ChannelShuffle

class myoverse.transforms.ChannelShuffle(p=0.5, **kwargs)

Randomly shuffle channel order.

Parameters:

p (float) – Probability of applying shuffle.

Examples

>>> x = torch.randn(64, 2048, device='cuda', names=('channel', 'time'))
>>> shuffle = ChannelShuffle(p=0.5)
>>> y = shuffle(x)

Methods

__init__([p])
_apply(x)	Apply the transform.