No performance gain after reloading policy.

jc-bao / policy-adaptation-survey

This repository is for comparing the prevailing adaptive control method in both control and learning communities.

Apache License 2.0

7 stars 1 forks source link

No performance gain after reloading policy. #6

Closed jc-bao closed 1 year ago

jc-bao commented 1 year ago

Using agent.act.load_state_dict(loaded_agent['actor'].state_dict()) to load policy rather than agent.act = loaded_agent['actor'] where the policy gradient is disabled.