secretflow.ml.boost.sgb_v.factory

Classes:

SGBFactory()

You can customize your own boosting algorithms which are based on any combination of ideas of secureboost, XGB, and lightGBM.

class secretflow.ml.boost.sgb_v.factory.SGBFactory

Bases: object

You can customize your own boosting algorithms which are based on any combination of ideas of secureboost, XGB, and lightGBM. The parameters for the produced booster algorithm depends on what components it consists of. See components’ parameters.

params_dict

A dict contain params for the factory, booster and its components.

Type:

dict

factory_params

validated params for the factory.

Type:

SGBFactoryParams

heu

the device for HE computations. must be set before training.

Methods:

__init__()
set_params(params)	Set params by a dictionary.
set_heu(heu)
get_params([detailed])	get the params set
fit(dataset, label)
train(params, dataset, label)

__init__()

set_params(params: dict)

Set params by a dictionary.

set_heu(heu: HEU)

get_params(detailed: bool = False) → dict

get the params set

Parameters:

detailed (bool, optional) – If include default settings. Defaults to False.

Returns:

current params.

Return type:

dict

fit(dataset: Union[FedNdarray, VDataFrame], label: Union[FedNdarray, VDataFrame]) → SgbModel

train(params: dict, dataset: Union[FedNdarray, VDataFrame], label: Union[FedNdarray, VDataFrame]) → SgbModel

• secretflow.ml.boost.sgb_v.factory.booster
  • GlobalOrdermapBooster
  • secretflow.ml.boost.sgb_v.factory.booster.global_ordermap_booster
• secretflow.ml.boost.sgb_v.factory.components
  • OrderMapManager
  • GradientEncryptor
  • Sampler
  • LeafManager
  • DataPreprocessor
  • ModelBuilder
  • LossComputer
  • TreeTrainer
  • NodeSelector
  • LevelWiseCache
  • Shuffler
  • BucketSumCalculator
  • SplitFinder
  • SplitTreeBuilder
  • LevelWiseTreeTrainer
  • secretflow.ml.boost.sgb_v.factory.components.bucket_sum_calculator
  • secretflow.ml.boost.sgb_v.factory.components.cache
  • secretflow.ml.boost.sgb_v.factory.components.data_preprocessor
  • secretflow.ml.boost.sgb_v.factory.components.gradient_encryptor
  • secretflow.ml.boost.sgb_v.factory.components.leaf_manager
  • secretflow.ml.boost.sgb_v.factory.components.loss_computer
  • secretflow.ml.boost.sgb_v.factory.components.model_builder
  • secretflow.ml.boost.sgb_v.factory.components.node_selector
  • secretflow.ml.boost.sgb_v.factory.components.order_map_manager
  • secretflow.ml.boost.sgb_v.factory.components.sampler
  • secretflow.ml.boost.sgb_v.factory.components.shuffler
  • secretflow.ml.boost.sgb_v.factory.components.split_finder
  • secretflow.ml.boost.sgb_v.factory.components.split_tree_builder
  • secretflow.ml.boost.sgb_v.factory.components.tree_trainer
  • secretflow.ml.boost.sgb_v.factory.components.component

secretflow.ml.boost.sgb_v.factory.factory

Classes:

TreeGrowingMethod(value)	An enumeration.
SGBFactoryParams(tree_growing_method)
SGBFactory()	You can customize your own boosting algorithms which are based on any combination of ideas of secureboost, XGB, and lightGBM.

class secretflow.ml.boost.sgb_v.factory.factory.TreeGrowingMethod(value)

Bases: Enum

An enumeration.

Attributes:

LEVEL
LEAF

LEVEL = 'level'

LEAF = 'leaf'

class secretflow.ml.boost.sgb_v.factory.factory.SGBFactoryParams(tree_growing_method: secretflow.ml.boost.sgb_v.factory.factory.TreeGrowingMethod = <TreeGrowingMethod.LEVEL: 'level'>)

Bases: object

Attributes:

tree_growing_method

Methods:

__init__([tree_growing_method])

tree_growing_method: TreeGrowingMethod = 'level'

__init__(tree_growing_method: TreeGrowingMethod = TreeGrowingMethod.LEVEL) → None

class secretflow.ml.boost.sgb_v.factory.factory.SGBFactory

Bases: object

You can customize your own boosting algorithms which are based on any combination of ideas of secureboost, XGB, and lightGBM. The parameters for the produced booster algorithm depends on what components it consists of. See components’ parameters.

params_dict

A dict contain params for the factory, booster and its components.

Type:

dict

factory_params

validated params for the factory.

Type:

SGBFactoryParams

heu

the device for HE computations. must be set before training.

Methods:

__init__()
set_params(params)	Set params by a dictionary.
set_heu(heu)
get_params([detailed])	get the params set
fit(dataset, label)
train(params, dataset, label)

__init__()

set_params(params: dict)

Set params by a dictionary.

set_heu(heu: HEU)

get_params(detailed: bool = False) → dict

get the params set

Parameters:

detailed (bool, optional) – If include default settings. Defaults to False.

Returns:

current params.

Return type:

dict

fit(dataset: Union[FedNdarray, VDataFrame], label: Union[FedNdarray, VDataFrame]) → SgbModel

train(params: dict, dataset: Union[FedNdarray, VDataFrame], label: Union[FedNdarray, VDataFrame]) → SgbModel