Hi Reuben,

Thanks for your code.
I am trying to use the pretrained model you provided to reproduce the results.
However, when I run:
python inference.py -n u_hemis -t1 ./data/BRATS2018/HGG3_T1.nii.gz -t2 ./data/BRATS2018/HGG3_T2.nii.gz -t1c ./data/BRATS2018/HGG3_T1c.nii.gz -fl ./data/BRATS2018/HGG3_Flair.nii.gz

I got an error that said:

Number of subjects 2, input section names: ['T1', 'subject_id', 'T1c', 'Flair', 'T2']
Dataset partitioning:
-- training 190 cases (9500.00%),
-- validation 95 cases (4750.00%),
-- inference 0 cases (0.00%).

CRITICAL:niftynet: Reader required input section name [T1], but in the filename list the column is empty.
CRITICAL:niftynet: file_list parameter should be a pandas.DataFrame instance and has input section name [T1] as a column name.
CRITICAL:niftynet: Reader requires section(s): ['T1', 'T1c', 'T2', 'Flair']
CRITICAL:niftynet: Configuration input sections are: ['T1', 'subject_id', 'T1c', 'Flair', 'T2']
Traceback (most recent call last):
File "/apdcephfs/share_1290796/medzhang_data/Task/missing_modal/U-HVED/uhved_python/bin/net_run", line 8, in
sys.exit(main())
File "/apdcephfs/share_1290796/medzhang_data/Task/missing_modal/U-HVED/uhved_python/lib/python3.6/site-packages/niftynet/init.py", line 141, in main
app_driver.initialise_application(system_param, input_data_param)
File "/apdcephfs/share_1290796/medzhang_data/Task/missing_modal/U-HVED/uhved_python/lib/python3.6/site-packages/niftynet/engine/application_driver.py", line 167, in initialise_application
data_param, app_param, self.data_partitioner)
File "/apdcephfs/share_1290796/medzhang_data/Task/missing_modal/U-HVED/extensions/multi_modal_application.py", line 97, in initialise_dataset_loader
data_param, task_param, file_list) for file_list in file_lists]
File "/apdcephfs/share_1290796/medzhang_data/Task/missing_modal/U-HVED/extensions/multi_modal_application.py", line 97, in
data_param, task_param, file_list) for file_list in file_lists]
File "/apdcephfs/share_1290796/medzhang_data/Task/missing_modal/U-HVED/uhved_python/lib/python3.6/site-packages/niftynet/io/image_reader.py", line 158, in initialise
raise ValueError
ValueError

I have checked the version of tensorflow and niftynet.
Do you have any idea about what is wrong here?

Best,
Yao