module documentation
Undocumented
| Function | pad |
Pad a tensor's size to a multiple of a number. |
def pad_to_multiple(tensor:
torch.Tensor, multiple: int, *, mode: str, value: float = 0.0) -> torch.Tensor:
(source)
¶
Pad a tensor's size to a multiple of a number.
| Parameters | |
tensor:torch.Tensor | Tensor to pad. |
multiple:int | Size multiple to pad to. |
mode:str | Padding mode; see torch.nn.functional.pad. |
value:float | Padding value; see torch.nn.functional.pad. |
| Returns | |
torch.Tensor | Padded tensor, or the original tensor if no padding was needed. |