Available Pipeline Steps
The following is a list of all the available steps that can be used to create a pipeline by name.
Features Preprocessing
Scalers
Name (str) |
Description |
Class |
---|---|---|
|
Removing mean and scale to unit variance |
|
|
Removing median and scale to IQR |
|
|
Scale to a given range |
|
|
Scale by max absolute value |
|
|
Normalize to unit norm |
|
|
Transform to uniform or normal distribution (robust) |
|
|
Gaussianise data |
Feature Selection
Name (str) |
Description |
Class |
---|---|---|
|
Removing mean and scale to unit variance |
|
|
Rank and select percentile |
|
|
Rank and select K |
|
|
Select based on estimated FDR |
|
|
Select based on FPR threshold |
|
|
Select based on FWE threshold |
|
|
Remove low variance features |
Confound Removal
Name (str) |
Description |
Class |
---|---|---|
|
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” |
Decomposition
Name (str) |
Description |
Class |
---|---|---|
|
Principal Component Analysis |
Target Preprocessing
Target Scalers
Name (str) |
Description |
Class |
---|---|---|
|
Removing mean and scale to unit variance |
Target Confound Removal
Name (str) |
Description |
Class |
---|---|---|
|
removing confounds from target, by subtracting the prediction of the target given all confounds. By default this is equal to “regressing out the confounds from the target” |
Models
Support Vector Machines
Name (str) |
Description |
Class |
Binary |
Multiclass |
Regression |
---|---|---|---|---|---|
|
Support Vector Machine |
Y |
Y |
Y |
Ensemble
Name (str) |
Description |
Class |
Binary |
Multiclass |
Regression |
---|---|---|---|---|---|
|
Random Forest |
Y |
Y |
Y |
|
|
Extra-Trees |
Y |
Y |
Y |
|
|
AdaBoost |
Y |
Y |
Y |
|
|
Bagging |
Y |
Y |
Y |
|
|
Gradient Boosting |
Y |
Y |
Y |
Gaussian Processes
Name (str) |
Description |
Class |
Binary |
Multiclass |
Regression |
---|---|---|---|---|---|
|
Gaussian Process |
Y |
Y |
Y |
Linear Models
Name (str) |
Description |
Class |
Binary |
Multiclass |
Regression |
---|---|---|---|---|---|
|
Logistic Regression (aka logit, MaxEnt). |
Y |
Y |
N |
|
|
Logistic Regression CV (aka logit, MaxEnt). |
Y |
Y |
N |
|
|
Least Squares regression. |
N |
N |
Y |
|
|
Linear least squares with l2 regularization. |
RidgeClassifier and Ridge |
Y |
Y |
Y |
|
Ridge regression with built-in cross-validation. |
Y |
Y |
Y |
|
|
Linear model fitted by minimizing a regularized empirical loss with SGD |
Y |
Y |
Y |
Naive Bayes
Name (str) |
Description |
Class |
Binary |
Multiclass |
Regression |
---|---|---|---|---|---|
|
Multivariate Bernoulli models. |
Y |
Y |
N |
|
|
Categorical features. |
Y |
Y |
N |
|
|
Complement Naive Bayes |
Y |
Y |
N |
|
|
Gaussian Naive Bayes |
Y |
Y |
N |
|
|
Multinomial models |
Y |
Y |
N |
Dummy
Name (str) |
Description |
Class |
Binary |
Multiclass |
Regression |
---|---|---|---|---|---|
|
Use simple rules (without features). |
Y |
Y |
Y |