Skip to content

Latest commit

 

History

History

Clustering_and_Dimensionality_Reduction_PCA

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
Graphs
Graphs
 
 
code
code
 
 
README.md
README.md
 
 
Report.pdf
Report.pdf
 
 
lab.pdf
lab.pdf
 
 

README.md

Lab 4 - Clustering and Dimensionality Reduction

Introduction

This project contains 3 parts

  1. K-means clustering on MNIST and experimenting on confusion matrix, misclassifications based on number of clusters

  2. Principal Component Analysis to reduce the dimensionality of the digit images and effect for reconstruction error based on number of princiapl components chosen.

  3. K-means clustering on the data projected onto lower dimensions

  • Please refer to Report.pdf for detailed analysis.
  • Please refer to lab.pdf for details about the project.
  • The code was developed and tested in Matlab 2017a; where in-built K-means clustering was used. If possible, use the same version.

Directory Structure

---README
---lab.pdf
---Report.pdf
---code
	|
	|---data.txt
	|---label.txt
	|---disptable.m
	|---readdata.m
	|---runclustering.m
	|---runpca.m
	|---Q1.m
	|---Q2.m
	|---Q3.m
---Graphs

To Run

Change directory to 'code' folder. In Matlab run scripts for corresponding parts

Q1.m

Q2.m

Q3.m

Graphs generated saved in Graphs

Additional Info

[ predict] = runclustering( rawX, label ,k) [ rawprojX, U, transX ] = runpca( rawX , elim)

Naman Goyal 2015CSB1021