Possible issue about the robust value iteration

[DanialTaheri]

Hi,

I was trying to run the experiment file using RVI algorithm. But, I face an error as follows. I appreciate it if you let me know if I am doing sth wrong. My bash command: python3 experiments.py evaluate configs/FiniteMDPEnv/large/env_2.json configs/FiniteMDPEnv/large/agents/value_iteration.json --train

The error: Traceback (most recent call last): File "experiments.py", line 149, in main() File "experiments.py", line 44, in main evaluate(opts[''], opts[''], opts) File "experiments.py", line 60, in evaluate env = load_environment(environment_config) File "/.local/lib/python3.6/site-packages/rl_agents/agents/common/factory.py", line 69, in load_environment import(env_config["import_module"]) ModuleNotFoundError: No module named 'finite_mdp'

[eleurent]

Hi @DanialTaheri Like the value iteration, the robust value iteration requires the knowledge of the MDP(s), described in the form of a FiniteMDPEnv, defined in this project. (It can also be used on environments that provide a to_finite_mdp() method that converts them into a FiniteMDPEnv.)

To fix your error, simply clone the project locally, or install it with pip: pip install --user git+https://github.com/eleurent/finite-mdp