sample-factory
tianshou
sample-factory | tianshou | |
---|---|---|
6 | 8 | |
743 | 7,435 | |
- | 1.3% | |
7.9 | 9.5 | |
5 days ago | about 16 hours ago | |
Python | Python | |
MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
sample-factory
-
A minimal RL library for infinite horizon tasks
I take a lot of inspiration from Sample Factory and RLlib for my own RL library's implementation. Although I thoroughly enjoy both of these libraries, they just didn't quite fit right with my use case which motivated me to start my own. Hopefully someone finds use in rlstack whether it be through direct usage or as inspiration for their own personalized library
-
Fast and hackable frameworks for RL research
I'm tired of having my 200m frames of Atari take 5 days to run with dopamine, so I'm looking for another framework to use. I haven't been able to find one that's fast and hackable, preferably distributed or with vectorized environments. Anybody have suggestions? seed-rl seems promising but is archived (and in TF2). sample-factory seems super fast but to the best of my knowledge doesn't work with replay buffers. I've been trying to get acme working but documentation is sparse and many of the features are broken.
-
Multi-agent Decentralized Training with a PettingZoo environment
Hi, try sample-factory
-
How is IMPALA as a framework?
Sample Factory: https://github.com/alex-petrenko/sample-factory
-
The Myth of a Superhuman AI
Everything in this reply is wrong.
In AlphaZero for example, there were 44 million training games total for 700,00x0 steps of training for the full 9 hours.
Turning that human-like numbers, 44million games with on average 60 moves, at 1 second thinking time per move,
> 44000000*60/60/60/24/365 = 83,7138508371 years of training experience in 9 hours
The whole field of Reinforcement learning has agents training and playing games for many orders of magnitude more time than a human ever will. In-fact, we can scale this to over 100k of actions per second, in a single machine:
https://github.com/alex-petrenko/sample-factory
Then, there is also distributed Reinforcement Learning, where hundreds of agents can play at different machines and share experience, see AlphaZero, LeelaZero, R2D2 agent, R2D3 agent, Apex, Acer, Asynchronous PPO.
> but the data isn't useful without the context of experience
The experience is the data in Reinforcement Learning.
> and all processing power can do it overfit model without experience.
That is wrong, the agents perform what is called exploration to avoid getting stuck in simple strategies.
> Even if we put AI into an army of robots running around and experiencing things, there are still scaling limits to encoding and communicating knowledge and understanding.
True, but machines scale better because they speak the same language, or they can learn to tune their language to get their message across.
> Human organizations are a great example of the scaling limits of intelligence.
Human organization is a testament to how far we can get with something as limiting as the commonly used language. The language that we use to communicate is subject to misinterpretation due to our subjective experiences, this limitation is not shared by machines.
-
Best PyTorch RL library for doing research
I borrow a lot of performance tricks from sample factory, which is awesome but hard to modify from its original APPO algorithm. rlpyt was more modular, and I borrowed more ideas from it (namedarraytuple), but still too limited.
tianshou
-
Is it better to not use the Target Update Frequency in Double DQN or depends on the application?
The tianshou implementation I found at https://github.com/thu-ml/tianshou/blob/master/tianshou/policy/modelfree/dqn.py is DQN by default.
- 他們能回來嗎
-
Multi-Agent Stable Baselines
https://github.com/thu-ml/tianshou Imho there isn't a library that has it all, RLlib is quite good too, but I think that Tianshou is more similar to Pytorch and that helps to change the internals more intuitively and know what you are doing.
-
Question about the old policy and new policy in TRPO code
Good point...I'll check in more detail when I get a chance later today! I would suggest looking at a more recent implementation like https://github.com/DLR-RM/stable-baselines3 or https://github.com/thu-ml/tianshou if you're trying to build. https://spinningup.openai.com/en/latest/algorithms/trpo.html is particularly good for understanding
-
Tensorflow vs PyTorch for A3C
Do you absolutely need A3C? A2C has become more widely used (see, e.g., the comment in https://github.com/ikostrikov/pytorch-a3c, and the fact that both https://github.com/thu-ml/tianshou and https://github.com/facebookresearch/salina have A2C implementations, but no A3C at first glance).
-
"Tianshou: a Highly Modularized Deep Reinforcement Learning Library", Weng et al 2021 (Python PyTorch MuJuCo; PPO, DQN, A2C, DDPG, SAC, TD3, REINFORCE, NPG, TRPO, ACKTR)
Code for https://arxiv.org/abs/2107.14171 found: https://github.com/thu-ml/tianshou/
Get the code for Tianshou here (GitHub).
-
Best PyTorch RL library for doing research
I tried tianshou and thought it was well-designed for modularity, but it was early in development when I tried and missing some basic features
What are some alternatives?
cleanrl - High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)
stable-baselines3 - PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
machin - Reinforcement learning library(framework) designed for PyTorch, implements DQN, DDPG, A2C, PPO, SAC, MADDPG, A3C, APEX, IMPALA ...
ElegantRL - Massively Parallel Deep Reinforcement Learning. 🔥
rl8 - A high throughput, end-to-end RL library for infinite horizon tasks.
pytorch-learn-reinforcement-learning - A collection of various RL algorithms like policy gradients, DQN and PPO. The goal of this repo will be to make it a go-to resource for learning about RL. How to visualize, debug and solve RL problems. I've additionally included playground.py for learning more about OpenAI gym, etc.
rlpyt - Reinforcement Learning in PyTorch
seed_rl - SEED RL: Scalable and Efficient Deep-RL with Accelerated Central Inference. Implements IMPALA and R2D2 algorithms in TF2 with SEED's architecture.
torchbeast - A PyTorch Platform for Distributed RL
pytorch-a3c - PyTorch implementation of Asynchronous Advantage Actor Critic (A3C) from "Asynchronous Methods for Deep Reinforcement Learning".