torch.NormOptions
export interface NormOptions<OutShape extends Shape = Shape> {
/** The order of the norm. Default: 2 (L2 norm). */
p?: number | 'fro' | 'nuc';
/** The dimension(s) to reduce. If undefined, reduces all dimensions. */
dim?: number | number[];
/** If true, keeps the reduced dimension with size 1 for broadcasting. */
keepdim?: boolean;
/** Data type of the output. */
dtype?: DType;
/** Pre-allocated output tensor. */
out?: Tensor<OutShape>;
}OutShapeextends Shapep(number | 'fro' | 'nuc')optional- – The order of the norm. Default: 2 (L2 norm).
dim(number | number[])optional- – The dimension(s) to reduce. If undefined, reduces all dimensions.
keepdim(boolean)optional- – If true, keeps the reduced dimension with size 1 for broadcasting.
dtype(DType)optional- – Data type of the output.
out(Tensor<OutShape>)optional- – Pre-allocated output tensor.
Options for norm operation