n",
        "  if (e+1)%2 == 0:\n",
        "    print(\"epoch:\",e+1,\"/\",num_epochs,\" loss:{0:.4f}\".format(tr_loss))\n",
        "  train_loss.append(tr_loss) \n",
        "\n",
        "#plotting loss per epoch \n",
        "plt.plot(train_loss,color=\"b\")\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Train_loss\")\n",
        "plt.title(\"Loss per epoch\")\n",
        "plt.show()\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "epoch: 2 / 20  loss:56.8218\n",
            "epoch: 4 / 20  loss:35.2122\n",
            "epoch: 6 / 20  loss:28.0955\n",
            "epoch: 8 / 20  loss:21.5663\n",
            "epoch: 10 / 20  loss:17.3700\n",
            "epoch: 12 / 20  loss:15.3233\n",
            "epoch: 14 / 20  loss:13.1308\n",
            "epoch: 16 / 20  loss:12.1135\n",
            "epoch: 18 / 20  loss:10.5857\n",
            "epoch: 20 / 20  loss:9.8093\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEWCAYAAABi5jCmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nO3deZhcZZn38e8v+0IgCekEsnQCBOKwaiYgCIMIDvtlnFdQeEFQ0IyOG68MgssI4zIDLqCODIqCwIAssmYQGBZBeFWWsIVdQgykQ5YOSYAQCAm554/nlF3V6epUurvqVHf9Ptd1rjp1ljp3VXfX3c9znkURgZmZWUG/vAMwM7P64sRgZmYlnBjMzKyEE4OZmZVwYjAzsxJODGZmVsKJwawBSQpJU/OOw+qTE4P1GpIWSPpg3nGY9XVODGY1JKl/3jGYbYoTg/V6kgZL+pGkl7PlR5IGZ/vGSLpZ0ipJKyTdJ6lftu90SYskvS7pOUkHlXn9SyT9TNId2bG/lzS5aP+7sn0rstf5aLtzL5B0i6Q3gA908PpbSbpI0uIsnu8UEoikT0j6g6SfSnpV0rPFcUoaL2l2du15kj5dtK+/pK9JeiGL+2FJk4ou/UFJz2efzfmS1PWfgvUlTgzWF3wd2Bt4N7AHsBfwjWzfqUAL0ASMA74GhKRpwOeBPSNiBHAIsKCTaxwHfBsYAzwGXAEgaThwB/BrYCxwDPCfknYuOvf/At8FRgD/v4PXvgRYD0wF3gMcDHyqaP97gReya58JXC9pdLbvquz9jQeOAv5N0oHZvi8DxwKHA1sCJwFril73SGBPYHfgo9lnYObEYH3CccC3ImJZRLQC/wp8PNu3DtgWmBwR6yLivkgDhL0DDAZ2ljQwIhZExAudXOO3EXFvRKwlJaJ9sv++jwQWRMSvImJ9RDwKXAccXXTuTRHxh4jYEBFvFb+opHGkL+5TIuKNiFgGnEdKMAXLgB9l8V8NPAcckV1/X+D0iHgrIh4DfgmckJ33KeAbEfFcJI9HxCtFr3t2RKyKiJeAu0mJ1cyJwfqE8cCLRc9fzLYBfB+YB9wuab6kMwAiYh5wCnAWsEzSVZLGU97CwkpErAZWZNeYDLw3q45ZJWkVKVFt09G5HZgMDAQWF53/c1Lpo2BRlI52WXh/44EVEfF6u30TsvVJpJJGOUuK1tcAW3RyrDUQJwbrC14mfcEWNGfbiIjXI+LUiNge+BDw5UIdfUT8OiL2y84N4JxOrvHXunlJWwCjs2ssBH4fESOLli0i4rNF53Y2hPFCYC0wpuj8LSNil6JjJrSr/y+8v5eB0ZJGtNu3qOi1d+jk2mYdcmKw3magpCFFywDgSuAbkpokjQG+CVwOIOlISVOzL9ZXSVVIGyRNk3RgdpP6LeBNYEMn1z1c0n6SBpHuNdwfEQuBm4GdJH1c0sBs2VPS31TyZiJiMXA78ENJW0rqJ2kHSe8vOmws8MXstY8G/ga4Jbv+H4F/zz6L3YGTC++dVK30bUk7Ktld0taVxGWNzYnBeptbSF/iheUs4DvAHGAu8ATwSLYNYEfgTmA18CfgPyPibtL9hbOB5aQqlbHAVzu57q9JN35XAH8LHA+pREK6WXwM6T/4JaSSx+DNeE8nAIOAp4GVwLWk+yIFD2TvYznpJvZRRfcKjgWmZNe+ATgzIu7M9p0LXENKPK8BFwFDNyMua1DyRD1mnZN0CdASEd/Y1LFVuPYngE9lVV5mNeESg5mZlXBiMDOzEq5KMjOzEi4xmJlZiQF5B9BdY8aMiSlTpuQdhplZr/Lwww8vj4imjvb1+sQwZcoU5syZk3cYZma9iqQXy+1zVZKZmZVwYjAzsxJODGZmVqKqiUHSxZKWSXqy3fYvZBOOPCXpe0Xbv5pNNvKcJI8Nb2aWg2rffL4E+ClwWWGDpA8AM4E9ImKtpLHZ9p1J483sQhpO+E5JO0XEO1WO0czMilS1xBAR95IGHSv2WdIEIWuzY5Zl22cCV0XE2oj4C2kM/b2qGZ+ZmW0sj3sMOwF/J+mBbO7cPbPtEyid0KSFtglHSkiaJWmOpDmtra1VDtfMrLHkkRgGkCY52Rs4Dbhmcychj4gLI2JGRMxoauqwf8Ym3X8/nH46vOOKKjOzEnkkhhbg+mwO2gdJk6OMIc06NanouIm0zUTV4x5/HL73PVi8uFpXMDPrnfJIDDcCHwCQtBNpgpLlwGzgGEmDJW1HmpjkwWoF0dycHl96qVpXMDPrnardXPVK0qxZ0yS1SDoZuBjYPmvCehVwYlZ6eIo029TTwG3A56rZImlyNkOwE4OZWamqNleNiGPL7Dq+zPHfJU1dWHWTskqrF8uOFmJm1pgatufziBEwapRLDGZm7TVsYoBUneTEYGZWqqETQ3Ozq5LMzNpr+MTgEoOZWamGTgyTJ8Orr6bFzMyShk4M7stgZrYxJwacGMzMijV0YnAnNzOzjTV0Yhg3DgYOdMskM7NiDZ0Y+vVLPaBdYjAza9PQiQHcyc3MrL2GTwzu5GZmVsqJoRlefhnWrcs7EjOz+uDE0AwbNqTkYGZmTgx/bbLq6iQzs6ThE4M7uZmZlWr4xFCYsMeJwcwsafjEMGwYNDW5KsnMrKDacz5fLGlZNr9z+32nSgpJY7LnkvQTSfMkzZU0vZqxFfPw22ZmbapdYrgEOLT9RkmTgIOB4q/jw4Ads2UWcEGVY/srJwYzszZVTQwRcS+wooNd5wFfAaJo20zgskjuB0ZK2raa8RVMnpyqkiI2fayZWV9X83sMkmYCiyLi8Xa7JgALi563ZNs6eo1ZkuZImtPa2trtmJqb4Y03YOXKbr+UmVmvV9PEIGkY8DXgm915nYi4MCJmRMSMpqambsflJqtmZm1qXWLYAdgOeFzSAmAi8IikbYBFwKSiYydm26rOndzMzNrUNDFExBMRMTYipkTEFFJ10fSIWALMBk7IWiftDbwaEYtrEZdLDGZmbardXPVK4E/ANEktkk7u5PBbgPnAPOAXwD9VM7ZiTU0wZIgTg5kZwIBqvnhEHLuJ/VOK1gP4XDXjKUfy8NtmZgUN3/O5wH0ZzMwSJ4aME4OZWeLEkJk8GRYvhrVr847EzCxfTgyZQsuklpZ84zAzy5sTQ8ZNVs3MEieGjDu5mZklTgyZiRPTo0sMZtbonBgygwfDNts4MZiZOTEUKQy/bWbWyJwYirgvg5mZE0OJQmLwhD1m1sicGIo0N8Nbb8Hy5XlHYmaWHyeGIm6yambmxFDCndzMzJwYSjgxmJk5MZQYPRqGD3dVkpk1NieGIoUJe1xiMLNGVu2pPS+WtEzSk0Xbvi/pWUlzJd0gaWTRvq9KmifpOUmHVDO2cpwYzKzRVbvEcAlwaLttdwC7RsTuwJ+BrwJI2hk4BtglO+c/JfWvcnwbce9nM2t0VU0MEXEvsKLdttsjYn329H4gG76OmcBVEbE2Iv4CzAP2qmZ8HWluhtZWePPNWl/ZzKw+5H2P4STg1mx9ArCwaF9Ltq2mCi2TFi7s/Dgzs74qt8Qg6evAeuCKLpw7S9IcSXNaW1t7NC53cjOzRpdLYpD0CeBI4LiIv45MtAiYVHTYxGzbRiLiwoiYEREzmpqaejQ292Uws0ZX88Qg6VDgK8CHImJN0a7ZwDGSBkvaDtgReLDW8U2YAP36OTGYWeMaUM0Xl3QlcAAwRlILcCapFdJg4A5JAPdHxGci4ilJ1wBPk6qYPhcR71Qzvo4MHAjjx7sqycwaV1UTQ0Qc28Hmizo5/rvAd6sXUWXcl8HMGlnerZLqkhODmTUyJ4YOTJ6cmqtu2JB3JGZmtefE0IHmZnj7bVi6NO9IzMxqz4mhA26yamaNzImhA+7kZmaNzImhAy4xmFkjc2LowFZbwZZbOjGYWWNyYijDw2+bWaNyYijDfRnMrFE5MZThxGBmjcqJoYzmZlixAlavzjsSM7PacmIoo9Bk1aUGM2s0TgxluMmqmTUqJ4YyConBLZPMrNE4MZQxfjz07+8Sg5k1HieGMvr3h4kTnRjMrPE4MXSiudlVSWbWeJwYOjF5sksMZtZ4qpoYJF0saZmkJ4u2jZZ0h6Tns8dR2XZJ+omkeZLmSppezdgq0dwMLS3wTs1nnjYzy0+1SwyXAIe223YGcFdE7AjclT0HOAzYMVtmARdUObZNam5OSeHll/OOxMysdipKDJK+JGnL7L/6iyQ9IungTZ0XEfcCK9ptnglcmq1fCny4aPtlkdwPjJS0bWVvozrcyc3MGlGlJYaTIuI14GBgFPBx4OwuXnNcRCzO1pcA47L1CcDCouNasm0bkTRL0hxJc1pbW7sYxqa5k5uZNaJKE4Oyx8OB/4qIp4q2dVlEBBBdOO/CiJgRETOampq6G0ZZ7uRmZo2o0sTwsKTbSYnhfySNADZ08ZpLC1VE2eOybPsiYFLRcROzbbnZYgsYPdolBjNrLJUmhpNJN4n3jIg1wEDgk1285mzgxGz9ROCmou0nZPcx9gZeLapyyo2H3zazRjOgwuP2AR6LiDckHQ9MB368qZMkXQkcAIyR1AKcSbo3cY2kk4EXgY9mh99CKpHMA9bQ9cTTo5qbYf78vKMwM6udShPDBcAekvYATgV+CVwGvL+zkyLi2DK7Durg2AA+V2E8NTN5MtxzT95RmJnVTqVVSeuzL+6ZwE8j4nxgRPXCqh/NzfDaa/Dqq3lHYmZWG5UmhtclfZXUTPW3kvqR7jP0eW6ZZGaNptLE8DFgLak/wxJSi6HvVy2qOuJObmbWaCpKDFkyuALYStKRwFsRcVlVI6sT7uRmZo2m0iExPgo8CBxNakX0gKSjqhlYvRg3DgYNclWSmTWOSlslfZ3Uh2EZgKQm4E7g2moFVi/69YNJk1xiMLPGUek9hn6FpJB5ZTPO7fXcyc3MGkmlJYbbJP0PcGX2/GOkDmkNobkZ7rwz7yjMzGqjosQQEadJ+giwb7bpwoi4oXph1Zfm5jQnw7p1MLAhGumaWSOrtMRARFwHXFfFWOrW5MkQAYsWwZQpeUdjZlZdnSYGSa/T8bDYIo1isWVVoqozxZ3cnBjMrK/rNDFEREXDXkgaFREreyak+uO+DGbWSHqqZdFdPfQ6dcmJwcwaSU8lhm7P5lbPhg6FpiZ3cjOzxtBTiWGzp+fsbdyXwcwaRcN0UuuuyZOdGMysMbgqqUKFEkP0+bKRmTW6ivsxSOoPjCs+JyIK/0NvNCNbX9PcDG+8AStWwNZb5x2NmVn1VDq66heApcAdwG+z5ebC/ohYsbkXlvT/JD0l6UlJV0oaImk7SQ9ImifpakmDNvd1q8XzMphZo6i0KulLwLSI2CUidsuW3bt6UUkTgC8CMyJiV6A/cAxwDnBeREwFVgInd/UaPc1NVs2sUVSaGBYCPT3r8QBgqKQBwDBgMXAgbUN5Xwp8uIev2WWe4tPMGkWl9xjmA/dI+i1pik8AIuLcrlw0IhZJ+gHwEvAmcDvwMLAqItZnh7UAEzo6X9IsYBZAc+Ebu8qammDIEJcYzKzvq7TE8BLp/sIgYETR0iWSRgEzge2A8cBw4NBKz4+ICyNiRkTMaGpq6moYm0VyXwYzawyVDrv9rz183Q8Cf4mIVgBJ15OG9B4paUBWapgILOrh63ZLc7Orksys79vU6Ko/iohTJP03HfRujogPdfG6LwF7SxpGqko6CJgD3A0cBVwFnAjc1MXXr4rJk+G3v807CjOz6tpUieG/sscf9ORFI+IBSdcCjwDrgUeBC0nNYK+S9J1s20U9ed3uam6GJUtg7VoYPDjvaMzMqmNTw24/nD3+vqcvHBFnAme22zwf2Kunr9VTCve5Fy6EqVPzjcXMrFoq7eC2o6RrJT0taX5hqXZw9cad3MysEVTaKulXwAWkap8PAJcBl1crqHrlTm5m1ggqTQxDI+IuQBHxYkScBRxRvbDq08SJ6dEtk8ysL6u0g9taSf2A5yV9ntSMdIvqhVWfBg+GbbZxicHM+rbNGStpGGl8o78Fjic1J204npfBzPq6TZYYsuG2PxYR/wysBj5Z9ajqWHMzPPZY3lGYmVVPpyWGrBfyO8B+NYqn7nnCHjPr6zZVYngQmA48Kmk28BvgjcLOiLi+irHVpcmTUwe31lYYOzbvaMzMel6lN5+HAK+QhsUO0lSeATRcYigeftuJwcz6ok0lhrGSvgw8SVtCKGjIypTivgx77plvLGZm1bCpxNCf1CxVHexryMTg3s9m1tdtKjEsjohv1SSSXmLUKBg+3J3czKzv2lQ/ho5KCg3NE/aYWV+3qcRwUE2i6GXcyc3M+rJOE0NErKhVIL2JZ3Izs76s0iExrEhzMyxfDmvW5B2JmVnPc2LogkLLpIUL843DzKwacksMkkZmk/88K+kZSftIGi3pDknPZ4+j8oqvM8Wd3MzM+po8Sww/Bm6LiHcBewDPAGcAd0XEjsBd2fO64wl7zKwvyyUxSNoK2B+4CCAi3o6IVcBM4NLssEuBD+cR36ZMmAD9+jkxmFnflFeJYTugFfiVpEcl/VLScGBcRCzOjlkCjOvoZEmzJM2RNKe1tbVGIbcZOBDGj3dVkpn1TXklhgGkUVsviIj3kEZsLak2ioigzLAbEXFhRMyIiBlNTU1VD7Yj7uRmZn1VXomhBWiJiAey59eSEsVSSdsCZI/Lcopvk9zJzcz6qlwSQ0QsARZKmpZtOgh4GphN25ShJwI35RBeRZqbU3PVDRvyjsTMrGfl2SrpC8AVkuYC7wb+DTgb+HtJzwMfzJ7XpZ12gnXr4LLL8o7EzKxnVTpRT4+LiMeAGR3s6hXjMx1/PPz61/DpT6dWSn//93lHZGbWM9zzuYsGDYLrroOdd4aPfAQeeyzviMzMeoYTQzdstRXcckt6PPxwN181s77BiaGbJkyAW29NA+oddhisXJl3RGZm3ePE0AN23RVuvBFeeAFmzoS33so7IjOzrnNi6CEHHACXXgr33QcnnOBmrGbWe+XWKqkvOuYYaGmB006DSZPghz/MOyIzs83nxNDDTj01dXw799yUHE45Je+IzMw2jxNDD5NSUmhpgS9/Od2cPvrovKMyM6uc7zFUQf/+cPnl8L73wcc/nu47mJn1Fk4MVTJ0KNx0E0yZkloqPfNM3hGZmVXGiaGKtt469XEYNAgOPRRefjnviMzMNs2Jocq22y71jn7lFTjiCHjttbwjMjPrnBNDDUyfDtdeC088AUcdlUZlNTOrV04MNXLoofCLX8Add6QRWaPDuenMzPLn5qo19MlPpj4OZ56Z+jh8+9t5R2RmtjEnhhr7l39JyeE730nJYdasvCMyMyvlqqQak+CCC9Iw3Z/5TOoEt2ZN3lGZmbVxYsjBgAHwm9/AZz8L550He+wB996bd1RmZkmuiUFSf0mPSro5e76dpAckzZN0taRBecZXTcOGwfnnw+9+l0Ziff/74QtfgNWr847MzBpd3iWGLwHFfYLPAc6LiKnASuDkXKKqoQ98AObOhS9+MSWK3XdPycLMLC+5JQZJE4EjgF9mzwUcCFybHXIp8OF8oqut4cPhxz9O1UkDBsBBB6VqptdfzzsyM2tEeZYYfgR8BShMabM1sCoi1mfPW4AJHZ0oaZakOZLmtLa2Vj/SGtlvP3jssTR0989/nmaGu/32vKMys0aTS2KQdCSwLCIe7sr5EXFhRMyIiBlNTU09HF2+hg2DH/wA/vCHtH7IIfCpT8Grr+YdmZk1irxKDPsCH5K0ALiKVIX0Y2CkpELfionAonzCy98++8Cjj8Lpp8OvfgW77JLGXDIzq7ZcEkNEfDUiJkbEFOAY4HcRcRxwN3BUdtiJwE15xFcvhgyBs8+G+++HkSPTIHwnnggrV+YdmZn1ZXm3SmrvdODLkuaR7jlclHM8dWHPPeHhh+Eb34ArroCdd4bZs/OOysz6qtwTQ0TcExFHZuvzI2KviJgaEUdHxNq846sXgwensZUefBDGjk2T/xxzTCpNeEA+M+tJuScG2zzTp8NDD8FZZ6UZ4vbZB6ZNg299C154Ie/ozKwvcGLohQYNSiO0LlkCF10EEyak51Onwr77ws9+BitW5B2lmfVWTgy92FZbwUknwd13w4svwr//O6xalTrHbbMN/MM/wPXXw1pXyJnZZnBi6COam+GMM+DJJ+GRR+Dzn4c//Qk+8pGUJP7xH+G++9K4TGZmnXFi6GMkeM974NxzoaUFbrstNXO9/HLYf3/YYYc0J8Rzz+UdqZnVK0Uvb9IyY8aMmDNnTt5h1L3Vq+GGG1KCuPPOVHKYPj0N4rfvvvC+98G4cXlHaWa1IunhiJjR4T4nhsbz8stw1VXp/sNDD8Hbb6ft22/fliT23Tf1l+jfP99Yzaw6nBisrLVr0z2JP/wB/vjH9LhsWdq35Zaw995tyeK974URI/KN18x6RmeJwXM+N7jBg1NfiH32Sc8jYP780kRx1llpe79+ab6IQqLYb79009vM+haXGGyTVq2CBx5oSxb33w9vvJH2HXFEGib8gAPSjW8z6x1cYrBuGTkyDf99yCHp+fr18MQTabym88+HAw9MN7JPPRWOPhoGDsw3XjPrHjdXtc02YEBqEnvmmalj3YUXphLEccel5rDnnguvvZZ3lGbWVU4M1i1Dh8KnPw1PPw3//d+pZdOpp8KkSXDaabBwYd4RmtnmcmKwHtGvHxx5JNxzTxoB9rDD4LzzUqI4/vg06ZCZ9Q5ODNbj9twz9ZOYNy8NzXHjjekexEEHwa23ephws3rnxGBVM2VKKjW0tMA558Czz8Lhh8Nuu6XpSj24n1l9cnNVq5m3304liR/+EObOhdGjU1XTuHEbL2PHtq2PHp2qqsys59Rdz2dJk4DLgHFAABdGxI8ljQauBqYAC4CPRkSnMxw7MfQ+EWm8piuvTMNzLF2almXL4J13Nj5+wABoato4eWy7bRq2Y7fd0rr7UZhVrh4Tw7bAthHxiKQRwMPAh4FPACsi4mxJZwCjIuL0zl7LiaHv2LABVq5sSxTllmXL0mNxVdTWW6cEsdtuqXf2brvBLrvAFlvk937M6lnddXCLiMXA4mz9dUnPABOAmcAB2WGXAvcAnSYG6zv69Utf8FtvnUoCnYmAV15J80888URa5s6Fiy9u65Utpaqq4mSx225ppjsPDmhWXu73GCRNAe4FdgVeioiR2XYBKwvP250zC5gF0Nzc/LcvvvhizeK1+rZhAyxYkJJEccJ4/vm2SYqGDm2rgiqULHbdFcaPd3WUNY66q0r668WlLYDfA9+NiOslrSpOBJJWRsSozl7DVUlWiTffhGee2ThhLF3adszIkW1JorDssku6v2HW19RdVRKApIHAdcAVEXF9tnmppG0jYnF2H2JZXvFZ3zJ0aOpLMX166fbly+Gpp1KVVOHxmmvg5z9vO2bs2LYkUZwwttqqtu/BrFZySQxZNdFFwDMRcW7RrtnAicDZ2eNNOYRnDWTMGHj/+9NSEAFLlqQkUVieeqr0/gXAxIkwbdrGS3Ozm9da75ZXq6T9gPuAJ4DC9PRfAx4ArgGagRdJzVVXdPZarkqyWtmwAV56qbR08dxzaSkeNHDIENhxR9hpp42TxsiN7piZ5aNu7zH0BCcGy1tEakJbSBLFy/z5pX0zxo5tSxhTp8KwYWmY8oEDU3+Njh7L7Rs0KFWRDR+eliFDfPPcKleX9xjM+gqprdPd/vuX7nv77ZQc/vzn0oQxeza0tvZsHP36pURTSBSbWkaMSNVeU6emZfTono3Hei8nBrMqGjQI3vWutLS3enXqpLduXZr8aN260vVyj4X1t9+GNWvSfY9yS2H/8uUb72vfy3zkyJQgdtihLVkU1rfZxqWRRuLEYJaTLbbIr2d2REoaCxbACy+kkXDnzUvrDz0E115bmjiGDWtLEoXH7bdPyaTwPoYPT4+DBuXznqznODGYNSApfZHvskta2lu3Ls3OV0gWhcTx7LNwyy2dj4w7cGBbkig8llsfNqyyZejQtnVPHVt9TgxmtpGBA9uqk9rbsAEWLYK//CW1xnrjjVQttnp123pH25YuTfdbiretW7f5sQ0Y0HYvZbvt2qrqCst226VjrOv88ZnZZunXL03dOmlS919r3brUK33Nms1fXnstlWZuvjn1MSkYODA1F542rTRhTJvmTomVcmIws9wUmuNuuWX3XmflytTa69ln25bCPOTr17cdt+22bUlip53SYIpr18Jbb6XH4vVNbVu7NpVMhgyBwYPbHjdnfciQtqqyckthfy2r0JwYzKzXGzUK9t47LcXWrUvVV8UJ47nn0oRRq1aVHiu1fVmX+0IfMSL1lh8yJN1kX7++NGG8/npqAVYusXSl6qygf/+Nk8bXvgYnnND11yzHicHM+qyBA9t6nc+c2bY9AlZkYyoUvvwHDKh+k9wNG0qTxZtvbt6yZk3p8zFjqhOnE4OZNRwpzftRa/36tf23X8881JeZmZVwYjAzsxJODGZmVsKJwczMSjgxmJlZCScGMzMr4cRgZmYlnBjMzKxEr5/aU1IraX7orhgDLO/BcHqa4+ueeo8P6j9Gx9c99Rzf5Iho6mhHr08M3SFpTrk5T+uB4+ueeo8P6j9Gx9c99R5fOa5KMjOzEk4MZmZWotETw4V5B7AJjq976j0+qP8YHV/31Ht8HWroewxmZraxRi8xmJlZO04MZmZWoiESg6RDJT0naZ6kMzrYP1jS1dn+ByRNqWFskyTdLelpSU9J+lIHxxwg6VVJj2XLN2sVX3b9BZKeyK49p4P9kvST7PObK2l6DWObVvS5PCbpNUmntDum5p+fpIslLZP0ZNG20ZLukPR89jiqzLknZsc8L+nEGsb3fUnPZj/DGySNLHNup78PVYzvLEmLin6Oh5c5t9O/9yrGd3VRbAskPVbm3Kp/ft0WEX16AfoDLwDbA4OAx4Gd2x3zT8DPsvVjgKtrGN+2wPRsfQTw5w7iOwC4OcfPcAEwppP9hwO3AgL2Bh7I8We9hNRxJ9fPD9gfmA48WbTte8AZ2foZwDkdnDcamJ89jsrWR9UovoOBAdn6OR3FV8nvQxXjOwv45wp+Bzr9e69WfO32/xD4Zl6fX3eXRigx7AXMi4j5EfE2cBUws90xM4FLs/VrgYOkasgkd0YAAAUESURBVM/+mkTE4oh4JFt/HXgGmFCLa/egmcBlkdwPjJS0bQ5xHAS8EBFd7QnfYyLiXmBFu83Fv2eXAh/u4NRDgDsiYkVErATuAA6tRXwRcXtErM+e3g9M7OnrVqrM51eJSv7eu62z+LLvjo8CV/b0dWulERLDBGBh0fMWNv7i/esx2R/Gq0DNZ4TNqrDeAzzQwe59JD0u6VZJu9Q0MAjgdkkPS5rVwf5KPuNaOIbyf4x5fn4F4