Apache SINGA
A distributed deep learning platform .
|
swap two axis of a tensor input: Tensor<Device,dim>: ishape output: Tensor<Device,dimdst> oshape[a1],oshape[a2] = ishape[a2],oshape[a1] More...
#include <tensor_expr_ext.h>
Public Member Functions | |
SwapAxisExp (const SrcExp &src) | |
constructor | |
Public Member Functions inherited from mshadow::expr::MakeTensorExp< SwapAxisExp< SrcExp, dimsrc, a1, a2 >, SrcExp, dimsrc > | |
const SwapAxisExp< SrcExp, dimsrc, a1, a2 > & | real_self (void) const |
true self of subtype | |
Public Member Functions inherited from mshadow::expr::Exp< MakeTensorExp< SwapAxisExp< SrcExp, dimsrc, a1, a2 >, SrcExp, dim >, type::kMapper > | |
const MakeTensorExp < SwapAxisExp< SrcExp, dimsrc, a1, a2 >, SrcExp, dim > & | self (void) const |
MakeTensorExp< SwapAxisExp < SrcExp, dimsrc, a1, a2 > , SrcExp, dim > & | refself (void) |
Public Attributes | |
const SrcExp & | src_ |
source expression | |
Public Attributes inherited from mshadow::expr::MakeTensorExp< SwapAxisExp< SrcExp, dimsrc, a1, a2 >, SrcExp, dimsrc > | |
Shape< dim > | shape_ |
the shape of this expression | |
swap two axis of a tensor input: Tensor<Device,dim>: ishape output: Tensor<Device,dimdst> oshape[a1],oshape[a2] = ishape[a2],oshape[a1]
SrcExp | type of source expression |
dimsrc | source dimension |
a1 | smaller dimension to be swapped |
a2 | larger dimension to be swapped |