ubicomplab / rPPG-Toolbox

rPPG-Toolbox: Deep Remote PPG Toolbox (NeurIPS 2023)
https://arxiv.org/abs/2210.00716
Other
442 stars 106 forks source link

Data loading issues #223

Closed peizhengfu closed 10 months ago

peizhengfu commented 10 months ago

Hello, I try to train on the COHFACE dataset and test on the UFBC-rppg dataset, modify the existing .yaml and save it as a new COHFACE_COHFACE_UBFC-rPPG_DEEPPHYS_BASIC.yaml file, and then I use this file for training , but I received the following error, how should I solve it. Traceback (most recent call last): File "main.py", line 164, in train_data_loader = train_loader( File "/root/autodl-tmp/rPPG-Toolbox-main/dataset/data_loader/COHFACELoader.py", line 49, in init super().init(name, data_path, config_data) File "/root/autodl-tmp/rPPG-Toolbox-main/dataset/data_loader/BaseLoader.py", line 67, in init self.preprocess_dataset(self.raw_data_dirs, config_data.PREPROCESS, config_data.BEGIN, config_data.END) TypeError: preprocess_dataset() takes 3 positional arguments but 5 were given

yahskapar commented 10 months ago

Hi @peizhengfu,

As noted in our datasets section of the README, we currently only support SCAMPS, UBFC-rPPG, PURE, BP4D+, UBFC-Phys, and MMPD. Take a look at #185 for a few more details regarding the lack of COHFACE support.

If you still want to try and use COHFACE with this toolbox, you will have to update the dataloader yourself for the time being. I recommend looking at other dataloaders (e.g., UBFC-rPPG) for comparison and to better understand certain errors you might come across.