7. Overview of available Pipeline Steps

The following is a list of all available steps that can be used to create a pipeline by name. The overview is sorted based on the type of the step: Transformers or Models (Estimators).

• The column Name refers to the string-name of the respective step, i.e. how it should be specified when passed to e.g., the PipelineCreator.

• The column Description gives a short description of what the step is doing.

• The column Class either indicates the underlying scikit-learn class of the respective pipeline step together with a link to the class in the scikit-learn documentation (follow the link to see the valid parameters) or indicates the class in julearn, so one can have a closer look at it in julearn’s API Reference.

For feature transformations, the Transformers are to be used with the PipelineCreator and for target transformations, the Transformers are to be used with the TargetPipelineCreator.

7.1. Transformers

Scalers

Name	Description	Class
zscore	Removing mean and scale to unit variance	StandardScaler
scaler_robust	Removing median and scale to IQR	RobustScaler
scaler_minmax	Scale to a given range	MinMaxScaler
scaler_maxabs	Scale by max absolute value	MaxAbsScaler
scaler_normalizer	Normalize to unit norm	Normalizer
scaler_quantile	Transform to uniform or normal distribution (robust)	QuantileTransformer
scaler_power	Gaussianise data	PowerTransformer

Feature Selection

Name	Description	Class
select_univariate	Removing mean and scale to unit variance	GenericUnivariateSelect
select_percentile	Rank and select percentile	SelectPercentile
select_k	Rank and select K	SelectKBest
select_fdr	Select based on estimated FDR	SelectFdr
select_fpr	Select based on FPR threshold	SelectFpr
select_fwe	Select based on FWE threshold	SelectFwe
select_variance	Remove low variance features	VarianceThreshold

DataFrame operations

Name	Description	Class
confound_removal	Removing confounds from features, by subtracting the prediction of each feature given all confounds. By default this is equal to “independently regressing out the confounds from the features”	ConfoundRemover
drop_columns	Drop columns from the DataFrame	DropColumns
change_column_types	Change the type of a column in a DataFrame	ChangeColumnTypes
filter_columns	Filter columns in a DataFrame	FilterColumns

Decomposition

Name	Description	Class
pca	Principal Component Analysis	PCA

Custom

Name	Description	Class
cbpm	Connectome-based Predictive Modeling (CBPM)	CBPM

7.2. Models (Estimators)

Support Vector Machines

Name	Description	Class	Binary	Multiclass	Regression
svm	Support Vector Machine	SVC and SVR	Y	Y	Y

Ensemble

Name	Description	Class	Binary	Multiclass	Regression
rf	Random Forest	RandomForestClassifier and RandomForestRegressor	Y	Y	Y
et	Extra-Trees	ExtraTreesClassifier and ExtraTreesRegressor	Y	Y	Y
adaboost	AdaBoost	AdaBoostClassifier and AdaBoostRegressor	Y	Y	Y
bagging	Bagging	BaggingClassifier and BaggingRegressor	Y	Y	Y
gradientboost	Gradient Boosting	GradientBoostingClassifier and GradientBoostingRegressor	Y	Y	Y
stacking	Stacking	StackingClassifier and StackingRegressor	Y	Y	Y

Gaussian Processes

Name	Description	Class	Binary	Multiclass	Regression
gauss	Gaussian Process	GaussianProcessClassifier and GaussianProcessRegressor	Y	Y	Y

Linear Models

Name	Description	Class	Binary	Multiclass	Regression
logit	Logistic Regression (aka logit, MaxEnt).	LogisticRegression	Y	Y	N
logitcv	Logistic Regression CV (aka logit, MaxEnt).	LogisticRegressionCV	Y	Y	N
linreg	Least Squares regression.	LinearRegression	N	N	Y
ridge	Linear least squares with l2 regularization.	RidgeClassifier and Ridge	Y	Y	Y
ridgecv	Ridge regression with built-in cross-validation.	RidgeClassifierCV and RidgeCV	Y	Y	Y
sgd	Linear model fitted by minimizing a regularized empirical loss with SGD	SGDClassifier and SGDRegressor	Y	Y	Y

Naive Bayes

Name	Description	Class	Binary	Multiclass	Regression
nb_bernoulli	Multivariate Bernoulli models.	BernoulliNB	Y	Y	N
nb_categorical	Categorical features.	CategoricalNB	Y	Y	N
nb_complement	Complement Naive Bayes	ComplementNB	Y	Y	N
nb_gaussian	Gaussian Naive Bayes	GaussianNB	Y	Y	N
nb_multinomial	Multinomial models	MultinomialNB	Y	Y	N

Dynamic Selection

Name	Description	Class	Binary	Multiclass	Regression
ds	Support for DESlib models	DynamicSelection	Y	Y	Y

Dummy

Name	Description	Class	Binary	Multiclass	Regression
dummy	Use simple rules (without features).	DummyClassifier and DummyRegressor	Y	Y	Y