TFF
TFF copied to clipboard
an end to end framework for analyzing fMRI time-series data (4D) using transformers
Thanks again for your excellent work! I'm following the same pipeline of your using the fMRI. I notice that you load the TR sequence randomly (the random_TR argument). Could you...
Thank you for your excellent works! Could you provide the code for preprocessing the ucla datasets as well? Or is it identical to the HCP datasets?
0it [00:00, ?it/s] 0it [00:00, ?it/s] ______epoch summary 1/10_____ Traceback (most recent call last): File "main.py", line 137, in main(base_path) File "main.py", line 118, in main model_weights_path_phase1 = run_phase(args,None,'1','autoencoder_reconstruction') File...
def test(args, model_weights_path): experiment_folder = '{}_{}_{}'.format(args.dataset_name, 'test_{}'.format(args.fine_tune_task), datestamp()) experiment_folder = os.path.join(args.base_path, 'tests', experiment_folder) os.makedirs(experiment_folder) # _ = run_phase(args, model_weights_path, '3', 'test') trainer = Trainer(experiment_folder, '3', args, ['test'], model_weights_path) trainer.testing()
hey, i am currently facing a similar issue. Could you please possibly share how i can make the code work for ucla dataset ? _Originally posted by @kri14 in https://github.com/GonyRosenman/TFF/issues/2#issuecomment-1500881552_
Thank you for your very outstanding work, I would like to ask where is the multiprocessing from multiprocessing import Process, Queue in prepreocessing.py in data_preprocess_and_load, I don't find it!