Reinforcement Learning Quickstart

To get setup with the repository,

git clone https://github.com/datares/rl-quickstart.git && cd rl-quickstart

To setup the development environment

conda create -n python=3.9 rl
conda activate rl
pip install -r requirements.txt

Then finally to train the agents

python main.py

Training metrics will be printed in stdout

Viewing Results

At the end of training, a window should open to display the trained agent.

To run tensorboard, run the following in a new terminal window

tensorboard --logdir=logs

Videos are also saved to the video directory, which has .mp4 videos of the testing results.