A modular AI agent experimentation tool.
Project description
piSTAR Lab
WARNING: Under active development - not ready for public use.
Overview
piSTAR Lab is a modular deep reinforcement learning platform built to make AI development accessible and fun.
Release: 0.0.1-dev (early release)
Features:
- Web UI
- Plugin System for adding new agents, environments or tasks types
- Python API, anthing you can do in the UI, you can do in Python as well
- Run agents in single and multi player environments
- Experiment tracking
- Built in web-based IDE (via Theia (https://theia-ide.org/))
- Uses Ray Project (https://ray.io/) under the hood for distributed processing
Licensed under an Apache-2.0 license.
UI Screenshots
Quick Start
These instructions are for single node only. For cluster mode, see TODO
Installation with Anaconda
NOTE: Only tested on Ubuntu, but should also work on OS X. MS Windows users see Installation using Docker
-
Install Anaconda or Miniconda Visit https://www.anaconda.com/products/individual for instructions
-
Install PIP
conda install pip
-
Clone Repo and install
git clone https://github.com/pistarlab/pistarlab cd pistarlab pip install -e .
-
build Redis
bash ./install_redis.sh_
-
install node for UI and IDE
bash ./install_node.sh bash ./build_ui.sh bash ./build_ide.sh #optional
-
install additional dependencies
- XVFB to render without display (No MS Windows Support)
- ffmpeg for video processing
sudo apt-get install -y xvfb ffmpeg
Usage
To launching piSTAR Lab, run:
pistarlab
-
Launcher Control Panel: http://localhost:7776
Installation using Docker
Recommended for MS Windows users
-
Install Docker:
-
Clone Repo
git clone https://github.com/pistarlab/pistarlab cd pistarlab
-
Build Docker Image
./build_docker.sh
Usage with Docker
Launching piSTAR Lab
./bin/docker_launcher.sh
Contributing
We are still in an early phase of this release and have many loose ends to wrap up. If you are interested in contributing to piSTAR Lab, please reach out.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for pistarlab-0.0.1.dev0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | d9692447db03168cb5613c6ea4ec45e1590cc4d29fc0cff9f59c76f23755ce9b |
|
MD5 | fabb15d5017091d7ce8250cd61beda38 |
|
BLAKE2b-256 | d2b2679f668c42d089aa14daa57a8632ae5f757aeb9e05162e10bbb5cfaf65da |