Package |
|
---|---|
mlr3pipelines: Preprocessing Operators and Pipelines for 'mlr3' |
|
Building Blocks |
|
PipeOp Base Class |
|
Graph Base Class |
|
Task Preprocessing Base Class |
|
Simple Task Preprocessing Base Class |
|
Imputation Base Class |
|
Graph Tools |
|
PipeOp Composition Operator |
|
Disjoint Union of Graphs |
|
Create Disjoint Graph Union of Copies of a Graph |
|
PipeOps |
|
Dictionary of PipeOps |
|
Shorthand PipeOp Constructor |
|
Box-Cox Transformation of Numeric Features |
|
Path Branching |
|
Chunk Input into Multiple Outputs |
|
Class Balancing |
|
Majority Vote Prediction |
|
Class Weights for Sample Weighting |
|
Apply a Function to each Column of a Task |
|
Collapse Factors |
|
Change Column Roles of a Task |
|
Copy Input Multiple Times |
|
Preprocess Date Features |
|
Factor Encoding |
|
Conditional Target Value Impact Encoding |
|
Impact Encoding with Random Intercept Models |
|
Aggregate Features from Multiple Inputs |
|
Feature Filtering |
|
Fix Factor Levels |
|
Split Numeric Features into Equally Spaced Bins |
|
Independent Component Analysis |
|
Impute Features by a Constant |
|
Impute Numerical Features by Histogram |
|
Impute Features by Fitting a Learner |
|
Impute Numerical Features by their Mean |
|
Impute Numerical Features by their Median |
|
Impute Features by their Mode |
|
Out of Range Imputation |
|
Impute Features by Sampling |
|
Kernelized Principle Component Analysis |
|
Wrap a Learner into a PipeOp |
|
Wrap a Learner into a PipeOp with Cross-validated Predictions as Features |
|
Add Missing Indicator Columns |
|
Transform Columns by Constructing a Model Matrix |
|
Explicate a Multiplicity |
|
Implicate a Multiplicity |
|
Add Features According to Expressions |
|
Non-negative Matrix Factorization |
|
Simply Push Input Forward |
|
Split a Classification Task into Binary Classification Tasks |
|
Unite Binary Classification Tasks |
|
Principle Component Analysis |
|
Wrap another PipeOp or Graph as a Hyperparameter |
|
Split Numeric Features into Quantile Bins |
|
Project Numeric Features onto a Randomly Sampled Subspace |
|
Generate a Randomized Response Prediction |
|
Weighted Prediction Averaging |
|
Remove Constant Features |
|
Rename Columns |
|
Replicate the Input as a Multiplicity |
|
Center and Scale Numeric Features |
|
Scale Numeric Features with Respect to their Maximum Absolute Value |
|
Linearly Transform Numeric Features to Match Given Boundaries |
|
Remove Features Depending on a Selector |
|
SMOTE Balancing |
|
Normalize Data Row-wise |
|
Subsampling |
|
Invert Target Transformations |
|
Transform a Target by a Function |
|
Linearly Transform a Numeric Target to Match Given Boundaries |
|
Bag-of-word Representation of Character Features |
|
Change the Threshold of a Classification Prediction |
|
Tune the Threshold of a Classification Prediction |
|
Unbranch Different Paths |
|
Transform a Target without an Explicit Inversion |
|
Interface to the vtreat Package |
|
Yeo-Johnson Transformation of Numeric Features |
|
Learners |
|
Encapsulate a Graph as a Learner |
|
Optimized Weighted Average of Features for Classification and Regression |
|
Helpers |
|
|
Selector Functions |
Conversion to mlr3pipelines Graph |
|
Assertion for mlr3pipelines Graph |
|
Conversion to mlr3pipelines PipeOp |
|
Assertion for mlr3pipelines PipeOp |
|
Test for NO_OP |
|
No-Op Sentinel Used for Alternative Branching |
|
Remove NO_OPs from a List |
|
Abstract PipeOps |
|
Ensembling Base Class |
|
PipeOp Type Inference |
|
Add a Class Hierarchy to the Cache |
|
Reset the Class Hierarchy Cache |
|
Add Autoconvert Function to Conversion Register |
|
Reset Autoconvert Register |