The problem of the dataset

[angelailei]

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]

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]

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