lale.lib.lale.topk_voting_classifier module¶
- class lale.lib.lale.topk_voting_classifier.TopKVotingClassifier(*, estimator=None, optimizer=None, args_to_optimizer=None, k=10)¶
Bases:
PlannedIndividualOp
This operator creates a voting ensemble from top k performing pipelines from the given planned pipeline.
This documentation is auto-generated from JSON schemas.
- Parameters
estimator (union type, not for optimizer, default None) –
Planned Lale individual operator or pipeline.
operator of None
or None
optimizer (union type, optional, not for optimizer, default None) –
- Optimizer class to be used during the two stages of optimization.
Default of None uses Hyperopt internally. Currently, only Hyperopt is supported as an optimizer.
operator of None
or None
args_to_optimizer (union type, optional, not for optimizer, default None) –
- Dictionary of keyword arguments required to be used for the given optimizer
as applicable for the given task. For example, max_evals, cv, scoring etc. for Hyperopt. If None, default values for the optimizer would be used.
dict
or None
k (integer, >=1, optional, not for optimizer, default 10) –
- Number of top pipelines to be used for the voting ensemble. If the number of
successful trials of the optimizer are less than k, the ensemble will use only successful trials.
- fit(X, y=None, **fit_params)¶
Train the operator.
Note: The fit method is not available until this operator is trainable.
Once this method is available, it will have the following signature:
- Parameters
X (any type) –
y (any type) –
- predict(X, **predict_params)¶
Make predictions.
Note: The predict method is not available until this operator is trained.
Once this method is available, it will have the following signature:
- Parameters
X (any type) –
- Returns
result
- Return type
any type