APL

This is the source code for the paper "Adaptive Policy Learning for Offline-to-Online Reinforcement Learning"(AAAI2023).

Environment

Tested on pytorch1.4, mojuco210, python3.8

How to run

1. Set the agent/engine name in config/exp.yaml, e.g., GCQL
2. Run "python off_on_interleave.py"

Acknowledgement

The code implementation is based on spinningup,rlkit and cql