ReinforcementLearning.jl
ReinforcementLearning.jl copied to clipboard
JuliaReinforcementLearning
A reinforcement learning package for Julia
I built a custom RL environment and tested it with some RL algorithms, namely DDPG, TD3 and SAC. Results were pretty good overall, just SAC somehow fails to converge (though...
I'm creating this issue to pin down that MultiThreadedEnvs do not currently work. It steps each inner env in parallel and stores the traces, but there's nothing that can be...
Quite schockingly, I realised `test/runtests.jl` is empty. What's the current way to test this package ?
Hi, I am trying to work with you Gym env wrapper but when I call GymEnv("HalfCheetah-v3") I have the following error : ```julia ERROR: Gym environment HalfCheetah-v3 not found. Run...
Hi there, in my spare time I've been making a [CleanRL](https://github.com/vwxyzjn/cleanrl) inspired library in julia ([CleanRL.jl](https://github.com/sash-a/CleanRL.jl)) and of course I've been using a fair amount of this library, however I...
`MultiThreadEnv` has a property called `terminals` which is a `BitArray{1}` containing the termination information for all of the wrapped environments. However, the `is_terminated` function behaves in an unexpected way: ```julia...
This fixes what appears to be a copy/paste bug that affects multi-threaded envs where the state space of the wrapped environment has different dimensionality than the action space of that...
The number of 'unmaintained' policies and significant subject-matter knowledge required to implement and verify correctness of many of the policies is making further progress on the v0.11 refactor difficult. I...
Hello @jeremiahpslewis @findmyway and any other who wishes. The upcoming 1.9 version will introduce the package extensions feature, which RL.jl could greatly make use of. I figured it would be...
Metadata
Owner
Metadata
A reinforcement learning package for Julia