{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Dimensionality Reduction\n", "\n", "In this lab, we will be implementing Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) using Scikit-learn.\n", "\n", "## Part 1: Feature Reduction using PCA" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "# Import the standard modules to be used in this lab\n", "import pandas as pd\n", "import numpy as np\n", "\n", "%matplotlib inline" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Iris Flower dataset is available in sklearn. So we have to import the dataset into IPython and store it in a variable called iris. Then, convert the data into a Pandas DataFrame and set the feature names" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "from sklearn.datasets import load_iris\n", "iris = load_iris()\n", "X = pd.DataFrame(iris.data)\n", "X.columns = ['sepal_length', 'sepal_width', 'petal_length', 'petal_width']\n", "y = iris.target" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Display the scatter plot of the data." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "from matplotlib import pyplot as plt\n", "plt.figure(figsize=(8,10))\n", "plt.subplot(211)\n", "plt.scatter(X.petal_length,X.petal_width, c=y)\n", "plt.xlabel('petal length')\n", "plt.ylabel('petal width')\n", "plt.subplot(212)\n", "plt.scatter(X.sepal_length,X.sepal_width, c=y)\n", "plt.xlabel('sepal length')\n", "plt.ylabel('sepal width')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now, let's reduce the number of dimension using PCA. We are going to import PCA from Scikit-learn and then create an object of PCA to transform the data. We will be using attribute petal length and petal width only." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Part 2: Feature Reduction using LDA\n", "To reduce the number of dimension using LDA, import LinearDiscriminantAnalysis from Scikit-learn and then create an object of LinearDiscriminantAnalysis." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "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": "3.6.12" }, "varInspector": { "cols": { "lenName": 16, "lenType": 16, "lenVar": 40 }, "kernels_config": { "python": { "delete_cmd_postfix": "", "delete_cmd_prefix": "del ", "library": "var_list.py", "varRefreshCmd": "print(var_dic_list())" }, "r": { "delete_cmd_postfix": ") ", "delete_cmd_prefix": "rm(", "library": "var_list.r", "varRefreshCmd": "cat(var_dic_list()) " } }, "types_to_exclude": [ "module", "function", "builtin_function_or_method", "instance", "_Feature" ], "window_display": false } }, "nbformat": 4, "nbformat_minor": 2 }