Reinforcement-learning-with-tensorflow icon indicating copy to clipboard operation
Reinforcement-learning-with-tensorflow copied to clipboard

Published 20 hours ago • verified publisher icon MorvanZhou

Save and Reuse of DDPG model

Open lyjge opened this issue 6 years ago • 1 comments

Hello Dr. MorvanZhou,

Thanks for sharing this great repository.

May I ask you 2 simple (stupid, as I am a rookie of RL~lol) questions:

  1. How can I save the trained model of your DDPG_update2.py code.

  2. Then how can I load and reuse the saved model to get action and the new state (when doing this the model is not going to be trained).

Many Thanks, YL

lyjge avatar May 03 '19 22:05 lyjge

Hi lyjge,

I acctually got that setup for the DDPG Code, both. If you could figure how to add batch normalization in the Actor network between L1 and L2 i'd be happy to share, so we could help each other?

Jan

ghost avatar May 04 '19 16:05 ghost