Dynamically change a block input during Deep Reinforcement Learning after each episode. How To?

2 visualizaciones (últimos 30 días)
I have a constant block that takes a single input from the base MATLAB workspace. How do I set up my enviroment such that it accepts one of seven available inputs (at random or in a cycle)? This input should change between episodes.
Thank you.

Respuestas (1)

Emmanouil Tzorakoleftherakis
Emmanouil Tzorakoleftherakis el 12 de Oct. de 2020
Hello,
Please have a look at this example, and in particular the 'localResetFcn' at the bottom ('setBlockParameter'in the reset function is one option).

Categorías

Más información sobre Deep Learning with Simulink en Help Center y File Exchange.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by