⚠️ SARL is pre-release software, in active development!
SARL is a toolkit for performing reinforcement learning on unconventionally specified action spaces. Such action-spaces differ from the non-hierarchical, exclusively discrete or continuous action-spaces explored in conventional reinforcement learning.
The most common structured action space is the parameterized action-space, requiring an agent to select from a discrete set of actions, then specify a continuous vector from that action's corresponding parameter-space (Masson et. al. 2016).
Name | Inclusion | Notes |
---|---|---|
H-PPO | ||
MP-DQN | ||
SP-DQN | ||
P-DQN | In Progress | |
PASVG(0) | ||
PATRPO | ||
PADDPG | ||
Q-PAMDP | In Progress |
Name | Inclusion | Notes |
---|---|---|
Chase & Attack | ||
Catching Point | ||
King of Glory | ||
Simulation / Moving | ||
Half Field Offense (HFO) | ||
Platform | ✔ | * |
RoboCup | ✔ | * |
* Custom wrappers exist, yet to be included in environment options.
Temporary state, subject to change.
Instructions for use on Linux, Mac OSX, or Windows (via WSL).
- Ensure Python is installed
python --version
- Ensure Poetry is installed
poetry --version
- You can install poetry with
curl -sSL https://install.python-poetry.org | python3 -
- Full instructions and alternative methods can be found here
- You can install poetry with
- Clone this repository
git clone https://github.com/ttollet/SARL.git
- Change directory
cd SARL
- Install dependencies to virtual environment
poetry install
- Check functionality
poetry run pytest