search

Jathurshan0330 / Cross-Modal-Transformer

Official repository of cross-modal transformer for interpretable automatic sleep stage classification. https://arxiv.org/abs/2208.06991
45 stars 3 forks source link

The problem of the dataset #1

Open angelailei opened 1 year ago

angelailei commented 1 year ago

hello, what bothers me is that I don't know the exact form and format of the sleepEDF dataset, can you explain how to apply the dataset

Jathurshan0330 commented 1 year ago

hi, we have added the data preparations scripts under folder /data_preparations. These scripts can be used to generate the datasets. The dataset is separated into 5 sets (5-fold validation), where subjects in each set are independent. Then the above script extracts data for each channel. You could change the channel to "eog" and other PSG channels to extact data separately.

angelailei commented 1 year ago

Thankyou sincerely for my problem, but i was still geting error when trying to run the .py, and i use the default sleep-EDF databast including .EDF files. Is this the problems?

/headless/anaconda3/envs/env1-pt/bin/python "/home/EEG test/Cross-Modal-Transformer-master/cmt_training.py" Torch Version : 1.7.0 Training Arguements ====================================> project_path : ./results/ data_path : .datasets/sleep-telemetry/ train_data_list : [0, 1, 2, 3] val_data_list : [4] is_retrain : False model_path :
save_model_freq : 50 model_type : Epoch d_model : 256 dim_feedforward : 1024 window_size : 50 num_seq : 5 batch_size : 32 weigths : [1.0, 2.0, 1.0, 2.0, 2.0] lr : 0.001 beta_1 : 0.9 beta_2 : 0.999 eps : 1e-09 weight_decay : 0.0001 n_epochs : 200 step_size : 30 gamma : 0.5 is_neptune : False nep_project :
nep_api :
Project directory already available at ./results/ Getting Dataset ===================================> Traceback (most recent call last): File "/home/EEG test/Cross-Modal-Transformer-master/cmt_training.py", line 158, in main() File "/home/EEG test/Cross-Modal-Transformer-master/cmt_training.py", line 94, in main train_data_loader, val_data_loader = get_dataset(device, args) File "/home/EEG test/Cross-Modal-Transformer-master/datasets/sleep_edf.py", line 198, in get_dataset args.train_data_list = list(args.train_data_list[0]) TypeError: 'int' object is not iterable