Skip to content

Jax Implementation of Data-regularized Q (DrQ)

Notifications You must be signed in to change notification settings

mantle2048/drq_jax

Repository files navigation

drq_jax

Jax Implementation of Data-regularized Q (DrQ)

(It's my lecture project of Reinforcement Learning :)

How to run?

python drq.py cfg=walker_walk train_seed=0

Performance

new_performance_curve_

Wall clock time

new_time_curve

Disclaimers

Running the code requires ≈38 GB GPU memory.

As I can access large memory GPUs, so I did not implement a memory-efficient replay buffer for image observations.

Leave it for future work (下次一定!)

About

Jax Implementation of Data-regularized Q (DrQ)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages