tannic::expression::Argsum Struct Reference

Sums tensor values along an axis. More...

#include <reductions.hpp>

Public Member Functions
constexpr type	reduce (type dtype) const
constexpr Shape	reduce (Shape const &shape) const
void	forward (Tensor const &input, Tensor &output) const

Public Attributes
int	axis
bool	keepdim

Detailed Description

Sums tensor values along an axis.

Preserves input dtype. Use keepdim=true to maintain shape for broadcasting.

Example (NumPy/PyTorch behavior):

Tensor X = {{1, 2}, 
            {3, 4}};  // shape(2, 2)
 
Tensor Y = sum(X, 0);  // Sum over rows
// Y = [4, 6]          // 1+3=4, 2+4=6
 
Tensor Z = sum(X, 1, /*keepdim=* /true);
// Z = [[3],           // 1+2=3
//      [7]]           // 3+4=7

Member Function Documentation

forward()

void tannic::expression::Argsum::forward	(	Tensor const &	input,
		Tensor &	output
	)		const

reduce() [1/2]

constexpr Shape tannic::expression::Argsum::reduce ( Shape const &  shape ) const

inlineconstexpr

reduce() [2/2]

constexpr type tannic::expression::Argsum::reduce ( type  dtype ) const

inlineconstexpr

Member Data Documentation

axis

int tannic::expression::Argsum::axis

keepdim

bool tannic::expression::Argsum::keepdim

---

The documentation for this struct was generated from the following file:

• include/tannic/reductions.hpp