secretflow.ml.boost.sgb_v.core.preprocessing

secretflow.ml.boost.sgb_v.core.preprocessing.params

Classes:

RegType(value)	An enumeration.
LabelHolderInfo(seed, reg_lambda, gamma, ...)
SGBTrainParams(num_boost_round, max_depth, ...)

class secretflow.ml.boost.sgb_v.core.preprocessing.params.RegType(value)

基类：Enum

An enumeration.

Attributes:

Linear
Logistic

Linear = 'linear'

Logistic = 'logistic'

class secretflow.ml.boost.sgb_v.core.preprocessing.params.LabelHolderInfo(seed: int, reg_lambda: float, gamma: float, learning_rate: float, base_score: float, sample_num: int, subsample_rate: float, obj_type: secretflow.ml.boost.sgb_v.core.preprocessing.params.RegType)

基类：object

Attributes:

seed
reg_lambda
gamma
learning_rate
base_score
sample_num
subsample_rate
obj_type

Methods:

__init__(seed, reg_lambda, gamma, ...)

seed: int

reg_lambda: float

gamma: float

learning_rate: float

base_score: float

sample_num: int

subsample_rate: float

obj_type: RegType

__init__(seed: int, reg_lambda: float, gamma: float, learning_rate: float, base_score: float, sample_num: int, subsample_rate: float, obj_type: RegType) → None

class secretflow.ml.boost.sgb_v.core.preprocessing.params.SGBTrainParams(num_boost_round: int, max_depth: int, learning_rate: float, objective: secretflow.ml.boost.sgb_v.core.preprocessing.params.RegType, reg_lambda: float, gamma: float, subsample: float, colsample_by_tree: float, base_score: float, sketch_eps: float, seed: int, fixed_point_parameter: int)

基类：object

Attributes:

num_boost_round
max_depth
learning_rate
objective
reg_lambda
gamma
subsample
colsample_by_tree
base_score
sketch_eps
seed
fixed_point_parameter

Methods:

__init__(num_boost_round, max_depth, ...)

num_boost_round: int

max_depth: int

learning_rate: float

objective: RegType

reg_lambda: float

gamma: float

subsample: float

colsample_by_tree: float

base_score: float

sketch_eps: float

seed: int

fixed_point_parameter: int

__init__(num_boost_round: int, max_depth: int, learning_rate: float, objective: RegType, reg_lambda: float, gamma: float, subsample: float, colsample_by_tree: float, base_score: float, sketch_eps: float, seed: int, fixed_point_parameter: int) → None

secretflow.ml.boost.sgb_v.core.preprocessing.preprocessing

Functions:

prepare_dataset(ds)	check data setting and get total shape.
validate_sgb_params_dict(params)

secretflow.ml.boost.sgb_v.core.preprocessing.preprocessing.prepare_dataset(ds: Union[FedNdarray, VDataFrame]) → Tuple[FedNdarray, Tuple[int, int]]

check data setting and get total shape.

参数:

ds – input dataset

返回:

dataset in unified type Second: shape concat all partition.

返回类型:

First

secretflow.ml.boost.sgb_v.core.preprocessing.preprocessing.validate_sgb_params_dict(params: dict) → SGBTrainParams