{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "from matplotlib import pyplot as plt\n", "from scipy.cluster.hierarchy import dendrogram, linkage\n", "import numpy as np" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ ":36: RuntimeWarning: covariance is not positive-semidefinite.\n", " datasetA = np.random.multivariate_normal(muA, covarianceA, size= sizeA)\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Data generation\n", "# if we want to repeat our results\n", "np.random.seed(1234) \n", "\n", "# First we set mean for datasetA and datasetB\n", "muA = [-5,-5]\n", "muB = [5,5]\n", "muC = [-5,10]\n", "\n", "# Then we set standard deviation for datasetA and datasetB\n", "sigmaAx = 10.0\n", "sigmaAy = 3.0\n", "# covariance\n", "sigmaAxy = -10.0\n", "# covariance matrix\n", "covarianceA = [[sigmaAx, sigmaAxy], [sigmaAxy, sigmaAy]]\n", "\n", "sigmaBx = 3.0\n", "sigmaBy = 3.0\n", "# covariance \n", "sigmaBxy = 0.0\n", "# covariance matrix\n", "covarianceB = [[sigmaBx, sigmaBxy], [sigmaBxy, sigmaBy]]\n", "\n", "sigmaCx = 2.0\n", "sigmaCy = 2.0\n", "# covariance \n", "sigmaCxy = 0.0\n", "# covariance matrix\n", "covarianceC = [[sigmaCx, sigmaCxy], [sigmaCxy, sigmaCy]]\n", "\n", "sizeA = 30\n", "sizeB = 30\n", "sizeC = 30\n", "\n", "datasetA = np.random.multivariate_normal(muA, covarianceA, size= sizeA)\n", "datasetB = np.random.multivariate_normal(muB, covarianceB, size= sizeB)\n", "datasetC = np.random.multivariate_normal(muC, covarianceC, size= sizeC)\n", "\n", "labelsA = [\"A\" for i in range(sizeA)]\n", "labelsB = [\"B\" for i in range(sizeB)]\n", "labelsC = [\"C\" for i in range(sizeC)]\n", "labelsABC = labelsA + labelsB + labelsC\n", "\n", "dataT = np.concatenate((datasetA, datasetB, datasetC),)\n", "\n", "plt.scatter(dataT[:,0], dataT[:,1])\n", "plt.axis('equal')\n", "plt.show()\n", "\n" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "# Compute linkage matrix\n", "# Read https://docs.scipy.org/doc/scipy/reference/generated/scipy.cluster.hierarchy.linkage.html\n", "# to learn other options\n", "\n", "LinkageSC = linkage(dataT, method='single', metric='cosine', optimal_ordering=False)\n", "LinkageAC = linkage(dataT, method='average', metric='cosine', optimal_ordering=False)\n", "LinkageCC = linkage(dataT, method='complete', metric='cosine', optimal_ordering=False)" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Plot dendrogram\n", "# Read https://docs.scipy.org/doc/scipy-0.14.0/reference/generated/scipy.cluster.hierarchy.dendrogram.html\n", "# to learn other options\n", "\n", "plt.figure(figsize=(25, 10))\n", "plt.title('Hierarchical Clustering Dendrogram - Cosine, single')\n", "plt.xlabel('Index of the samples')\n", "plt.ylabel('Distance')\n", "SE = dendrogram(LinkageSC,labels=labelsABC,leaf_rotation=0.0,leaf_font_size=9.0)\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Plot dendrogram\n", "# Read https://docs.scipy.org/doc/scipy-0.14.0/reference/generated/scipy.cluster.hierarchy.dendrogram.html\n", "# to learn other options\n", "\n", "plt.figure(figsize=(25, 10))\n", "plt.title('Hierarchical Clustering Dendrogram - Cosine, average')\n", "plt.xlabel('Index of the samples')\n", "plt.ylabel('Distance')\n", "AE = dendrogram(LinkageAC,labels=labelsABC,leaf_rotation=0.0,leaf_font_size=9.0)\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Plot dendrogram\n", "# Read https://docs.scipy.org/doc/scipy-0.14.0/reference/generated/scipy.cluster.hierarchy.dendrogram.html\n", "# to learn other options\n", "\n", "plt.figure(figsize=(25, 10))\n", "plt.title('Hierarchical Clustering Dendrogram - Cosine, complete')\n", "plt.xlabel('Index of the samples')\n", "plt.ylabel('Distance')\n", "CE = dendrogram(LinkageCC,labels=labelsABC,leaf_rotation=0.0,leaf_font_size=9.0)\n", "plt.show()" ] } ], "metadata": { "celltoolbar": "Slideshow", "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": "3.8.5" }, "latex_envs": { "LaTeX_envs_menu_present": true, "bibliofile": "biblio.bib", "cite_by": "apalike", "current_citInitial": 1, "eqLabelWithNumbers": true, "eqNumInitial": 1, "hotkeys": { "equation": "Ctrl-E", "itemize": "Ctrl-I" }, "labels_anchors": false, "latex_user_defs": false, "report_style_numbering": false, "user_envs_cfg": false }, "toc": { "colors": { "hover_highlight": "#DAA520", "running_highlight": "#FF0000", "selected_highlight": "#FFD700" }, "moveMenuLeft": true, "nav_menu": { "height": "12px", "width": "252px" }, "navigate_menu": true, "number_sections": true, "sideBar": true, "threshold": 4, "toc_cell": false, "toc_section_display": "block", "toc_window_display": false } }, "nbformat": 4, "nbformat_minor": 2 }