Skip to content

🏆 gym-cooking: Code for "Too many cooks: Bayesian inference for coordinating multi-agent collaboration", Winner of the CogSci 2020 Computational Modeling Prize in High Cognition, and a NeurIPS 2020 CoopAI Workshop Best Paper.

License

Notifications You must be signed in to change notification settings

SPL-BGU/gym-cooking

 
 

Repository files navigation

Maps To create new maps add textual maps manually to utils/levels following the instructions in design.md There is also a script to generate random maps located at utils/map_generator.py The pddl maps can be generated using the pddl_problem_generator.py script. simply provide the path to the textual

Run To run the environment see instructions in gym_cooking/README.md All runs are logged in experiment_log.txt located in gym_cooking All vector representations of the states are logged in state_logs folder located in env_log (relevant for rl only)

Experiments To run experiments use the run_experiments.sh script located in gym_cooking. You can specify the planner, map, number of runs, and other parameters. The results of the experiments are also logged in experiment_log.txt The content of the log can be exported to a csv file using the export_log_to_csv.py script located in gym_cooking/utils.

About

🏆 gym-cooking: Code for "Too many cooks: Bayesian inference for coordinating multi-agent collaboration", Winner of the CogSci 2020 Computational Modeling Prize in High Cognition, and a NeurIPS 2020 CoopAI Workshop Best Paper.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • PDDL 99.5%
  • Makefile 0.4%
  • C++ 0.1%
  • Python 0.0%
  • CMake 0.0%
  • C 0.0%