RandomGrayscale — Torchvision 0.22 documentation (original) (raw)

class torchvision.transforms.v2.RandomGrayscale(p: float = 0.1)[source]¶

Randomly convert image or videos to grayscale with a probability of p (default 0.1).

If the input is a torch.Tensor, it is expected to have […, 3 or 1, H, W] shape, where … means an arbitrary number of leading dimensions

The output has the same number of channels as the input.

Parameters:

p (float) – probability that image should be converted to grayscale.

make_params(flat_inputs: List[Any]) → Dict[str, Any][source]¶

Method to override for custom transforms.

See How to write your own v2 transforms

transform(inpt: Any, params: Dict[str, Any]) → Any[source]¶

Method to override for custom transforms.

See How to write your own v2 transforms