search

knowledgedefinednetworking / DRL-GNN

BSD 3-Clause "New" or "Revised" License
191 stars 35 forks source link

about the code #12

Closed dontwakemeup closed 1 year ago

dontwakemeup commented 1 year ago

excuse me,could you please tell me what ‘SAP’ and 'RAND' in evaluate_DQN.py means?

paulalmasan commented 1 year ago

The issue is solved in the last commit https://github.com/knowledgedefinednetworking/DRL-GNN/commit/61723d24afe774d9f023dd98fac7166974a13d54

paulalmasan commented 1 year ago

To which dense layer are you referring to? Readout or Message? In any case, in https://github.com/knowledgedefinednetworking/DRL-GNN/blob/61723d24afe774d9f023dd98fac7166974a13d54/DQN/mpnn.py#L11 you can modify the number of neurons on each layer for all NNs. Changing the number of neurons in the last layer of the NNs will indicate the number of values to output for each NN