How do initialize ddpg with a pretrained policy.

[jasonzliang]

Hi, If I have a policy (a tensorflow model) that has been trained using some other RL method, is there way for me to initalize ddpg with this policy?

[ctmakro]

i don't think there is; ddpg needs both actor and critic to work together. You need a replacement for each.

2017-11-04 9:30 GMT+08:00 Jason Liang [email protected]:

Hi, If I have a policy (a tensorflow model) that has been trained using some other RL method, is there way for me to initalize ddpg with this policy?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/ctmakro/stanford-osrl/issues/2, or mute the thread https://github.com/notifications/unsubscribe-auth/AJgudK2EISMULH0A2ueVExe99_JKiXxKks5sy75KgaJpZM4QR2B0 .