Trying to load previously saved agent fails (Reinforcement Learning Toolbox)

Hello,

I have problems to load an agent, that I have trained before. When I try to open the preconfigured examples from Mathworks (e.g. the Watertank model) I can load the existing agent and everything works fine. However, when I train the model first, save the agent and then load this agent I get a warning saying: "Warning: Variable 'agent' not found."

This happens also for other models and I don´t understand why... Can somebody help me out on this?

Answer 1

[Jooo]

Just managed to solve this problem by myself.

In case anyone else has the same problem:

In order to load your saved agent you have to use the following code snippet:

load('Agent123.mat','saved_agent');
agent = saved_agent;

[Wenkai Wu]

Solved, thanks a lot

[beni hadi]

This was my problem, thanks a lot.

[Chao Wang]

• Wouldn't that work if it was

load(' agent123.mat ')

[Kushal Gowda N]

I get this error of saved_agent

Please help