nisargfrd / CampX

Tensor Based Environment Framework for Training RL Agents - Pre Alpha

CampX

Agent Training environments that are entirely created using tensor operations. The API is a modified port of PyColab and should be loosely compatible with worlds created for PyColab.

Cloning:

In order to get everything, use git's --recursive flag:

git clone --recursive https://github.com/OpenMined/CampX.git

This will also download the safe-grid-agents and ai-safety-gridworlds repos, which can be used for training fast plaintext agents in safety-minded gridworlds. See those repos' respective READMEs for installation/usage instructions.

Install

python3 -m venv venv
source venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt
python setup.py install # installs campx from latest git build
python -m ipykernel install --user --name=campx
jupyter notebook # select campx kernel

Update the Code

git pull
python setup.py install
# install the software and make requirements
# remove the campx from requirements
pip freeze | sed 's/@[a-z0-9]\+//' > requirements.txt