├── .gitignore ├── LICENSE ├── Readme.md ├── agents ├── __init__.py ├── base_agent.py ├── categorical_dqn_agent.py ├── dqn_agent.py ├── evaluation_agent.py └── random_agent.py ├── categorical.yml ├── configs ├── atari_bench.yaml ├── atari_dev.yaml ├── catch_categorical.yaml ├── catch_dev.yaml └── catch_dqn.yaml ├── data_structures ├── __init__.py ├── ntuple_experience_replay.py └── tensor_experience_replay.py ├── estimators ├── __init__.py ├── atari_net.py └── catch_net.py ├── img └── learning_curves.png ├── main.py ├── policy_evaluation ├── __init__.py ├── categorical.py ├── deterministic.py └── exploration_schedules.py ├── policy_improvement ├── __init__.py ├── categorical_update.py ├── dqn_update.py └── optim_utils.py └── utils ├── __init__.py ├── parse_config.py ├── torch_types.py ├── utils.py └── wrappers.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/.gitignore -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/LICENSE -------------------------------------------------------------------------------- /Readme.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/Readme.md -------------------------------------------------------------------------------- /agents/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/agents/__init__.py -------------------------------------------------------------------------------- /agents/base_agent.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/agents/base_agent.py -------------------------------------------------------------------------------- /agents/categorical_dqn_agent.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/agents/categorical_dqn_agent.py -------------------------------------------------------------------------------- /agents/dqn_agent.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/agents/dqn_agent.py -------------------------------------------------------------------------------- /agents/evaluation_agent.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/agents/evaluation_agent.py -------------------------------------------------------------------------------- /agents/random_agent.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/agents/random_agent.py -------------------------------------------------------------------------------- /categorical.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/categorical.yml -------------------------------------------------------------------------------- /configs/atari_bench.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/configs/atari_bench.yaml -------------------------------------------------------------------------------- /configs/atari_dev.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/configs/atari_dev.yaml -------------------------------------------------------------------------------- /configs/catch_categorical.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/configs/catch_categorical.yaml -------------------------------------------------------------------------------- /configs/catch_dev.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/configs/catch_dev.yaml -------------------------------------------------------------------------------- /configs/catch_dqn.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/configs/catch_dqn.yaml -------------------------------------------------------------------------------- /data_structures/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/data_structures/__init__.py -------------------------------------------------------------------------------- /data_structures/ntuple_experience_replay.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/data_structures/ntuple_experience_replay.py -------------------------------------------------------------------------------- /data_structures/tensor_experience_replay.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/data_structures/tensor_experience_replay.py -------------------------------------------------------------------------------- /estimators/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/estimators/__init__.py -------------------------------------------------------------------------------- /estimators/atari_net.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/estimators/atari_net.py -------------------------------------------------------------------------------- /estimators/catch_net.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/estimators/catch_net.py -------------------------------------------------------------------------------- /img/learning_curves.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/img/learning_curves.png -------------------------------------------------------------------------------- /main.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/main.py -------------------------------------------------------------------------------- /policy_evaluation/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/policy_evaluation/__init__.py -------------------------------------------------------------------------------- /policy_evaluation/categorical.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/policy_evaluation/categorical.py -------------------------------------------------------------------------------- /policy_evaluation/deterministic.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/policy_evaluation/deterministic.py -------------------------------------------------------------------------------- /policy_evaluation/exploration_schedules.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/policy_evaluation/exploration_schedules.py -------------------------------------------------------------------------------- /policy_improvement/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/policy_improvement/__init__.py -------------------------------------------------------------------------------- /policy_improvement/categorical_update.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/policy_improvement/categorical_update.py -------------------------------------------------------------------------------- /policy_improvement/dqn_update.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/policy_improvement/dqn_update.py -------------------------------------------------------------------------------- /policy_improvement/optim_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/policy_improvement/optim_utils.py -------------------------------------------------------------------------------- /utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/utils/__init__.py -------------------------------------------------------------------------------- /utils/parse_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/utils/parse_config.py -------------------------------------------------------------------------------- /utils/torch_types.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/utils/torch_types.py -------------------------------------------------------------------------------- /utils/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/utils/utils.py -------------------------------------------------------------------------------- /utils/wrappers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/floringogianu/categorical-dqn/HEAD/utils/wrappers.py --------------------------------------------------------------------------------