import albumentations as A
return A.Compose([
A.RandomResizedCrop(size=(image_size, image_size), scale=(0.5, 1.0), p=1.0),
A.GaussianBlur(blur_limit=(3, 7), sigma_limit=0.1, p=0.5),
ToTensorV2()
])
The algorithm should pick the closes by similarity.
This should be an option since it will increase processing time.
Add albumentations augmentation as an option to the http://doris.shore.mbari.org:8000/docs#/default/knn
endpoint
The algorithm should pick the closes by similarity.
This should be an option since it will increase processing time.