search

greenelab / tybalt

Training and evaluating a variational autoencoder for pan-cancer gene expression data
BSD 3-Clause "New" or "Revised" License
164 stars 62 forks source link

Implement Warmup #12

Closed gwaybio closed 7 years ago

gwaybio commented 7 years ago

As discussed in Sonderby et al. 2016 a warm up period from deterministic to stochastic autoencoder model as training epochs proceed may enhance parameter usage.

This stackoverflow post may be useful to review.

gwaybio commented 7 years ago

addressed by #14