SE label problem

[redeemme]

Hello, when I run the preprocess.py to process IDRID dataset I encounter such an error:
label_all[np.where(label4 > 0)] = 4
TypeError: '>' not supported between instances of 'NoneType' and 'int'
It seems that the mask number of SE is less than other labels, so some images do not have SE masks, could you plz tell me how to avoid this?
I just copied the original data and masks from IDRiD dataset, is there any problem?