How do I solve this error?

I am getting this error when I try to train a TD3 RL agent.

Thanking You

Apoorv Pandey

[Emmanouil Tzorakoleftherakis]

If you share a reproduction model it would be easier to debug

Answer 1

[Cris LaPierre]

When defining your rlQValueFunction, include the ActionInputNames and OvservationInputNames name-value pairs.

See this example: https://www.mathworks.com/help/reinforcement-learning/ref/rl.function.rlqvaluefunction.html#mw_da4065e4-5b9a-41c6-b11b-6692d8698a76

% Observation path layers
obsPath = [featureInputLayer( ...
                prod(obsInfo.Dimension), ...
                Name="netObsInput") 
           fullyConnectedLayer(16)
           reluLayer
           fullyConnectedLayer(5,Name="obsout")];

% Action path layers
actPath = [featureInputLayer( ...
                prod(actInfo.Dimension), ...
                Name="netActInput") 
           fullyConnectedLayer(16)
           reluLayer
           fullyConnectedLayer(5,Name="actout")];

%<snip>
    
critic = rlQValueFunction(net,...
     obsInfo,actInfo, ...
     ObservationInputNames="netObsInput",...
     ActionInputNames="netActInput")

[Apoorv Pandey]

SAC

I have used the exact same code as mentioned in the link and still getting the error. Please help

[Cris LaPierre]

Please share your data and your code. You can attach files using the paperclip icon. If it's easier,save your workspace variables to a mat file and attach that.