Ray VS stable-baselines

22. Ray | Github | tutorial

Training times for GSM8k are mentioned here: https://github.com/ray-project/ray/tree/master/doc/source/te...

Here's a reference from RLlib that points to docs and an example, and here's one from one of my projects that includes all my own implementations

RLlib

⦿ AWS Glue Data Quality - Automatic data quality rule recommendations based on your data AWS Glue for Ray - Data integration with Ray (ray.io), a popular new open-source compute framework that helps you scale Python workloads

https://ray.io (just dropping the link)

I've wondered whether it's easier to add data analyst stuff to Elixir that Python seems to have, or add features to Python that Erlang (and by extension Elixir) provides out of the box.

By what I can see, if you want multiprocessing on Python in an easier way (let's say running async), you have to use something like ray core[0], then if you want multiple machines you need redis(?). Elixir/Erlang supports this out of the box.

Explorer[1] is an interesting approach, where it uses Rust via Rustler (Elixir library to call Rust code) and uses Polars as its dataframe library. I think Rustler needs to be reworked for this usecase, as it can be slow to return data. I made initial improvements which drastically improves encoding (https://github.com/elixir-nx/explorer/pull/282 and https://github.com/elixir-nx/explorer/pull/286, tldr 20+ seconds down to 3).

[0] https://github.com/ray-project/ray

Adding these for anyone who had a similar question about Ray vs dask 1, 2, 3

As underlined by gold-panda, you can give a try with multiprocessing. I once implemented a version based on what is done in stable_baselines v1 (https://github.com/hill-a/stable-baselines/blob/master/stable_baselines/common/vec_env/subproc_vec_env.py)

Found relevant code at https://github.com/hill-a/stable-baselines + all code implementations here

Depends what you wanna do. Universal answer would be https://stable-baselines.readthedocs.io/

As you will see here it is recommended to augment this safety measure with target kl_divergence, that will ensure even smoother learning and enforce early stopping to prevent learning collapses.

Code for https://arxiv.org/abs/1707.06347 found: https://github.com/hill-a/stable-baselines

I haven't read the code, but stable-baselines doesn't support multi-agent environments (https://github.com/hill-a/stable-baselines/issues/423), so I think they're trying to make learning multi-agent easier with Environment.train().

- https://github.com/hill-a/stable-baselines (Tensorflow)

- tf2 speed: https://github.com/hill-a/stable-baselines/issues/576#issuecomment-573331715