cleanrl
cleanrl copied to clipboard
vwxyzjn
High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)
Removing the unused `noise_clip` from `sac_continuous_actions.py` (probably a relic from TD3)
## Problem Description ## Checklist - [x] I have installed dependencies via `poetry install` (see [CleanRL's installation guideline](https://docs.cleanrl.dev/get-started/installation/). - [x] I have checked that there is no similar [issue](https://github.com/vwxyzjn/cleanrl/issues) in...
## Problem Description ## Checklist - [x] I have installed dependencies via `poetry install` (see [CleanRL's installation guideline](https://docs.cleanrl.dev/get-started/installation/). - [x] I have checked that there is no similar [issue](https://github.com/vwxyzjn/cleanrl/issues) in...
## Problem Description Docker image that CleanRL is built on is out of date as per [this issue](https://github.com/NVIDIA/nvidia-docker/issues/1735) ## Checklist - [x] I have installed dependencies via `poetry install` (see...
## Problem Description I'm trying to train different models using CleanRL with a custom PyBullet environment. I couldn't use PyBullet directly due to a conflict with Gymnasium, so I installed...
## Problem Description The pytorch CUDA-enabled libraries are more capable than the CPU-only one. They can also run on the CPU if no CUDA device is available. However, due to...
## Description ## Types of changes - [ ] Bug fix - [ ] New feature - [ ] New algorithm - [ ] Documentation ## Checklist: - [ ]...
Why doesn't SAC and DDPG support multi-environment? I can try to implement it if needed.
## Problem Description I propose to add a new algorithm : 'Munchausen Reinforcement Learning' [Paper link](https://arxiv.org/pdf/2007.14430) ## Checklist - [x] I have installed dependencies via `poetry install` (see [CleanRL's installation...
## Problem Description I would like to implement the CrossQ algorithm as proposed in https://openreview.net/pdf?id=PczQtTsTIX. I already did some testing and it looks like its not that much effort to...
