-
Notifications
You must be signed in to change notification settings - Fork 82
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fix/recurrent maddpg with launchpad #121
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good @DriesSmit !
@@ -92,7 +93,7 @@ def make_networks( | |||
observation_network, | |||
snt.Flatten(), | |||
snt.nets.MLP(policy_networks_layer_sizes[key]), | |||
snt.LSTM(20), | |||
snt.LSTM(25), |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Any particular reason for this or just seeing if it improved performance?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This was used to make it easier to debug, as now it is different than the sequence length. This can change back to 20, but I find it easier to see what is going on in the adders when something breaks if the size is not the same as the sequence length.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good, thanks @DriesSmit! 👍
What?
Fix minor bugs to allow for recurrent training of MA-DDPG using Launchpad.
Why?
The bugs broke training if one used MA-DDPG with a recurrent executor.
How?
There were some minor bugs inside the MA-DDPG system, especially with the signatures, which meant incorrect data was being sent through the adders.
Extra
I also updated the transition adder to conform to the signature setup of the sequence adder.