Caffe2 - C++ API
A deep learning, cross platform ML framework
Static Public Member Functions | Static Public Attributes
std::numeric_limits< c10::Half > Class Template Reference

Static Public Member Functions

static constexpr c10::Half min ()
 
static constexpr c10::Half lowest ()
 
static constexpr c10::Half max ()
 
static constexpr c10::Half epsilon ()
 
static constexpr c10::Half round_error ()
 
static constexpr c10::Half infinity ()
 
static constexpr c10::Half quiet_NaN ()
 
static constexpr c10::Half signaling_NaN ()
 
static constexpr c10::Half denorm_min ()
 

Static Public Attributes

static constexpr bool is_specialized = true
 
static constexpr bool is_signed = true
 
static constexpr bool is_integer = false
 
static constexpr bool is_exact = false
 
static constexpr bool has_infinity = true
 
static constexpr bool has_quiet_NaN = true
 
static constexpr bool has_signaling_NaN = true
 
static constexpr auto has_denorm = numeric_limits<float>::has_denorm
 
static constexpr auto has_denorm_loss
 
static constexpr auto round_style = numeric_limits<float>::round_style
 
static constexpr bool is_iec559 = true
 
static constexpr bool is_bounded = true
 
static constexpr bool is_modulo = false
 
static constexpr int digits = 11
 
static constexpr int digits10 = 3
 
static constexpr int max_digits10 = 5
 
static constexpr int radix = 2
 
static constexpr int min_exponent = -13
 
static constexpr int min_exponent10 = -4
 
static constexpr int max_exponent = 16
 
static constexpr int max_exponent10 = 4
 
static constexpr auto traps = numeric_limits<float>::traps
 
static constexpr auto tinyness_before
 

Detailed Description

template<>
class std::numeric_limits< c10::Half >

Definition at line 229 of file Half-inl.h.

Field Documentation

constexpr auto std::numeric_limits< c10::Half >::has_denorm_loss
static
Initial value:
=
numeric_limits<float>::has_denorm_loss

Definition at line 239 of file Half-inl.h.

constexpr auto std::numeric_limits< c10::Half >::tinyness_before
static
Initial value:
=
numeric_limits<float>::tinyness_before

Definition at line 254 of file Half-inl.h.


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