Installation.txt

Follow these steps to set up a virtual environment to run our algorithms. We use conda environments here.

1) Install Mujoco (https://github.com/openai/mujoco-py)

2) Create the conda environment from the yml file

		conda env create -f environment.yml

3) Activate the conda environment
	
		conda activate DEOC

4) Install baselines through the setup file in the project folder

		pip install -e .

5) Install gym-extensions through the setup file in the gym-extensions-master folder

		cd gym-extensions-master
		pip install -e .

6) Install gym-miniworld through the setup file in the gym-miniworld-master folder

		git clone https://github.com/maximecb/gym-miniworld.git
		cd gym-miniworld-master
		pip install -e .

** Before running experiments, please consult the hyperparameters we provide in the Appendix

7) For four-rooms task:
	
		cd TDEOC_fourrooms
		python TDEOC_Tabular.py --tdeoc

8) For Mujoco (and TMaze transfer) tasks:
	
		cd baselines/Termination_DEOC

		python run_mujoco.py --env='Walker2d-v2' --tradeoff=0.2 --tdeoc

9) For Miniworld tasks:
	
		cd baselines/Termination_DEOC

		python run_atari_miniworld.py --env='MiniWorld-OneRoom-v0' --tradeoff=0.0 --tdeoc