Why I keep receiving this message: "For deep learning, the simulation target language must be set to C++"?

Question

Ismaeel 2021-9-26

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1460819-why-i-keep-receiving-this-message-for-deep-learning-the-simulation-target-language-must-be-set-to

评论： Walter Roberson 2022-7-25

在 MATLAB Online 中打开

I am using MATLAB R2021b (the latest version). This error also occured in the previous version.

I replaced the "RL Agent" block with a "MATLAB Function" block containing the policy generated for my controller.

I keep receiving this message even after I changed the taget language from "C" to 'C++".

For deep learning, the simulation target language must be set to C++. Function 'MATLAB Function' (#66.286.341), line 13, column 11: "coder.loadDeepLearningNetwork('agentData.mat','policy')" Launch diagnostic report.

Component:MATLAB Function | Category:Coder error

Persistent variable 'policy' must be assigned before it is used. The only exception is a check using 'isempty(policy)' that can be performed prior to assignment. Function 'MATLAB Function' (#66.398.404), line 16, column 19: "policy" Launch diagnostic report.

Component:MATLAB Function | Category:Coder error

Function call failed. Function 'MATLAB Function' (#66.140.167), line 7, column 11: "localEvaluate(observation1)" Launch diagnostic report.

Component:MATLAB Function | Category:Coder error

Errors occurred during parsing of MATLAB function 'IntegratedSys/MATLAB Function'

Component:MATLAB Function | Category:Coder error

Errors occurred during parsing of MATLAB function 'IntegratedSys/MATLAB Function'

This is the code inside the "MATLAB Function" blcok (which I generated it using the function "generatePolicyFunction"):

function action1 = evaluatePolicy(observation1)
%#codegen
% Reinforcement Learning Toolbox
% Generated on: 26-Sep-2021 07:01:45
action1 = localEvaluate(observation1);
end
%% Local Functions
function action1 = localEvaluate(observation1)
persistent policy
if isempty(policy)
	policy = coder.loadDeepLearningNetwork('agentData.mat','policy');
end
observation1 = observation1(:)';
action1 = predict(policy, observation1);
end

3 个评论
显示 1更早的评论隐藏 1更早的评论

Ismaeel 2021-9-26

Thank you Walter.

I am not professional in this filed and this term is new to me. I was wondering if there is a suported version of it or anything else that solves the problem.

I am also wondering why the message is not clear to the user. I followed what the message tells me and changed the setting accordingly, however, nothing changed.