Agent.choose_action: replace the mode string argument with internal self.training #211
Description
Sync. RL baselines by calling agent.train() and agent.eval() explicitly.
{{ message }}
Sync. RL baselines by calling agent.train() and agent.eval() explicitly.