Deploy Trained Reinforcement Learning Policy as Microservice Docker Image
To deploy a trained RL policy as a microservice Docker® image, follow three steps.
Package a MATLAB® function that evaluates a reinforcement learning policy into a deployable archive.
Create a Docker image that contains the archive and a minimal MATLAB Runtime package.
Run the image in Docker and make calls to the service using any of the MATLAB Production Server™ client APIs.
For an example on how to do this, see Deploy Trained Reinforcement Learning Policy as Microservice Docker Image (MATLAB Compiler SDK).
See Also
Functions
generatePolicyFunction|generatePolicyBlock|policyParameters|updatePolicyParameters|train|trainFromData
