Struct GroupNormOptions¶
Defined in File normalization.h
Page Contents
Struct Documentation¶
-
struct
torch::nn
::
GroupNormOptions
¶ Options for the
GroupNorm
module.Example:
GroupNorm model(GroupNormOptions(2, 2).eps(2e-5).affine(false));
Public Functions
-
GroupNormOptions
(int64_t num_groups, int64_t num_channels)¶
-
auto
num_groups
(const int64_t &new_num_groups) -> decltype(*this)¶ number of groups to separate the channels into
-
auto
num_groups
(int64_t &&new_num_groups) -> decltype(*this)¶
-
const int64_t &
num_groups
() const noexcept¶
-
int64_t &
num_groups
() noexcept¶
-
auto
num_channels
(const int64_t &new_num_channels) -> decltype(*this)¶ number of channels expected in input
-
auto
num_channels
(int64_t &&new_num_channels) -> decltype(*this)¶
-
const int64_t &
num_channels
() const noexcept¶
-
int64_t &
num_channels
() noexcept¶
-
auto
eps
(const double &new_eps) -> decltype(*this)¶ a value added to the denominator for numerical stability. Default: 1e-5
-
auto
eps
(double &&new_eps) -> decltype(*this)¶
-
const double &
eps
() const noexcept¶
-
double &
eps
() noexcept¶
-
auto
affine
(const bool &new_affine) -> decltype(*this)¶ a boolean value that when set to
true
, this module has learnable per-channel affine parameters initialized to ones (for weights) and zeros (for biases).Default:
true
.
-
auto
affine
(bool &&new_affine) -> decltype(*this)¶
-
const bool &
affine
() const noexcept¶
-
bool &
affine
() noexcept¶
-