NDHU AI Lab - AICup Tutorial
There are 3 AI models required to train in this contest, namely Dan, Kyu, and Play Style models. The baseline models are trained by using TensorFlow 2.9 and shared in this repository.
- CSVs: Dataset folder for training and testing
- Dan Training Tutorial.ipynb: Baseline model training of Dan dataset
- Kyu Training Tutorial.ipynb: Baseline model training of Kyu dataset
- PlayStyle Training Tutorial.ipynb: Baseline model training of PlayStyle dataset
- Create Public Upload CSV.ipynb: Predictions by using the baseline models
- model_dan_tutorial.h5: Baseline model of Dan dataset
- model_kyu_tutorial.h5: Baseline model of Kyu dataset
- model_playstyle_tutorial.h5: Baseline model of PlayStyle dataset
- public_submission_template: Predictions for public testing datasets
To generate the baseline models, you can follow tutorial notebooks.
For submission, you can follow Create Public Upload CSV.ipynb notebook.