Giters

han-liu / ModDropPlusPlus

Official PyTorch Implementation of ModDrop++ [MICCAI 2022 (early accept)]. A simple yet effective approach to tackle missing-modality problem for multi-modality medical imaging data.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Data Conversion has issues

zzs95 opened this issue · comments

commented

Data conversion description and code do not match.
Should uncomment this line for the first time running in train.py

data_generator.build_dataset(val_index, test_index, opt.test_mode) # uncomment for online data generation

commented

Thanks for pointing it out. This line of code should indeed be uncommented for the first time of training. Once the dataset has been generated, we can comment this line to skip the data generation process. I've uncommented this line in the train.py.