gymnasium_rs's Introduction

gymnasium_rs

Rust implementation of Gymnasium API for reinforcement learning. This implementation is compatible and interoperable with the Python implementation.

Overview

The workspace contains these packages:

gymnasium: Public API
gymnasium_cli: [WIP] CLI tool
gymnasium_core: Core library
gymnasium_py: [WIP] Python module named gymnasium_rs for interoperability with Rust environments (Rust -> Python)
gymnasium_derive: [WIP] Procedural macros for deriving traits
gymnasium_sys: Rust FFI bindings for the official Python implementation of gymnasium

Instructions

Rust

Add gymnasium as a Rust dependency to your Cargo.toml manifest.

[dependencies]
gymnasium = "0.1"

Examples: gymnasium/examples/

CLI tool

Install the gymnasium_rs executable with cargo.

cargo install --locked gymnasium_cli

Afterwards, run the gymnasium_rs executable.

# Pass `--help` to show the usage and available options
gymnasium_rs

Docker

To install Docker on your system, you can run .docker/host/install_docker.bash to configure Docker with NVIDIA GPU support.
.docker/host/install_docker.bash

Build Image

To build a new Docker image from Dockerfile, you can run .docker/build.bash as shown below.

.docker/build.bash ${TAG:-latest} ${BUILD_ARGS}

Run Container

To run the Docker container, you can use .docker/run.bash as shown below.

.docker/run.bash ${TAG:-latest} ${CMD}

Run Dev Container

To run the Docker container in a development mode (source code mounted as a volume), you can use .docker/dev.bash as shown below.

.docker/dev.bash ${TAG:-latest} ${CMD}

As an alternative, users familiar with Dev Containers can modify the included .devcontainer/devcontainer.json to their needs. For convenience, .devcontainer/open.bash script is available to open this repository as a Dev Container in VS Code.

.devcontainer/open.bash

Join Container

To join a running Docker container from another terminal, you can use .docker/join.bash as shown below.

.docker/join.bash ${CMD:-bash}

License

This project is dual-licensed to be compatible with the Rust project, under either the MIT or Apache 2.0 licenses.

Contributing

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.

gymnasium_rs's People

Contributors

Stargazers

Watchers

gymnasium_rs's Issues

Rust implementation of space (`gymnasium::Space` trait, `gymnasium::space::*Space` impl)

Features

This meta-issue tracks the progress towards the next releases and describes the intended structure of the project.
Any form of feedback is welcome and appreciated!

Note: Before the first major release, the public API is subject to changes that are required to accommodate these features.

`gymnasium`

* Use-case: Python-based RL algorithm/framework used inside a Rust binary on Rust environments directly via a Python C library wrapper (e.g. pyo3)

TBD

Rust derive macro for vectorised/parallel environment (e.g. #[derive(gymnasium::VectorEnv)] macro)
Rust implementation for certain environment "wrappers" (e.g. gymnasium::GoalEnv: gymnasium::Env trait)

`gymnasium_py`

Interface for loading Rust-registered gymnasium::Env impl in Python (inside a standalone Python script)

`gymnasium_cli`

gymnasium_rs new ... — CLI for templating new Rust crates
gymnasium_rs list ... — CLI for listing all registered environments of Python modules and Rust crates
gymnasium_rs info ... — CLI for printing detailed information about registered environments
gymnasium_rs rand ... — CLI for running a random agent in a specific environment

andrejorsula / gymnasium_rs Goto Github PK