sklearn-utilities 0.3.0

Utilities for scikit-learn.

Sklearn Utilities

Utilities for scikit-learn.

Installation

Install this via pip (or your favourite package manager):

pip install sklearn-utilities

API

See Docs for more information.

• EstimatorWrapperBase: base class for wrappers. Redirects all attributes which are not in the wrapper to the wrapped estimator.
• DataFrameWrapper: tries to convert every estimator output to a pandas DataFrame or Series.
• FeatureUnionPandas: a FeatureUnion that works with pandas DataFrames.
• IncludedColumnTransformerPandas, ExcludedColumnTransformerPandas: select columns by name.
• AppendPredictionToX: appends the prediction of y to X.
• AppendXPredictionToX: appends the prediction of X to X.
• DropByNoisePrediction: drops columns which has high importance in predicting noise.
• DropMissingColumns: drops columns with missing values above a threshold.
• DropMissingRowsY: drops rows with missing values in y. Use feature_engine.DropMissingData for X.
• IntersectXY: drops rows where the index of X and y do not intersect. Use with feature_engine.DropMissingData.
• IdTransformer: a transformer that does nothing.
• RecursiveFitSubtractRegressor: a regressor that recursively fits a regressor and subtracts the prediction from the target.
• SmartMultioutputEstimator: a MultiOutputEstimator that supports tuple of arrays in predict() and supports pandas Series and DataFrame.
• until_event(), since_event(): calculates the time since or until events (Series[bool])
• ComposeVarEstimator: composes mean and std/var estimators.
• DummyRegressorVar: DummyRegressor that returns 1.0 for std/var.
• TransformedTargetRegressorVar: TransformedTargetRegressor with std/var support.
• StandardScalerVar: StandardScaler with std/var support.

sklearn_utilities.dataset

• add_missing_values(): adds missing values to a dataset.

sklearn_utilities.torch

• PCATorch: faster PCA using PyTorch with GPU support.

sklearn_utilities.torch.skorch

• SkorchReshaper, SkorchCNNReshaper: reshapes X and y for nn.Linear and nn.Conv1d/2d respectively. (For nn.Conv2d, uses np.sliding_window_view().)
• AllowNaN: wraps a loss module and assign 0 to y and y_hat for indices where y contains NaN in forward()..

See also

• ml-tooling/best-of-ml-python

Contributors ✨

Thanks goes to these wonderful people (emoji key):

This project follows the all-contributors specification. Contributions of any kind welcome!