Does anyone is getting the below error in Starter Notebook? If yes then how to fix the issue
ValueError: All bounding boxes should have positive height and width. Found invalid box [776.5625, 137.5, 776.5625, 143.75] for target at index 3.
Try to check the image augmentations you did. Most likely one of them resulted to 'out of bounds' for the bounding boxes. i.e. Try removing some(the augs) and reCheck if the error persists. Good Luck!
Hello @NAC_ENGINE. I am following the starter notebook . I am getting error in the starter notebook
import albumentations as A
from albumentations.pytorch.transforms import ToTensorV2
def get_train_transform():
return A.Compose([A.Flip(0.5),ToTensorV2(p=1.0), bbox_params={'format': 'pascal_voc', 'label_fields': ['labels']})
def get_valid_transform():
return A.Compose([ToTensorV2(p=1.0)], bbox_params={'format': 'pascal_voc', 'label_fields': ['labels']})
Delet A.Flip(0.5)
Try to check the image augmentations you did. Most likely one of them resulted to 'out of bounds' for the bounding boxes. i.e. Try removing some(the augs) and reCheck if the error persists. Good Luck!
Hello @NAC_ENGINE. I am following the starter notebook . I am getting error in the starter notebook
import albumentations as A
from albumentations.pytorch.transforms import ToTensorV2
def get_train_transform():
return A.Compose([A.Flip(0.5),ToTensorV2(p=1.0), bbox_params={'format': 'pascal_voc', 'label_fields': ['labels']})
def get_valid_transform():
return A.Compose([ToTensorV2(p=1.0)], bbox_params={'format': 'pascal_voc', 'label_fields': ['labels']})
Delet A.Flip(0.5)