Reinforcement Learning for playing arcade fighting game KOF'97

Introduction

In this project, we intend to develop an RL Agent which can not only win the game: KOF’97 The following works are done in this project

We made a gym envrionment of KOF 97. see Kof97EnvironmentSR.py
Train models for the specific opponent. see trainVecEnvV2.py and trainVecEnvLSTM.py
Evaluate the performance of those models. see Kof97_Evaluation.ipynb
visualise the evaluation result. see visualize.ipynb

Name		Name	Last commit message	Last commit date
Latest commit History 17 Commits
.idea		.idea
data		data
images		images
kof97		kof97
resource		resource
Callback.py		Callback.py
Kof97Environment.py		Kof97Environment.py
Kof97EnvironmentSR.py		Kof97EnvironmentSR.py
Kof97_Evaluation.ipynb		Kof97_Evaluation.ipynb
KofActions.py		KofActions.py
KofSteps.py		KofSteps.py
README.md		README.md
trainVecEnvLSTM.py		trainVecEnvLSTM.py
trainVecEnvV2.py		trainVecEnvV2.py
visualize.ipynb		visualize.ipynb