search

XifengGuo / DEC-keras

Keras implementation for Deep Embedding Clustering (DEC)
MIT License
477 stars 162 forks source link

Can using different autoencoder architectures help improve results? #18

Closed vivektreddy closed 4 years ago

vivektreddy commented 4 years ago

If I am not getting the desired results with my dataset, will modifying the architecture of the autoencoder help? Do I have to keep the embedding layer below a certain size?

XifengGuo commented 4 years ago

@vivektreddy You can choose a better autoencoder structure for your own dataset. The dimension of the embedding layer (10 in this code) is not necessarily the best choice either.