search

hpi-epic / BP2021

Working repository in context of the bachelor project "Online Marketplace Simulation: A Testbed for Self-Learning Agents" at the research group Enterprise Platform and Integration Concepts
MIT License
3 stars 0 forks source link

q_learning_config hardcoded #544

Open davidhennemann opened 2 years ago

davidhennemann commented 2 years ago

in training_scenario.py in train_from_config() the RL config seems to be hardcoded to q_learning_config.

config_rl: AttrDict = HyperparameterConfigLoader.load('q_learning_config', config_environment.agent[0]['agent_class'])

According to the wiki the relevant config file should be named rl_config.json . The same issue can also be found in the exampleprinter.py.