aminmax

→tuple[Tensor, Tensor]

aminmax(input: Tensor, dim: _int | None = ..., keepdim: _bool = ...)

Compute the minimum and maximum of input.

inputTensor

Input tensor.

dimint or None

Dimension along which to compute minimum and maximum. When None (default), reduces over the entire tensor.

keepdimbool

If True, retains the reduced dimensions with size 1.

tuple[Tensor, Tensor]

Named pair (min, max) with .min and .max attributes.

>>> import lucid
>>> out = lucid.aminmax(lucid.tensor([3.0, -1.0, 4.0]))
>>> (out.min.item(), out.max.item())
(-1.0, 4.0)