Skip to main content

base_models

Base forest-related models using LightGBM as the gradient booster.

Classes​

BaseLGBMRandomForest​

class BaseLGBMRandomForest(    gradient_boosting: bool = False,    num_leaves: Optional[int] = None,    max_depth: Optional[int] = None,    subsample_for_bin: Optional[int] = None,    num_iterations: Optional[int] = None,    learning_rate: Optional[float] = None,    reg_alpha: Optional[float] = None,    reg_lambda: Optional[float] = None,    bagging_freq: Optional[float] = None,    bagging_fraction: Optional[float] = None,    feature_fraction: Optional[float] = None,    early_stopping_rounds: Optional[int] = None,    verbose: Optional[int] = None,    min_split_gain: Optional[float] = None,    **kwargs: Any,):

Implements an (optionally Gradient Boosted) Random Forest from LightGBM.

Ancestors​

  • bitfount.models.base_models._BaseModel
  • bitfount.models.base_models._BaseModelRegistryMixIn
  • bitfount.types._BaseSerializableObjectMixIn
  • abc.ABC
  • typing.Generic

Subclasses​

Methods​

def deserialize(self, filename: Union[str, os.PathLike]) ‑> None:

Deserialize model.

def evaluate(    self,    test_dl: Optional[bitfount.data.dataloaders._BitfountDataLoader] = None,    *args: Any,    **kwargs: Any,) ‑> Tuple[numpy.ndarray, numpy.ndarray]:

Perform inference on test set and save dictionary of metrics.

def fit(    self,    data: Optional[BaseSource] = None,    *args: Any,    **kwargs: Any,) ‑> None:

Trains a model using the training set provided by the BaseSource object.

def get_params(self) ‑> Dict[~KT, ~VT]:

Create an instance of the model.

def serialize(self, filename: Union[str, os.PathLike]) ‑> None:

Serialize model.

Variables​