wenxi-yue / SurgicalSAM

[AAAI2024] Official implementation of SurgicalSAM
MIT License
70 stars 9 forks source link

about KeyError: 'seq5\\00060_class1.png/00060.png' #7

Closed Kingxudong closed 6 months ago

Kingxudong commented 8 months ago

In the eval_endovis function in surgicalSAM/utils.py

for file_name, prediction in endovis_masks.items(): full_mask = gt_endovis_masks[file_name]

"KeyError: 'seq5\00060_class1.png/00060.png'" appears As far as I know, gt_endovis_masks: {'seq15/00000.png': is such a dictionary, And endovis_masks: {'seq5\00060_class1.png/00060.png' is such a dictionary , they should not be able to correspond to, I directly downloaded the processed endovis_2018 data set provided by you.

wenxi-yue commented 8 months ago

Hi,

Thanks for your interest in our work.

The keys in your endovis_masks appear to have an issue; it shouldn't include "\00060_class1.png" in the middle, in your example.

Below I present the keys of gt_endovis_masks and endovis_masks for your reference.

image

image