Skip to content

umd-huang-lab/PDML

 
 

Repository files navigation

Overview

Code of ICML 2023 paper: "Live in the Moment: Learning Dynamics Model Adapted to Evolving Policy"

Usage

MBPO

CUDA_VISIBLE_DEVICES=0 python main_pmac.py --env_name 'Humanoid-v2' --num_epoch 400 --exp_name humanoid_mbpo --seed 6

PDML-MBPO

CUDA_VISIBLE_DEVICES=0 python main_pmac_tvd.py --env_name 'Humanoid-v2' --num_epoch 400 --exp_name humanoid_pdml --seed 4 --reweight_model TV --reweight_rollout TV

code structure

During training, 'exp' folder will created aside 'PDML' folder.

Dependencies

MuJoCo 1.5 & MuJoCo 2.0

Reference

This code is built on a pytorch implementation MBPO: https://github.com/Xingyu-Lin/mbpo_pytorch

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%