xgboost
Public Member Functions | Public Attributes | List of all members
xgboost::LearnerModelParam Struct Reference

Basic model parameters, used to describe the booster. More...

#include <learner.h>

Collaboration diagram for xgboost::LearnerModelParam:
Collaboration graph

Public Member Functions

 LearnerModelParam ()=default
 
 LearnerModelParam (Context const *ctx, LearnerModelParamLegacy const &user_param, linalg::Tensor< float, 1 > base_margin, ObjInfo t, MultiStrategy multi_strategy)
 
 LearnerModelParam (LearnerModelParamLegacy const &user_param, ObjInfo t, MultiStrategy multi_strategy)
 
 LearnerModelParam (bst_feature_t n_features, linalg::Tensor< float, 1 > base_score, std::uint32_t n_groups, bst_target_t n_targets, MultiStrategy multi_strategy)
 
linalg::TensorView< float const, 1 > BaseScore (Context const *ctx) const
 
linalg::TensorView< float const, 1 > BaseScore (DeviceOrd device) const
 
void Copy (LearnerModelParam const &that)
 
bool IsVectorLeaf () const noexcept
 
bst_target_t OutputLength () const noexcept
 
bst_target_t LeafLength () const noexcept
 
bool Initialized () const
 

Public Attributes

bst_feature_t num_feature {0}
 The number of features. More...
 
std::uint32_t num_output_group {0}
 The number of classes or targets. More...
 
ObjInfo task {ObjInfo::kRegression}
 Current task, determined by objective. More...
 
MultiStrategy multi_strategy {MultiStrategy::kOneOutputPerTree}
 Strategy for building multi-target models. More...
 

Detailed Description

Basic model parameters, used to describe the booster.

Constructor & Destructor Documentation

◆ LearnerModelParam() [1/4]

xgboost::LearnerModelParam::LearnerModelParam ( )
default

◆ LearnerModelParam() [2/4]

xgboost::LearnerModelParam::LearnerModelParam ( Context const *  ctx,
LearnerModelParamLegacy const &  user_param,
linalg::Tensor< float, 1 >  base_margin,
ObjInfo  t,
MultiStrategy  multi_strategy 
)

◆ LearnerModelParam() [3/4]

xgboost::LearnerModelParam::LearnerModelParam ( LearnerModelParamLegacy const &  user_param,
ObjInfo  t,
MultiStrategy  multi_strategy 
)

◆ LearnerModelParam() [4/4]

xgboost::LearnerModelParam::LearnerModelParam ( bst_feature_t  n_features,
linalg::Tensor< float, 1 >  base_score,
std::uint32_t  n_groups,
bst_target_t  n_targets,
MultiStrategy  multi_strategy 
)
inline

Member Function Documentation

◆ BaseScore() [1/2]

linalg::TensorView<float const, 1> xgboost::LearnerModelParam::BaseScore ( Context const *  ctx) const

◆ BaseScore() [2/2]

linalg::TensorView<float const, 1> xgboost::LearnerModelParam::BaseScore ( DeviceOrd  device) const

◆ Copy()

void xgboost::LearnerModelParam::Copy ( LearnerModelParam const &  that)

◆ Initialized()

bool xgboost::LearnerModelParam::Initialized ( ) const
inline

◆ IsVectorLeaf()

bool xgboost::LearnerModelParam::IsVectorLeaf ( ) const
inlinenoexcept

◆ LeafLength()

bst_target_t xgboost::LearnerModelParam::LeafLength ( ) const
inlinenoexcept

◆ OutputLength()

bst_target_t xgboost::LearnerModelParam::OutputLength ( ) const
inlinenoexcept

Member Data Documentation

◆ multi_strategy

MultiStrategy xgboost::LearnerModelParam::multi_strategy {MultiStrategy::kOneOutputPerTree}

Strategy for building multi-target models.

◆ num_feature

bst_feature_t xgboost::LearnerModelParam::num_feature {0}

The number of features.

◆ num_output_group

std::uint32_t xgboost::LearnerModelParam::num_output_group {0}

The number of classes or targets.

◆ task

ObjInfo xgboost::LearnerModelParam::task {ObjInfo::kRegression}

Current task, determined by objective.


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