subhadarship / kmeans_pytorch

kmeans using PyTorch
https://subhadarship.github.io/kmeans_pytorch
MIT License
490 stars 80 forks source link

Input of the same data leads to different results. #35

Open zchwang opened 2 years ago

zchwang commented 2 years ago

Hi,

I tried to cluster the same data with Kmeans_pytorch, but got different clustering results. What is the cause of this?

Looking forward to your reply.

qbeer commented 1 year ago

Haven't went through the code, but K-means itself is an inherently random algorithm and the outcome depends on the initialization. However, I think for an easily clustering dataset, the difference in the outcomes should be minimal.