Skip to content

here's code for PCA and it's importance in data reduction to avoid multi-collinearity in dataset.

Notifications You must be signed in to change notification settings

avikumart/Principal-Component-Analysis

Repository files navigation

Principal-Component-Analysis

This notebook contains simple explanation of PCA and it's importance in data reduction to avoid multi-collinearity in given variables of data.

Overview

  • Principal component analysis is a dimention reduction technique that finds the variance maximizing directions onto which to project the data.
  • Linear dimensionality reduction using Singular Value Decomposition of the data to project it to a lower dimensional space. The input data is centered but not scaled for each feature before applying the SVD.
  • Principal components are calculated to reduce variance of features and thus reducing dimentionality of variables in dataset.

image

Table of contents of this repo:

  1. Library imports
  2. Plotting continues numerical features
  3. PCA Object
  4. Plotting explained variance ratio
  5. Plotting explained variance ratio
  6. Plotting cummulative explained variance

About

here's code for PCA and it's importance in data reduction to avoid multi-collinearity in dataset.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published