labelme2coco.py crash when just contines __ignore__ label

[withkun]

Provide environment information

(DET2) D:\FPCs>python --version
Python 3.9.13

(DET2) D:\FPCs>pip list labelme
Package Version Editable project location

---

labelme 5.0.5

What OS are you using?

windows11

Describe the Bug

(DET2) D:\FPCs>python D:/DeepLearning/labelme/examples/instance_segmentation/labelme2coco.py --labels classes.txt 000003 labels
Creating dataset: labels
Generating dataset from: 000003\dog.json
Traceback (most recent call last):
File "D:\DeepLearning\labelme\examples\instance_segmentation\labelme2coco.py", line 209, in
main()
File "D:\DeepLearning\labelme\examples\instance_segmentation\labelme2coco.py", line 184, in main
labels, captions, masks = zip(
ValueError: not enough values to unpack (expected 3, got 0)

Expected Behavior

Skip the label file

To Reproduce

1. Just set ignore label of an image
2. Create classes.txt file
3. run labelme2coco.py

[withkun]

label file of dog.json
{
"version": "5.0.5",
"flags": {},
"shapes": [
{
"label": "ignore",
"points": [
[
1248.0740740740741,
445.7530864197531
],
[
1586.3456790123457,
1012.4197530864197
]
],
"group_id": null,
"shape_type": "rectangle",
"flags": {}
}
],
"imagePath": "dog.png",
"imageData": null,
"imageHeight": 1152,
"imageWidth": 2048
}

[HUBUwg96]

how to deal with？

[withkun]

Just skip it?

[qinli-jian]

The solution is here: #774