`bosk.block.zoo.models.classification.ferns`#

Module Contents#

Classes#

RandomFernsBlock

Random Ferns Classifier Block.

class bosk.block.zoo.models.classification.ferns.RandomFernsBlock(n_groups=10, n_ferns_in_group=20, fern_size=7, kind='unary', bootstrap=False, n_jobs=None, random_state=None)#

Bases: bosk.block.base.BaseBlock

Random Ferns Classifier Block.

Parameters:

n_groups (int) – Number of ferns groups (like a number of estimators).
n_ferns_in_group (int) – Number of ferns in a group.
fern_size (int) – Number of tests in fern.
kind (str) – Kind of tests (‘unary’ or ‘binary’).
bootstrap (bool) – Apply data bootstrap or not.
n_jobs (Optional[int]) – Number of threads.
random_state (Optional[int]) – Random state.

Input slots#

Fit inputs#

X: Input features.

y: Ground truth labels.

Transform inputs#

X: Input features.

Output slots#

probas: Predicted probabilities.

meta#

__getstate__()#

Return type:: dict

__setstate__(state)#

Parameters:: state (dict) –

_classifier_init(y)#

_parallel_calc_bucket_stats(bucket_indices, y, group_data_indices)#

_make_ferns(X, prng_key)#

_apply_ferns(X, ferns)#

fit(inputs)#

Fit the Random Ferns Block. The implementation is device-agnostic.

Parameters:: inputs (bosk.block.base.BlockInputData) –
Return type:: RandomFernsBlock

_predict_proba(X)#

transform(inputs)#

Transform the given input data, i.e. compute values for each output slot.

Parameters:: inputs (bosk.block.base.BlockInputData) – Block input data for the transforming stage.
Returns:: Outputs calculated for the given inputs.
Return type:: bosk.block.base.TransformOutputData

bosk.block.zoo.models.classification.ferns#