Flip top-bottom with probability p (Albumentations VerticalFlip).
Mirrors the image about its horizontal centre-line, mirrors the mask the same way, and reflects bounding-box / keypoint y-coordinates via the canvas height so every target stays aligned.
Parameters
pfloat= 0.5Probability of applying the flip.
Examples
>>> import lucid, lucid.utils.transforms as T
>>> tf = T.VerticalFlip(p=1.0)
>>> tuple(tf(T.Image(lucid.rand(3, 32, 32))).data.shape)
(3, 32, 32)