zhampel / clusterGAN

Pytorch Implementation of ClusterGAN (arXiv:1809.03627)
MIT License
95 stars 22 forks source link

Clustering during infernce #4

Open EmreOzkose opened 3 years ago

EmreOzkose commented 3 years ago

Hi, thank you for sharing code.

I can't understand the final part of the pipeline exactly. We train a GAN to be able to extract a representation for each image. After that, we can visualize images using these encodings via TSNE . However how can we seperate data into K groups? For instance, we can run K-means over these representations, right?