Week_3.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Rakshit-Sawarn-iitb/WIDS-Brain-Tumor-Detection/blob/main/Week_3.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4lFWhSEEwc9F"
      },
      "source": [
        "#MNIST\n",
        "Our objective is to build a neural network for the classification of the MNIST dataset. This neural network will comprise two layers, each with 10 nodes, and an input layer with 784 nodes corresponding to the image pixels. The specific structure of the neural network is outlined below, where $X$ represents the input, $A^{[0]}$ denotes the first layer, $Z^{[1]}$ signifies the unactivated layer 1, $A^{[1]}$ stands for the activated layer 1, and so forth. The weights and biases are represented by $W$ and $b$ respectively:\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dDanK4nEwfhh"
      },
      "source": [
        "<div align=\"center\">\n",
        "\n",
        "$A^{[0]}=X$\n",
        "\n",
        "$Z^{[1]}=W^{[1]}A^{[0]}+b^{[1]}$\n",
        "\n",
        "$A^{[1]}=\\text{ReLU}(Z^{[1]})$\n",
        "\n",
        "$Z^{[2]}=W^{[2]}A^{[1]}+b^{[2]}$\n",
        "\n",
        "$A^{[2]}=\\text{softmax}(Z^{[2]})$\n",
        "</div>\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YzBJgODl4aDp"
      },
      "source": [
        "You have the flexibility to create any function within or outside the class, allowing you to modify parameters as needed"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 88,
      "metadata": {
        "id": "8A6ScQ-8lzWy"
      },
      "outputs": [],
      "source": [
        "#importing libraries\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from keras.datasets import mnist\n",
        "import matplotlib.pyplot as plt\n",
        "import random"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TGVN81yBnufX"
      },
      "source": [
        "### Required functions"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 89,
      "metadata": {
        "id": "N4A3X8zjh5My"
      },
      "outputs": [],
      "source": [
        "# activation and loss functions\n",
        "def ReLU(x):\n",
        "  a = np.maximum(x,0)\n",
        "  return a\n",
        "\n",
        "def derivative_ReLU(x):\n",
        "  b = np.array(x>0,dtype = np.float32)\n",
        "  return b\n",
        "\n",
        "def softmax(x):\n",
        "  expx = np.exp(x)\n",
        "  return expx/np.sum(expx, axis = 0,keepdims = True)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 117,
      "metadata": {
        "id": "Vqdn0Wv0mFNE"
      },
      "outputs": [],
      "source": [
        "#complete the class of neural network\n",
        "\n",
        "class NN:\n",
        "  def __init__(self, lr = 0.001, n_iters = 100):\n",
        "      self.lr = lr\n",
        "      self.n_iters = n_iters\n",
        "      self.w1 = np.random.randn(10,784)*0.01\n",
        "      self.b1 = np.zeros((10,1))\n",
        "      self.w2 = np.random.randn(10,10)*0.01\n",
        "      self.b2 = np.zeros((10,1))\n",
        "      self.a1 = None\n",
        "      self.a2 = None\n",
        "      self.dw1=None\n",
        "      self.dw2 = None\n",
        "      self.db1 = None\n",
        "      self.db2 = None\n",
        "\n",
        "  def forward_propagation(self,x):\n",
        "    z1 = np.dot(self.w1,x.T)+self.b1\n",
        "    self.a1 = ReLU(z1)\n",
        "    z2 = np.dot(self.w2,self.a1)+self.b2\n",
        "    self.a2 = softmax(z2)\n",
        "\n",
        "    return self.a2\n",
        "\n",
        "  def one_hot(self, labels, num_classes): #return a 0 vector with 1 only in the position corresponding to the value in test target\n",
        "    num_samples = len(labels)\n",
        "    one_hot_labels = np.zeros((num_classes, num_samples))\n",
        "\n",
        "    for i in range(num_samples):\n",
        "      one_hot_labels[labels[i], i] = 1\n",
        "\n",
        "    return one_hot_labels\n",
        "\n",
        "  def cost_function(self,a,y):\n",
        "    m = (y.T).shape[0]\n",
        "    cost = -(1/m) * np.sum(y * np.log(a))\n",
        "    return cost\n",
        "\n",
        "  def backward_propagation(self,x,y):\n",
        "    w1 = self.w1\n",
        "    w2 = self.w2\n",
        "    b1 = self.b1\n",
        "    b2 = self.b2\n",
        "    a1 = self.a1\n",
        "    a2 = self.a2\n",
        "\n",
        "    m = x.shape[1]\n",
        "\n",
        "    dz2 = (a2 - y)\n",
        "    dw2 = (1/m)*(np.dot(dz2,a1.T))\n",
        "    db2 = (1/m)*(np.sum(dz2, axis=1, keepdims = True))\n",
        "    dz1 = (1/m)*(np.dot(w2.T,dz2))*(derivative_ReLU(a1))\n",
        "    dw1 = (1/m)*(np.dot(dz1,x))\n",
        "    db1 = (1/m)*(np.sum(dz1, axis=1, keepdims = True))\n",
        "\n",
        "    self.dw1 = dw1\n",
        "    self.dw2 = dw2\n",
        "    self.db1 = db1\n",
        "    self.db2 = db2\n",
        "\n",
        "  def update_params(self):\n",
        "\n",
        "    #dw1 = gradients['dw1']\n",
        "   # db1 = gradients['db1']\n",
        "    #dw2 = gradients['dw2']\n",
        "    #db2 = gradients['db2']\n",
        "\n",
        "    self.w1 -= self.lr*(self.dw1)\n",
        "    self.w2 -= self.lr*(self.dw2)\n",
        "    self.b1 -= self.lr*(self.db1)\n",
        "    self.b2 -= self.lr*(self.db2)\n",
        "\n",
        "  def get_predictions(self):\n",
        "    pass\n",
        "\n",
        "  def get_accuracy(self,y):\n",
        "    ypred_out = np.argmax(self.a2,0)\n",
        "    yout = np.argmax(y,0)\n",
        "\n",
        "    acc = np.mean(yout == ypred_out)*100\n",
        "\n",
        "    return acc\n",
        "\n",
        "  def gradient_descent(self, x, y):\n",
