fix: shared weights with agent type #428
Conversation
arnupretorius
left a comment
There was a problem hiding this comment.
Thanks @AsadJeewa! 😄
Just see my few comments. Also, did we test this? At the very least we should do some runs on environments with and without different agent types.
|
I tested that everything works in multiple different environments with different (openspiel Tic Tac Toe, Petting Zoo Pong) or single-agent types (debugging, Petting Zoo Multiwalker, Flatland) to make sure that the networks were being assigned correctly and that execution proceeds. Since shared_weights default value is True, I think that we should run a small set of benchmarks @RuanJohn |
mmorris44
left a comment
There was a problem hiding this comment.
- 1 potential issue in 3 places
- 1 debug print statement
- 1 minor comment
|
This line (https://github.com/instadeepai/Mava/blob/develop/mava/specs.py#L68) assumes that all environments will follow the naming convention of |
mmorris44
left a comment
There was a problem hiding this comment.
PR looks good to go from my side.
Thanks @AsadJeewa!
arnupretorius
left a comment
There was a problem hiding this comment.
Thanks @AsadJeewa! 👍
What?
Updated agent network types to only share weights between agent types
#427
Why?
When setting shared_weights to true, a single network is created for all agents. It should not be possible to share weights across agent types
How?
changed self._agent_net_keys in systems.py to assign the correct keys (for MADQN, MAPPO, MADDPG)
Extra
If we are happy with this change, would we then need to benchmark?