yJicba+BBjXwTH18lmeRCoFdmRTvw/V9PmsquviMlVx9fD5/R2wNCKeL7M/z8+vIo2QGHoFSVsA1wGnRMRr7XY/Qqoe2QP4D+DGGoe3X0RMBw4DPidp/xpff5MkDQI+BPymg915f34biVSnUJdtxSV9HVgPXFHmkLx+Hy4AdgDeDSwmVdfUo2PpvLRQ939PjZAYFgGTip5PzLZ1eIykAcBWwCs1iS5dcyApKVwREde33x8Rr0XE6mz9FmCgpDG1ii8iFmWPy4AbSMX1YpV8xtV2GPBIRCxtvyPvz6/I0kIVW/a4rINjcv0sJX0COBI4LkteG6ng96EqImJpRLwTERuAX5S5bt6f3wDg/wBXlzsmr89vczRCYngI2FHSdtl/lccAs9sdMxsotP44CvhduT+KnpbVR14EPBMR55Y5ZpvCPQ9Je5F+bjVJXJKGSxpRWCfdoHyy3WGzgROy1kl7A68WVZnUStn/0vL8/Nop/j07Ebipg2P+BzhY0qisquTgbFvVSToU+ArwoYhYU+aYSn4fqhVf8X2rfyhz3Ur+3qvpg8CzEdHS0c48P7/Nkvfd71ospFYzfya1Vvh6tu1bpD8AgCGkKoh5wIPA9jWMbT9SlcJc4LFsORz4DPCZ7JjPA0+RWljcD7yvhvFtn1338SyGwudXHJ+A87PP9wlgRo1/vsNJX/RbFW3L9fMjJanFwDpSPffJpPtWdwHPA3cCo7NjZwC/LDr3pOx3cR7wyRrGN49UP1/4PSy01BsP3NLZ70ON4vuv7PdrLunLftv28WXPN/p7r0V82fZLCr93RcfW/PPr7uIhMczMrEQjVCWZmdlmcGIwM7MSTgxmZlbCicHMzEo4MZiZWQknBrMyJL3TbuTWHhupU9KU4pE5zerJgLwDMKtjb0bEu/MOwqzWXGIw20zZePrfy8bUf1DS1Gz7FEm/ywZ5u0tSc7Z9XDa/wePZ8r7spfpL+oXSPBy3SxqaHf9Fpfk55kq6Kqe3aQ3MicGsvKHtqpI+VrTv1YjYDfgp8KNs238Al0bE7qQB6H6Sbf8J8PtIg/hNJ/V4BdgROD8idgFWAR/Jtp8BvCd7nc9U682ZleOez2ZlSFodEVt0sH0BcGBEzM8GQFwSEVtLWk4apmFdtn1xRIyR1ApMjIi1Ra8xhTTvwo7Z89OBgRHxHUm3AatJo8DeGNkAgGa14hKDWddEmfXNsbZo/R3a7vkdQRp7ajrwUDZip1nNODGYdc3Hih7/lK3/kTSaJ8BxwH3Z+l3AZwEk9Ze0VbkXldQPmBQRdwOnk4aA36jUYlZN/k/ErLyh7SZ0vy0iCk1WR0maS/qv/9hs2xeAX0k6DWgFPplt/xJwoaSTSSWDz5JG5uxIf+DyLHkI+ElErOqxd2RWAd9jMNtM2T2GGRGxPO9YzKrBVUlmZlbCJQYzMyvhEoOZmZVwYjAzsxJODGZmVsKJwczMSjgxmJlZif8FRX3Nah9nRCQAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_Awe_tLzKcnG",
        "colab_type": "code",
        "outputId": "435f1529-710d-4efb-c149-52fd63dcee93",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 218
        }
      },
      "source": [
        "#evaluating our trained model on test data\n",
        "with torch.no_grad():\n",
        "  \n",
        "  model.eval()\n",
        "  n_samples = 0\n",
        "  n_correct = 0\n",
        "  n_class_correct = [0 for i in range(10)]\n",
        "  n_class_samples = [0 for i in range(10)]\n",
