Rate this Page

Struct SoftmaxFuncOptions#

Page Contents

Struct Documentation#

struct SoftmaxFuncOptions#

Options for torch::nn::functional::softmax.

Example:

namespace F = torch::nn::functional;
F::softmax(input, F::SoftmaxFuncOptions(1));

Public Functions

SoftmaxFuncOptions(int64_t dim)#
inline auto dim(const int64_t &new_dim) -> decltype(*this)#

Dimension along which Softmax will be computed.

inline auto dim(int64_t &&new_dim) -> decltype(*this)#
inline const int64_t &dim() const noexcept#
inline int64_t &dim() noexcept#
inline auto dtype(const std::optional<torch::Dtype> &new_dtype) -> decltype(*this)#

the desired data type of returned tensor.

If specified, the input tensor is casted to dtype before the operation is performed. This is useful for preventing data type overflows. Default: None.

inline auto dtype(std::optional<torch::Dtype> &&new_dtype) -> decltype(*this)#
inline const std::optional<torch::Dtype> &dtype() const noexcept#
inline std::optional<torch::Dtype> &dtype() noexcept#