Incorrect plot for the same code

saksham_thukral · April 20, 2020, 10:22am

i Have replicated the same code as implemented in the video, then i got the wrong plot but same error was encountered in the video as well , but in video the error that came at 5:48 was resolved by just removing the color property , and infact after rentering the same property the plot was correct, but in my case i have tried everything but i am not getting the colored plot at all, i even re-ran the code of video by taking github , but it was still not giving the colored plot. The irony is that for exactly same code i am not getting the colored plot at all plz help!

S18CRX0120 · April 20, 2020, 10:42am

Hey @saksham_thukral, match your code line by line from here, if you didn’t find any mistake than try to run this code only, and check it this works fine or not.

github.com

coding-blocks-archives/machine-learning-online-2018/blob/master/10. K-Means and Color Extraction/K-Means Clustering.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# K-Means\n",
    "- Common clustering technique, helps to solve many problems efficiently.\n",
    "- K-Means is an example of Hard Clustering, where every point belongs only to one cluster.\n",
    "\n",
    "## Motivation Problem - *Pizza Parlour Study*\n",
    "\n",
    "**Problem Statement** Given the list of locations of customers who frequently order Pizza we want to find out optimal locations of Pizza Parlours where they should be opened.\n",
    "\n",
    "<img src=\"pizza1.png\" alt=\"Pizza-1\" style=\"width: 400px;\"/>\n",
    "\n",
    "**Step-1** : Initialise Pizza Parlour Centers randomly.\n",
    "<img src=\"pizza2.png\" alt=\"Pizza-2\" style=\"width: 400px;\"/>\n",
    "\n",
    "**Step-2** : Assign each customer to its nearest Pizza Center.\n",

This file has been truncated. show original

Hope this resolved your doubt.
Plz mark the doubt as resolved in my doubts section.

sushil_d_loser · May 26, 2020, 6:08am

I think there is a bug in the code here. Before calling the function plotClusters(clusters) in In(85), we must call the function assignPointToClusters(clusters) first. Otherwise the points of datasets will not be displayed, i mean the same error will occur as 5.12 in the K-Means implementing M-step video.