NestedTabularMLAlgo
- class lightautoml.pipelines.ml.nested_ml_pipe.NestedTabularMLAlgo(ml_algo, tuner=None, refit_tuner=False, cv=5, n_folds=None)[source]
Bases:
TabularMLAlgo
,ImportanceEstimator
Wrapper for MLAlgo to make it trainable over nested folds.
Limitations - only for
TabularMLAlgo
.- property params
Parameters of ml_algo.
- init_params_on_input(train_valid_iterator)[source]
Init params depending on input data.
- Parameters:
train_valid_iterator (
TrainValidIterator
) – Iterator over input data.- Return type:
- Returns:
dict with model hyperparameters.
- fit_predict_single_fold(train, valid)[source]
Implements training and prediction on single fold.
- Parameters:
train (
Union
[NumpyDataset
,CSRSparseDataset
,PandasDataset
]) – TabularDataset to train.valid (
Union
[NumpyDataset
,CSRSparseDataset
,PandasDataset
]) – TabularDataset to validate.
- Return type:
- Returns:
Tuple (model, predicted_values).
- predict_single_fold(model, dataset)[source]
Model prediction on a dataset.
- Parameters:
model (
Any
) – Model.dataset (
Union
[NumpyDataset
,CSRSparseDataset
,PandasDataset
]) – Dataset.
- Return type:
- Returns:
Predictions.
- fit(train_valid)[source]
Just to be compatible with
ImportanceEstimator
.- Parameters:
train_valid (
TrainValidIterator
) – Classic cv iterator.