This PR address the following logical bug in get_state. During get_state(complete=False) we should return the last state_size frames from agent.history. Previously we were returning the first agent.state_size frames.
At agent.optimize we get the first agent.state_size frames as state and last agent.state_size frames for next_state
This PR address the following logical bug in get_state. During
get_state(complete=False)we should return the laststate_sizeframes from agent.history. Previously we were returning the firstagent.state_sizeframes.At
agent.optimizewe get the firstagent.state_sizeframes as state and lastagent.state_sizeframes fornext_state