Skip to content

parv-nangalia/MLClassify_StarsQuasars

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Classifiaction-between-stars-Quasars

This is a simple ML introductory project aimed to showcase and familiarize you with the working of Logitic regression model.

Our aim was to use the given dataset and classify an entity as either a star or quasar using LR model and compare its accuracy with in-built LR classifier.

followed the basic ML approach i.e

  • Data gathering
  • Data cleaning/Manipulation
  • Train model
  • Test data
  • improve using feedback

Used four categories of dataset - 1) North galactic. 2) Equatorial. 3) Both (label and unlabelled).

Intially started with selecting the correct dataset with higher inferential parameters so as to add in classification accuracy. Then proceeded with data cleaning and normalization so as to reuce noise and outlier which results in lower accuracy. Based on different combination of u,g,r,i,z pararmeters zeroed in on the approximate activation function for the classifier which was yielding highest accuracy.

Finally on comparison, model performed upto the mark against the in-built model with the accuracy of 92%.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published