Commit b834f5e5 authored by lucas_miranda's avatar lucas_miranda
Browse files

Implemented weight saving callback in model_training.py

parent 8db9836c
......@@ -434,6 +434,12 @@ class SEQ_2_SEQ_GMVAE:
)
# TODO:
# - latent space metrics to control overregulatization (turned off dimensions). Useful for warmup tuning
# - Clustering metrics for model selection and aid training (eg early stopping)
# - Silhouette / likelihood / classifier accuracy metrics
# - design clustering-conscious hyperparameter tuing pipeline
# TODO (in the non-immediate future):
# - Try Bayesian nets!
# - MCMC sampling (n>1) (already suported by tfp! we should try it)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment