Hello, I am using V-REP and MATLAB for reinforcement learning. The training process runs smoothly, but when I try to run the trained agent with the following code:
%% Play the game with the trained agent
simOptions = rlSimulationOptions('MaxSteps',3000);
experience = sim(env,agent,simOptions);
I encounter an error:
Error using subscript
Subscript arguments for SUBSREF and SUBSASGN cannot be empty.
Error in subscript (line 30)
info = builtin('subsref', self.x__info, S(1:end-1));
Error in trainDDPGWT4 (line 122)
experience = sim(env,agent,simOptions);
The error seems to be occurring in the zmqRemoteApi\clients\matlab\RemoteAPlObject.m file, specifically at line 30. How can I resolve this issue?
Error using subscript Subscript arguments for SUBSREF and SUBSASGN cannot be empty.
Re: Error using subscript Subscript arguments for SUBSREF and SUBSASGN cannot be empty.
Hello, thank you for your reply. I also have a question on how to modify the default simulation speed control and visualization on/off for VRP, as I have found that increasing the simulation speed and visualization off would result in better graphics card usage.
Re: Error using subscript Subscript arguments for SUBSREF and SUBSASGN cannot be empty.
Hello,
various simulation step sizes can be set/adjusted in the simulation dialog. You can turn off rendering during simulation via the appropriate toolbar button (right to the rabbit button).
Cheers
various simulation step sizes can be set/adjusted in the simulation dialog. You can turn off rendering during simulation via the appropriate toolbar button (right to the rabbit button).
Cheers