Skip to content

0.6.0 - 2020-06-09


  • Added a new base class called SupervisedTransformer from which supervised transformers inherit from. Before this, supervised transformers has a is_supervised property.


  • Added compose.SelectType, which allows selecting feature subsets based on their type.
  • Added a score_one method to compose.Pipeline so that estimators from the anomaly module can be pipelined.
  • Added compose.Grouper, which allows applying transformers within different subgroups.


  • Added datasets.Music, which is a dataset for multi-output binary classification.
  • Added datasets.synth.Friedman, which is synthetic regression dataset.
  • The datasets.gen module has been renamed to datasets.synth
  • Each dataset now has a __repr__ method which displays some descriptive information.
  • Added datasets.Insects, which has 10 variants.


  • feature_extraction.Differ has been deprecated. We might put it back in a future if we find a better design.


  • impute.StatImputer has been completely refactored.


  • In metrics.SMAPE, instead of raising a ZeroDivisionError, the convention is now to use 0 when both y_true and y_pred are equal to 0.


  • Added the possibility to configure how the progress is printed in model_selection.progressive_val_score. For instance, the progress can now be printed to a file by providing the file argument.


  • Added multiclass.OutputCodeClassifier.
  • Added multiclass.OneVsOneClassifier.


  • Fixed a bug where multioutput.ClassifierChain and multioutput.RegressorChain could not be pickled.


  • Added stats.Shift, which can be used to compute statistics over a shifted version of a variable.
  • Added stats.Link, which can be used to compose univariate statistics. Univariate statistics can now be composed via the | operator.
  • Renamed stats.Covariance to stats.Cov.
  • Renamed stats.PearsonCorrelation to stats.PearsonCorr.
  • Renamed stats.AutoCorrelation to stats.AutoCorr.
  • Added stats.RollingCov, which computes covariance between two variables over a window.
  • Added stats.RollingPearsonCorr, which computes the Pearson correlation over a window.


  • Added a stream.iter_sql utility method to work with SQLAlchemy.
  • The target_name parameter of stream.iter_csv has been renamed to target. It can now be passed a list of values in order to support multi-output scenarios.
  • Added stream.iter_arff for handling ARFF files.


  • Cancelled the behavior where tree.DecisionTreeRegressor would raise an exception when no split was found.