HW1_MNIST_NN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "HW1_MNIST_NN.ipynb",
      "provenance": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyNOPvbpnElQYCfIUqHw6ddy",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ashaaher/Deep-Learning/blob/master/HW1_MNIST_NN.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZGlXFlKTLRTn",
        "colab_type": "text"
      },
      "source": [
        "# Importing Libraries\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZenAnuB3LN38",
        "colab_type": "text"
      },
      "source": [
        "import keras\n",
        "import pandas as pd\n",
        "import numpy as np,sys\n",
        "from keras.datasets import mnist\n",
        "import copy\n",
        "from matplotlib import pyplot\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "import os\n",
        "np.random.seed(1)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ufRJjG8RLEOH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 80
        },
        "outputId": "9bcafbb7-e9ec-49bd-e50e-e21de449f56e"
      },
      "source": [
        "import keras\n",
        "import pandas as pd\n",
        "import numpy as np,sys\n",
        "from keras.datasets import mnist\n",
        "import copy\n",
        "from matplotlib import pyplot\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "import os\n",
        "np.random.seed(1)"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<p style=\"color: red;\">\n",
              "The default version of TensorFlow in Colab will soon switch to TensorFlow 2.x.<br>\n",
              "We recommend you <a href=\"https://www.tensorflow.org/guide/migrate\" target=\"_blank\">upgrade</a> now \n",
              "or ensure your notebook will continue to use TensorFlow 1.x via the <code>%tensorflow_version 1.x</code> magic:\n",
              "<a href=\"https://colab.research.google.com/notebooks/tensorflow_version.ipynb\" target=\"_blank\">more info</a>.</p>\n"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Pr-jfxKHLYCK",
        "colab_type": "text"
      },
      "source": [
        "# Loading and Visualization of Dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iX4jmXa11AYh",
        "colab_type": "code",
        "outputId": "4143400d-49e9-4a25-99e1-25792bad00f2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 302
        }
      },
      "source": [
        "from keras.datasets import mnist\n",
        "(x_train,y_train),(x_test,y_test)=mnist.load_data()\n",
        "for i in range(0, 9):\n",
        "\tpyplot.subplot(330 + 1 + i)\n",
        "\tpyplot.imshow(x_train[i], cmap=pyplot.get_cmap('gray'))"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://s3.amazonaws.com/img-datasets/mnist.npz\n",
            "11493376/11490434 [==============================] - 0s 0us/step\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAU4AAAD7CAYAAAAFI30bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 9 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WOJx7rOoLeDy",
        "colab_type": "text"
      },
      "source": [
        "# Normalizing the inputs"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n5b18ZDj09Qv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "images,labels = (x_train.reshape(len(x_train), 28*28)/255,y_train)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5xQPmJgoLmu8",
        "colab_type": "text"
      },
      "source": [
        "# One hot encoding"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DK3VSOy83MXz",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "one_hot_labels = np.zeros((len(labels),10))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "garYKpLm3ePm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "for i, l in enumerate(labels):\n",
        "  one_hot_labels[i][l] = 1\n",
        "labels =one_hot_labels"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LYP9fJTW7XL0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "test_images = x_test.reshape(len(x_test),28*28)/255\n",
        "test_labels = np.zeros((len(y_test),10))\n",
        "for i,l in enumerate(y_test):\n",
        "    test_labels[i][l] =1"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dDTyIZPRLqtW",
        "colab_type": "text"
      },
      "source": [
