Error in Display Samples on inspect_dataset_via file

[StraigenDaigen]

Hello!, I'm looking your code inspect_dataset_via notebook, but in Display Samples cell I received this error

KeyError Traceback (most recent call last)
in
3 for image_id in image_ids:
4 image = dataset.load_image(image_id)
----> 5 mask, class_ids = dataset.load_mask(image_id)
6 visualize.display_top_masks(image, mask, class_ids, dataset.class_names)

D:\Steven\Mask_RCNN\maskrcnn\dataset.py in load_mask(self, image_id)
264
265 # find the class ID corresponding to the region's class attribute
--> 266 class_label = region_attributes["class"]
267 class_id = -1
268 for key, label in self.class_ids_to_labels.items():

KeyError: 'class'

I don't know why, because in the before cell "dataset" the code is working:

Image Count: 220

Class Count: 4
0. BG

1. rio
2. remocion
3. piscina

I annotated the images in VIA 2.0.8.

[monocongo]

Thanks for pointing this out, @StraigenDaigen.

Can you see the items (keys and values) of the region_attributes? If so do you see a key named "class"? If not then there's the problem, somehow that attribute has not been set, and this code is expecting it to be present.

I apologize I don't have much time to devote to this project as I am not focused on segmentation now as I was when I was developing the code in this repository. However, if you are able to find a fix for this please comment and/or submit a PR, your participation is welcome and appreciated!