Skip to content

Implementation of AutoEncoder in PyTorch for k-Means Clustering

Notifications You must be signed in to change notification settings

tschechlovdev/AutoEncoder_KMeans

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 

Repository files navigation

Implementation of Auto-Encoders for k-Means Clustering

This is the implementation of my medium article "Deep Auto-Encoders for Clustering: Understanding and Implementing in PyTorch".

This repository showcases how to apply Auto-Encoders to learn a lower-dimensional and easier separable presentation of the data. This is especially useful for tasks like clustering. To this end, the well-known MNIST dataset that comprises images of handwritten digits is used. The Auto-Encoder is implemented using PyTorch and scikit-learn is used to apply clustering.

The notebook shows that the accuracy of k-Means clustering can be improved by more than 20%-points using Auto-Encoders!

About

Implementation of AutoEncoder in PyTorch for k-Means Clustering

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published