        "\n",
        "  for (samples,labels) in test_loader:\n",
        "    samples = samples.to(device)\n",
        "    labels = labels.to(device)\n",
        "    y_pred = model(samples)\n",
        "    _,prediction = torch.max(y_pred,1)\n",
        "    n_samples+= labels.shape[0]\n",
        "    n_correct+= (prediction == labels).sum().item()\n",
        "\n",
        "    for i in range(batch_size):\n",
        "      label = labels[i].item()\n",
        "      predict = prediction[i].item()\n",
        "      if label == predict:\n",
        "        n_class_correct[label]+= 1\n",
        "      n_class_samples[label]+= 1\n",
        "  \n",
        "  #printing overall accuracy\n",
        "  accuracy = 100*n_correct/(float)(n_samples)\n",
        "  print(\"Overall accuracy of system: {0:.2f}%\".format(accuracy))  \n",
        "\n",
        "  #printing accuracy of each class\n",
        "  for i in range(10):\n",
        "    acc=100*n_class_correct[i]/n_class_samples[i]\n",
        "    print(\"Accuracy of digit\",i,\"= {:.2f}%\".format(acc)) \n",
        "\n",
        "#print(\"Time taken by CPU model= {:.2f} sec\".format(time.time()-start_time))-----output=2665.37sec\n",
        "print(\"Time taken by GPU model= {:.2f} sec\".format(time.time()-start_time))  "
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Overall accuracy of system: 99.26%\n",
            "Accuracy of digit 0 = 99.59%\n",
            "Accuracy of digit 1 = 99.91%\n",
            "Accuracy of digit 2 = 99.52%\n",
            "Accuracy of digit 3 = 99.11%\n",
            "Accuracy of digit 4 = 99.49%\n",
            "Accuracy of digit 5 = 99.22%\n",
            "Accuracy of digit 6 = 99.06%\n",
            "Accuracy of digit 7 = 98.74%\n",
            "Accuracy of digit 8 = 99.18%\n",
            "Accuracy of digit 9 = 98.71%\n",
            "Time taken by GPU model= 135.37 sec\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "76Ky421qui6z",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}