        "# Image augmentation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eVf5Qyxu0fks",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\n",
        "x_train = x_train.reshape((x_train.shape[0], 28, 28, 1))\n",
        "x_test = x_test.reshape((x_test.shape[0], 28, 28, 1))\n",
        "x_train = x_train.astype('float32')\n",
        "x_test = x_test.astype('float32')\n",
        "datagen = ImageDataGenerator()\n",
        "datagen.fit(x_train)\n",
        "os.makedirs('images6')\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e2yi2i9sLvBX",
        "colab_type": "text"
      },
      "source": [
        "# Defining Nework Arhitecture Using 3 Neuron Layers (Input and Hidden layers)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qrp2gfOv8Cwy",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "np.random.seed(1)\n",
        "batch_size = 100\n",
        "def tanh(x):\n",
        "  return np.tanh(x)\n",
        "def tanh2deriv(output):\n",
        "  return 1-(output**2)\n",
        "relu = lambda x:(x>=0)*x\n",
        "relu2deriv = lambda x:x>=0\n",
        "def softmax(x):\n",
        "   temp = np.exp(x)\n",
        "   return temp/np.sum(temp,axis =1,keepdims =True)\n",
        "alpha,iterations,hidden_size, pixels_per_image,num_labels =(0.01,300,300,784,10)\n",
        "weights_0_1 = np.random.randn(pixels_per_image,hidden_size) * np.sqrt(2.0/hidden_size)\n",
        "weights_1_2 = np.random.randn(hidden_size,100)*np.sqrt(2.0/100)\n",
        "weights_2_3 = np.random.randn(100,num_labels)*np.sqrt(2.0/num_labels)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8zVtteG6L91o",
        "colab_type": "text"
      },
      "source": [
        "# Implementing Minibatch gradient descent along with appropriate learning rate and various dropout rates."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jn5YE2vH-QUw",
        "colab_type": "code",
        "outputId": "28206387-1281-45d5-cf49-73e21544b85b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 561
        }
      },
      "source": [
        "for j in range (iterations):\n",
        "  correct_cnt = 0\n",
        "  error =0.0\n",
        "  for i in range(int(len(images)/batch_size)):\n",
        "    batch_start,batch_end = ((i*batch_size),((i+1)*batch_size))\n",
        "    layer_0=images[batch_start:batch_end]\n",
        "    layer_1=tanh(np.dot(layer_0,weights_0_1))\n",
        "    dropout_mask = np.random.randint(2,size=layer_1.shape)\n",
        "    layer_1 *= dropout_mask*2\n",
        "    layer_2 = relu(np.dot(layer_1,weights_1_2))\n",
        "    layer_3 = softmax(np.dot(layer_2,weights_2_3))\n",
        "    error+=np.sum((labels[batch_start:batch_end]-layer_3)**2)\n",
        "   \n",
        "\n",
        "    for k in range (batch_size):\n",
        "      correct_cnt+=int(np.argmax(layer_3[k:k+1])==np.argmax(labels[batch_start+k:batch_start+k+1]))\n",
        "    layer_3_delta = (labels[batch_start:batch_end]-layer_3)/(batch_size * layer_3.shape[0])\n",
        "    layer_2_delta = layer_3_delta.dot(weights_2_3.T)*relu2deriv(layer_2)\n",
        "   \n",
        "    layer_1_delta = layer_2_delta.dot(weights_1_2.T)*relu2deriv(layer_1)\n",
        "    layer_1_delta *= dropout_mask\n",
        "\n",
        "\n",
        "    weights_2_3 += alpha*layer_2.T.dot(layer_3_delta)\n",
        "    weights_1_2 += 0.1*layer_1.T.dot(layer_2_delta)\n",
        "    weights_0_1 += 0.2*layer_0.T.dot(layer_1_delta)\n",
        "  test_correct_cnt = 0\n",
        "  y_pred1=[]\n",
        "  act_labels=[]\n",
        "\n",
        "  for i in range (len(test_images)):\n",
        "    layer_0 = test_images[i:i+1]\n",
        "    layer_1 = tanh(np.dot(layer_0,weights_0_1))\n",
        "    layer_2 = relu(np.dot(layer_1,weights_1_2))\n",
        "    layer_3 = np.dot(layer_2,weights_2_3)\n",
        "    y_pred1 += [int(np.argmax(layer_3))]\n",
        "    act_labels += [np.argmax(test_labels[i:i+1])]\n",
        "    test_correct_cnt += int(np.argmax(layer_3)==np.argmax(test_labels[i:i+1]))\n",
        "   \n",
        "  if (j%10==0 or j==iterations-1):\n",
        "   sys.stdout.write(\"\\n\"+\"I:\"+str(j)+\" Test- Acc:\"+str(test_correct_cnt/float(len(test_images)))+\" Train-Acc:\"+'{:0.6f}'.format(correct_cnt/float(len(images))))\n",
        "  \n",
        "  "
