14 Artificial_Neural_Network.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c0c43f9b-0cd9-453f-9bfa-cda475c994f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "pip install seaborn"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "04640efc-20e2-4728-9b29-230e54590d29",
   "metadata": {
    "id": "04640efc-20e2-4728-9b29-230e54590d29"
   },
   "source": [
    "## Experiment 14\n",
    "### Build an Artificial Neural Network using Backpropagation algorithm and test the same with appropriate dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "b127ac3e-3ac3-4072-8395-7fd0bb61a099",
   "metadata": {
    "id": "b127ac3e-3ac3-4072-8395-7fd0bb61a099",
    "outputId": "37632508-a099-4092-8b58-4f8ffbd406e2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix:\n",
      "[[23  0  0]\n",
      " [ 1 18  0]\n",
      " [ 0  0 12]]\n",
      "Accuracy: 98.15%\n",
      "Wine types: ['class_0' 'class_1' 'class_2']\n",
      "Data prediction: class_0\n"
     ]
    }
   ],
   "source": [
    "# Import necessary libraries\n",
    "from sklearn.datasets import load_wine\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn import metrics\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix\n",
    "\n",
    "# Load the Wine dataset\n",
    "wine = load_wine()\n",
    "\n",
    "# Features (chemical properties of wines) and target (wine class)\n",
    "X = wine.data\n",
    "y = wine.target\n",
    "\n",
    "# Standardize the features to have mean=0 and variance=1\n",
    "scaler = StandardScaler()\n",
    "x_scaled = scaler.fit_transform(X)\n",
    "\n",
    "# Split the dataset into training and testing sets (70% train, 30% test)\n",
    "X_train, X_test, y_train, y_test = train_test_split(x_scaled, y, test_size=0.3, random_state=1)\n",
    "\n",
    "# Create an instance of MLPClassifier with specified parameters\n",
    "mlp = MLPClassifier(random_state=1, max_iter=1000, hidden_layer_sizes=(100,), solver='lbfgs', batch_size=50)\n",
    "\n",
    "# Fit the model on the training data\n",
    "mlp.fit(X_train, y_train)\n",
    "\n",
    "# Make predictions on the test set\n",
    "y_pred = mlp.predict(X_test)\n",
    "\n",
    "# Compute the confusion matrix to evaluate the predictions\n",
    "confusion_mtx = confusion_matrix(y_test, y_pred)\n",
    "print(\"Confusion Matrix:\")\n",
    "print(confusion_mtx)\n",
    "\n",
    "# Calculate and print the accuracy of the model\n",
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "print(f\"Accuracy: {accuracy * 100:.2f}%\")  # Format accuracy to two decimal places\n",
    "\n",
    "# Example test data for making a prediction\n",
    "test = np.array([[14.37, 1.95, 2.5, 16.8, 113, 3.85, 3.49, 0.24, 2.18, 7.8, 0.86, 3.45, 1480]])\n",
    "\n",
    "# Standardize the test data using the same scaler (important for consistency)\n",
    "test_scaled = scaler.transform(test)\n",
    "\n",
    "# Predict the class of the new sample\n",
    "prediction = mlp.predict(test_scaled)\n",
    "\n",
    "# Print the target names and the predicted class for the new sample\n",
    "print(f\"Wine types: {wine.target_names}\")\n",
    "print(f\"Data prediction: {wine.target_names[prediction][0]}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13f77e9c-7a16-4c75-ab6a-29059c50c854",
   "metadata": {},
   "source": [
    "# Following cells are not Required for the Experiment"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "abde38ea-0285-492b-aed5-d181adc43e13",
   "metadata": {},
   "source": [
    "## Confusion Matrix Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "66c8830f-db79-4cab-bc7a-fcca757edeb2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "sns.heatmap(confusion_mtx, annot=True, cmap=\"YlGnBu\", fmt=\"d\",\n",
    "            xticklabels=wine.target_names, yticklabels=wine.target_names)\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('Actual')\n",
    "plt.title('Confusion Matrix')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2910496-507f-4cff-84d0-843e55ebac23",
   "metadata": {},
   "source": [
    "## ROC Curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d9246f20-ccb5-4452-9c49-38b4e2b13e96",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve, auc\n",
    "from sklearn.preprocessing import label_binarize\n",
    "from sklearn.metrics import RocCurveDisplay\n",
    "\n",
    "# Binarize the output\n",
    "y_test_bin = label_binarize(y_test, classes=[0, 1, 2])\n",
    "y_score = mlp.predict_proba(X_test)\n",
    "\n",
    "plt.figure(figsize=(10, 8))\n",
    "for i in range(3):\n",
    "    fpr, tpr, _ = roc_curve(y_test_bin[:, i], y_score[:, i])\n",
    "    roc_auc = auc(fpr, tpr)\n",
    "    plt.plot(fpr, tpr, label=f'ROC curve for {wine.target_names[i]} (area = {roc_auc:.2f})')\n",
    "\n",
    "plt.plot([0, 1], [0, 1], 'k--')  # Diagonal line\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic (ROC) - Multi-Class')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb94b746-71a3-4afb-8b3b-bef8bb27dfac",
   "metadata": {},
   "source": [
    "## Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7671bb89-41f0-4539-b889-7b08c89bdaa4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     class_0       0.96      1.00      0.98        23\n",
      "     class_1       1.00      0.95      0.97        19\n",
      "     class_2       1.00      1.00      1.00        12\n",
      "\n",
      "    accuracy                           0.98        54\n",
      "   macro avg       0.99      0.98      0.98        54\n",
      "weighted avg       0.98      0.98      0.98        54\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "\n",
    "report = classification_report(y_test, y_pred, target_names=wine.target_names)\n",
    "print(\"Classification Report:\")\n",
    "print(report)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1bc830ab-e13d-40cc-b6f0-e16565d8c929",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}