        "    for i in range(self.n_iters):\n",
        "      self.a2 = self.forward_propagation(x)\n",
        "      self.backward_propagation(x,y)\n",
        "      self.update_params()\n",
        "      cost = self.cost_function(self.a2, y)\n",
        "      if i % (self.n_iters/10) == 0:\n",
        "        print(f\"Iteration {i}, Cost: {cost}\")\n",
        "\n",
        "    return self.w1, self.b1, self.w2, self.b2\n",
        "\n",
        "  def make_predictions(self,x):\n",
        "    ypred = self.forward_propagation(x)\n",
        "    predicted = np.argmax(ypred)\n",
        "    return predicted\n",
        "\n",
        "  def show_prediction(self,x): #show the prediction and actual output for an image in mnist dataset\n",
        "    idx = random.randrange(0, x.shape[0])\n",
        "    single_image = x[idx,: ].reshape((28, 28))\n",
        "    plt.imshow(single_image, cmap='gray')\n",
        "    plt.show()\n",
        "\n",
        "    input_image = x[idx, :].reshape(x[idx, :].shape[0], 1)\n",
        "    prediction = self.make_predictions(input_image.T)\n",
        "\n",
        "    print(\"Model predicts:\", prediction)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FsgNaz6qmoLI"
      },
      "source": [
        "## main"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 118,
      "metadata": {
        "id": "iIbC5z1Lmlcr"
      },
      "outputs": [],
      "source": [
        "(X_train, Y_train), (X_test, Y_test) = mnist.load_data()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ymme4NNNmws9"
      },
      "source": [
        "###preprocessing the data\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 119,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "B8YjLpUwm678",
        "outputId": "2ab8a5c9-2370-4240-8fcb-7e8e521c0f77"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(60000, 784)\n",
            "(10000, 784)\n"
          ]
        }
      ],
      "source": [
        "# all values of pixels should be in range[0,1]\n",
        "X_train = (X_train.reshape(X_train.shape[0], -1)/ 255.0)\n",
        "X_test = (X_test.reshape(X_test.shape[0], -1) / 255.0)\n",
        "print(X_train.shape)\n",
        "print(X_test.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 120,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OOpl2afQ7OOA",
        "outputId": "3c46f826-be82-4d8c-e2bd-78b12bc8045a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(10, 60000)\n"
          ]
        }
      ],
      "source": [
        "nn = NN(lr = 0.01,n_iters=1000)\n",
        "y_train_encoded = nn.one_hot(Y_train,10)\n",
        "y_test_encoded = nn.one_hot(Y_test,10)\n",
        "\n",
        "print(y_train_encoded.shape)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B5dqfE25m7ZD"
      },
      "source": [
        "###Model Training"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 121,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G05ggxM1m_n0",
        "outputId": "dd33f3fb-54ac-4917-f5df-0e9e1344c1ef"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Iteration 0, Cost: 2.3024862966435227\n",
            "Iteration 100, Cost: 2.222690588875354\n",
            "Iteration 200, Cost: 1.8242377501065739\n",
            "Iteration 300, Cost: 1.2634363065903036\n",
            "Iteration 400, Cost: 0.9033934375316045\n",
            "Iteration 500, Cost: 0.7201150489469026\n",
            "Iteration 600, Cost: 0.6203174941905154\n",
            "Iteration 700, Cost: 0.5585100449571108\n",
            "Iteration 800, Cost: 0.5163970559489997\n",
            "Iteration 900, Cost: 0.4858178877340743\n"
          ]
        }
      ],
      "source": [
        "#training model using gradient descent\n",
        "w1, b1, w2, b2 = nn.gradient_descent(X_train,y_train_encoded)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 122,
      "metadata": {
        "id": "KyTbyG1MvjLv"
      },
      "outputs": [],
      "source": [
        "nn.w1 = w1\n",
        "nn.w2 = w2\n",
        "nn.b1 = b1\n",
        "nn.b2 = b2"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 123,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8DIl2E1Dv5Ub",
        "outputId": "8186fd3f-1d3b-463d-9e4f-cf953b34643f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "87.28\n"
          ]
        }
      ],
      "source": [
        "ypred = nn.make_predictions(X_test)\n",
        "\n",
        "acc = nn.get_accuracy(y_test_encoded)\n",
        "print(acc)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sa-CT3UnnAsr"
      },
      "source": [
        "### Viewing Results\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 124,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "WV9UEIHbnJKd",
        "outputId": "f4c9807b-4c08-4bec-d954-7a5bc0b8f1a2"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model predicts: 4\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model predicts: 1\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model predicts: 5\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model predicts: 8\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model predicts: 0\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model predicts: 9\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model predicts: 6\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model predicts: 6\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model predicts: 6\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model predicts: 2\n"
          ]
        }
      ],
      "source": [
        "#viewing prediction for 10 random images in dataset\n",
        "for epoch in range(10) :\n",
        "  nn.show_prediction(X_train)"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}