search

hukaixuan19970627 / yolov5_obb

yolov5 + csl_label.(Oriented Object Detection)（Rotation Detection）（Rotated BBox）基于yolov5的旋转目标检测
GNU General Public License v3.0
1.81k stars 425 forks source link

ignoring corrupt image/label？ #598

Open nightrain-vampire opened 1 year ago

nightrain-vampire commented 1 year ago

I run the training command as the 'getStart.md', but it reports:
image

The others said there may be something wrong with data format. However, I use the ImgSplit_multi_process.py to generate image and labels. What's the matter? Can anyone help me?

MAhmedDev commented 9 months ago

I have the same issue, @nightrain-vampire were you able to resolve the issue?

daihongxia commented 8 months ago

maybe it's because of the numpy version? I modified L983 of utils/datasets.py as l_.append(np.concatenate(([cls_id], label[:8]), axis=None)) and it worked for me. and remember to delete the cache before rerunning the training script.

IsenLwww commented 5 months ago

Do you solve the issue?

IsenLwww commented 5 months ago

maybe it's because of the numpy version? I modified L983 of utils/datasets.py as and it worked for me. and remember to delete the cache before rerunning the training script.l_.append(np.concatenate(([cls_id], label[:8]), axis=None)) Do you have the same issue? I can't solve this problem.