DS-Project.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install matplotlib\n",
    "import numpy as np\n",
    "from tensorflow.keras.datasets import mnist\n",
    "from matplotlib import pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "class NeuralNetwork:\n",
    "    def __init__(self, input_size, hidden_size, output_size, lr=0.01):\n",
    "        self.input_size = input_size\n",
    "        self.hidden_size = hidden_size\n",
    "        self.output_size = output_size\n",
    "        self.learningrate = lr\n",
    "\n",
    "        self.W1 = np.random.randn(input_size, hidden_size)\n",
    "        self.b1 = np.random.randn(hidden_size)\n",
    "\n",
    "        self.W2 = np.random.randn(hidden_size, output_size)\n",
    "        self.b2 = np.random.randn(output_size)\n",
    "\n",
    "    def relu(self, x):\n",
    "        return np.maximum(0, x)\n",
    "\n",
    "    def softmax(self, x):\n",
    "        exp_x = np.exp(x - np.max(x))\n",
    "        return exp_x / np.sum(exp_x)\n",
    "\n",
    "    def forward(self, x):\n",
    "        z1 = x.dot(self.W1) + self.b1\n",
    "        a1 = self.relu(z1)\n",
    "        z2 = a1.dot(self.W2) + self.b2\n",
    "        y_pred = self.softmax(z2)\n",
    "        return y_pred, a1\n",
    "\n",
    "    def backward(self, X, y_true, a1):\n",
    "        n = X.shape[0]\n",
    "        total_loss = 0\n",
    "\n",
    "        for i in range(n):\n",
    "            y_pred, a1 = self.forward(X[i])\n",
    "            loss = self.calculate_loss(y_true[i], y_pred)\n",
    "            total_loss += loss\n",
    "\n",
    "            dz2 = (y_pred - y_true[i]) / 1\n",
    "            dW2 = a1.reshape(a1.shape[0], 1).dot(dz2.reshape(1, self.output_size))\n",
    "            db2 = np.sum(dz2)\n",
    "\n",
    "            da1 = dz2.dot(self.W2.T)\n",
    "            dz1 = (da1 * (a1 > 0)).reshape(-1, 1)\n",
    "            dW1 = X[i].reshape(1, -1).T.dot(dz1.T)\n",
    "            db1 = np.sum(dz1, axis=1)\n",
    "\n",
    "            self.W1 -= self.learningrate * dW1\n",
    "            self.b1 -= self.learningrate * db1\n",
    "            self.W2 -= self.learningrate * dW2\n",
    "            self.b2 -= self.learningrate * db2\n",
    "\n",
    "        self.loss = total_loss / n\n",
    "\n",
    "    def calculate_loss(self, Y_True, Y_Pred):\n",
    "        epsilon = 1e-7\n",
    "        loss = -(Y_True * np.log(Y_Pred + epsilon) + (1 - Y_True) * np.log(1 - Y_Pred + epsilon))\n",
    "        return loss\n",
    "\n",
    "    def predict(self, X):\n",
    "        y_pred = []\n",
    "        for x in X:\n",
    "            y_pred.append(self.forward(x)[0])\n",
    "        return np.array(y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "nn = NeuralNetwork(input_size=784, hidden_size=128, output_size=10, lr=0.001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the MNIST dataset\n",
    "(X_train, y_train), (X_test, y_test) = mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Flatten the images\n",
    "X_train = X_train.reshape(-1, 784)\n",
    "X_test = X_test.reshape(-1, 784)\n",
    "\n",
    "# Normalize the pixel values\n",
    "X_train = X_train.astype('float32') / 255\n",
    "X_test = X_test.astype('float32') / 255\n",
    "\n",
    "# Convert the labels to one-hot encoding\n",
    "y_train = np.eye(10)[y_train]\n",
    "y_test = np.eye(10)[y_test]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Store losses for each epoch to plot them later\n",
    "losses = []\n",
    "accuracys = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "Epoch: 1/5, Loss: 0.1612475808514769, Accuracy: 0.8783\n",
      "Epoch 2/5\n",
      "Epoch: 2/5, Loss: 0.1169863535102604, Accuracy: 0.8934\n",
      "Epoch 3/5\n",
      "Epoch: 3/5, Loss: 0.10649549194629053, Accuracy: 0.9004\n",
      "Epoch 4/5\n",
      "Epoch: 4/5, Loss: 0.09574898945529449, Accuracy: 0.9050\n",
      "Epoch 5/5\n",
      "Epoch: 5/5, Loss: 0.07938974082958085, Accuracy: 0.9083\n"
     ]
    }
   ],
   "source": [
    "# Train the model\n",
    "epochs = 5\n",
    "batch_size = 64\n",
    "for epoch in range(epochs):\n",
    "    for i in range(0, len(X_train), batch_size):\n",
    "        X_batch = X_train[i:i + batch_size]\n",
    "        y_batch = y_train[i:i + batch_size]\n",
    "        nn.backward(X_batch, y_batch, nn.relu(X_batch.dot(nn.W1) + nn.b1))\n",
    "    losses.append(np.mean(nn.loss))\n",
    "\n",
    "    # Test the model\n",
    "    y_pred = nn.predict(X_test)\n",
    "    accuracy = np.mean(np.argmax(y_pred, axis=1) == np.argmax(y_test, axis=1))\n",
    "    accuracys.append(accuracy)\n",
    "    print(f\"Epoch: {epoch + 1}/{epochs}, Loss: {np.mean(nn.loss)}, Accuracy: {accuracy:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the training loss for each epoch\n",
    "plt.plot(list(range(1,6)), losses)\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Training Loss')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(1,6), accuracys)\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Training Accuracy/ Epoch')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}