Trait dfdx::tensor_ops::SumTo

pub trait SumTo: HasErr + HasShape {
    // Required method
    fn try_sum<Dst: Shape, Ax: Axes>(
        self
    ) -> Result<Self::WithShape<Dst>, Self::Err>
       where Self::Shape: ReduceShapeTo<Dst, Ax>;

    // Provided method
    fn sum<Dst: Shape, Ax: Axes>(self) -> Self::WithShape<Dst>
       where Self::Shape: ReduceShapeTo<Dst, Ax> { ... }
}

Reduction along multiple axes using sum.

Required Methods

fn try_sum<Dst: Shape, Ax: Axes>( self ) -> Result<Self::WithShape<Dst>, Self::Err>where Self::Shape: ReduceShapeTo<Dst, Ax>,

Fallible version of SumTo::sum

Provided Methods

fn sum<Dst: Shape, Ax: Axes>(self) -> Self::WithShape<Dst>where Self::Shape: ReduceShapeTo<Dst, Ax>,

Sum reduction. Pytorch equivalent: t.sum(Ax)

Example reducing a single axis:

let t: Tensor<Rank2<2, 3>, f32, _> = dev.tensor([[1.0, 2.0, 3.0], [-1.0, -2.0, -3.0]]);
let r = t.sum::<Rank1<2>, _>(); // or `sum::<_, Axis<1>>()`
assert_eq!(r.array(), [6.0, -6.0]);

Reducing multiple axes:

let r = t.sum::<Rank0, _>(); // or `sum::<_, Axes2<0, 1>>()`
assert_eq!(r.array(), 0.0);

Implementors

impl<S: Shape, E: Dtype, D: SumKernel<E>, T: Tape<E, D>> SumTo for Tensor<S, E, D, T>