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "I:0 Test- Acc:0.7109Train-Acc:0.368417\n",
            "I:10 Test- Acc:0.8783Train-Acc:0.799350\n",
            "I:20 Test- Acc:0.8923Train-Acc:0.825667\n",
            "I:30 Test- Acc:0.8948Train-Acc:0.832967\n",
            "I:40 Test- Acc:0.896Train-Acc:0.837483\n",
            "I:50 Test- Acc:0.8995Train-Acc:0.839433\n",
            "I:60 Test- Acc:0.9016Train-Acc:0.842600\n",
            "I:70 Test- Acc:0.9026Train-Acc:0.841883\n",
            "I:80 Test- Acc:0.9015Train-Acc:0.843233\n",
            "I:90 Test- Acc:0.9044Train-Acc:0.845667\n",
            "I:100 Test- Acc:0.9053Train-Acc:0.848267\n",
            "I:110 Test- Acc:0.9062Train-Acc:0.849183\n",
            "I:120 Test- Acc:0.9074Train-Acc:0.850783\n",
            "I:130 Test- Acc:0.9086Train-Acc:0.851483\n",
            "I:140 Test- Acc:0.9107Train-Acc:0.851150\n",
            "I:150 Test- Acc:0.9139Train-Acc:0.853600\n",
            "I:160 Test- Acc:0.9148Train-Acc:0.854100\n",
            "I:170 Test- Acc:0.915Train-Acc:0.854983\n",
            "I:180 Test- Acc:0.9151Train-Acc:0.856050\n",
            "I:190 Test- Acc:0.9165Train-Acc:0.857050\n",
            "I:200 Test- Acc:0.9177Train-Acc:0.857067\n",
            "I:210 Test- Acc:0.9192Train-Acc:0.856850\n",
            "I:220 Test- Acc:0.9193Train-Acc:0.858433\n",
            "I:230 Test- Acc:0.9199Train-Acc:0.857867\n",
            "I:240 Test- Acc:0.9225Train-Acc:0.858550\n",
            "I:250 Test- Acc:0.9222Train-Acc:0.861117\n",
            "I:260 Test- Acc:0.9221Train-Acc:0.860067\n",
            "I:270 Test- Acc:0.9219Train-Acc:0.859817\n",
            "I:280 Test- Acc:0.9229Train-Acc:0.860633\n",
            "I:290 Test- Acc:0.9222Train-Acc:0.861400\n",
            "I:299 Test- Acc:0.9227Train-Acc:0.862983"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rvsFYf7QJa0y",
        "colab_type": "text"
      },
      "source": [
        "# Accuracy of Neural Network:\n",
        "Test Accuracy - 92 %, Train Accuracy - 86% with 300 epoch"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1fV406i6JCbe",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "10e51ffd-f1fb-4b4f-91c1-8d472493f630"
      },
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.metrics import accuracy_score\n",
        "accuracy_score(act_labels,y_pred1)*100"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "92.27"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "U6MTuaKnJQEN",
        "colab_type": "text"
      },
      "source": [
        "# Evaluating The Model with Confusion Matrix"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2hIbf7R-__1I",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 311
        },
        "outputId": "8d5b6e5f-7bbe-42d0-cec2-d41a78b4cab2"
      },
      "source": [
        "\n",
        "\n",
        "# Plotting Confusion Matrics\n",
        "import matplotlib.pyplot as plt\n",
        "import itertools\n",
        "def plot_conf_mat(cm,classes,\n",
        "                  normalize=False,\n",
        "                  title='Confusion-Matrix', \n",
        "                  cmap=plt.cm.Blues):\n",
        "  plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "  plt.title(title)\n",
        "  plt.colorbar()\n",
        "  tick_marks = np.arange(len(classes))\n",
        "  plt.xticks(tick_marks, classes, rotation=45)\n",
        "  plt.yticks(tick_marks, classes)\n",
        "\n",
        "  if normalize:\n",
        "    cm = cm.astype('float') / cm.sum(axis = 1)[:, np.newaxis]\n",
        "\n",
        "  thresh = cm.max() / 2.\n",
        "  for i,j in itertools.product(range(cm.shape[0]),range(cm.shape[1])):\n",
        "      plt.text(j,i, cm[i,j],horizontalalignment=\"center\",color=\"white\" if cm[i,j] > thresh else \"black\")\n",
        "\n",
        "  plt.tight_layout()\n",
        "  plt.ylabel('True label')\n",
        "  plt.xlabel('Predicted label')\n",
        "\n",
        "#Computing-Confusion Matrix\n",
        "conf_mat = confusion_matrix(act_labels, y_pred1)\n",
        "\n",
        "#plotting conf matrix\n",
        "plot_conf_mat(conf_mat, classes = range(10))"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVEAAAEmCAYAAADbUaM7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}