MIN2Net Usage
View source on GitHubTable of contents
About MIN2Net
Original authors have uploaded their code here https://github.com/IoBT-VISTEC/MIN2Net
If you use the MIN2Net model in your research, please cite the following paper:
@ARTICLE{9658165,
author={Autthasan, Phairot and Chaisaen, Rattanaphon and Sudhawiyangkul, Thapanun and Rangpong, Phurin and Kiatthaveephong, Suktipol and Dilokthanakul, Nat and Bhakdisongkhram, Gun and Phan, Huy and Guan, Cuntai and Wilaiprasitporn, Theerawit},
journal={IEEE Transactions on Biomedical Engineering},
title={MIN2Net: End-to-End Multi-Task Learning for Subject-Independent Motor Imagery EEG Classification},
year={2022},
volume={69},
number={6},
pages={2105-2118},
doi={10.1109/TBME.2021.3137184}}
Preparation for MIN2Net’s input
An example of MIN2Net’s input on the BCIC IV 2a dataset (All settings are set up as optimal settings in the original paper).
Downloading a particular dataset
python download_datasets.py --dataset 'BCIC2a'
Preprocessing based on time domain EEG over the considered dataset
python prep_time_domain.py --dataset 'BCIC2a'
Build, fit, and evaluate MIN2Net
# Subject-dependent MI classification
python run_MIN2Net.py --model_name 'MIN2Net_original' --dataset 'BCIC2a' --train_type 'subject_dependent' --data_type 'time_domain' --log_dir 'logs' --num_class 2 --num_chs 20 --GPU 0 --loss_weights 1.0 0.1 1.0 --margin 100.0
# Subject-independent MI classification
python run_MIN2Net.py --model_name 'MIN2Net_original' --dataset 'BCIC2a' --train_type 'subject_independent' --data_type 'time_domain' --log_dir 'logs' --num_class 2 --num_chs 20 --GPU 0 --loss_weights 0.5 0.1 1.0 --margin 1.0
Arguments:
| Arguments | Description | Default | |
|---|---|---|---|
| model_name | str prefix to save model | ‘MIN2Net_original’ | |
| dataset | str prefix to pick up a particular dataset and save model | ‘BCIC2a’ | |
| train_type | str prefix to pick up a particular traning manner and save model | ‘subject_dependent’ | |
| data_type | str prefix to select data type of input | ‘time_domain’ | |
| num_class | int number of classes | 2 | |
| num_chs | int number of classes | 20 | |
| latent_dim | int or None. If None, latent_dim is equal to number of channels C or 64 for 2- or 3-class classification, respectively. | None | |
| log_dir | str path to save model | ‘logs’ | |
| subjects | int or list of int or None list of range test subject, if None, use all subjects | None | |
| margin | float margin (alpha) of Triplet loss | 1.0 | |
| GPU | str select GPU ID | 0 | - |