search

mwhittaker / deeprl_project

Deep RL Final Project
1 stars 1 forks source link

Improve autoencoder training speed. #24

Open vlad17 opened 7 years ago

vlad17 commented 7 years ago

Currently the autoencoder is being trained epoch-by-epoch with feed_dict. Try using a TF Input pipeline and measure perf improvement.

SamvitJ commented 7 years ago

Dependent on #23 Deliverable: time before/time after change