R: Predictions with LiblineaR model

predict.LiblineaR {LiblineaR}

R Documentation

Predictions with LiblineaR model

Description

The function applies a model (classification or regression) produced by the LiblineaR function to every row of a data matrix and returns the model predictions.

Usage

## S3 method for class 'LiblineaR'
predict(object, newx, proba = FALSE,
  decisionValues = FALSE, ...)

Arguments

`object`	Object of class `"LiblineaR"`, created by `LiblineaR`.
`newx`	An n x p matrix containing the new input data. A vector will be transformed to a n x 1 matrix. A sparse matrix (from SparseM package) will also work.
`proba`	Logical indicating whether class probabilities should be computed and returned. Only possible if the model was fitted with `type`=0, `type`=6 or `type`=7, i.e. a Logistic Regression. Default is `FALSE`.
`decisionValues`	Logical indicating whether model decision values should be computed and returned. Only possible for classification models (`type`<10). Default is `FALSE`.
`...`	Currently not used

Value

By default, the returned value is a list with a single entry:

predictions

A vector of predicted labels (or values for regression).

If proba is set to TRUE, and the model is a logistic regression, an additional entry is returned:

probabilities

An n x k matrix (k number of classes) of the class probabilities. The columns of this matrix are named after class labels.

If decisionValues is set to TRUE, and the model is not a regression model, an additional entry is returned:

decisionValues

An n x k matrix (k number of classes) of the model decision values. The columns of this matrix are named after class labels.

Note

If the data on which the model has been fitted have been centered and/or scaled, it is very important to apply the same process on the newx data as well, with the scale and center values of the training data.

Author(s)

Thibault Helleputte thibault.helleputte@dnalytics.com and
Pierre Gramme pierre.gramme@dnalytics.com and
Jerome Paul jerome.paul@dnalytics.com.
Based on C/C++-code by Chih-Chung Chang and Chih-Jen Lin

References

For more information on 'LIBLINEAR' itself, refer to:
R.-E. Fan, K.-W. Chang, C.-J. Hsieh, X.-R. Wang, and C.-J. Lin.
LIBLINEAR: A Library for Large Linear Classification,
Journal of Machine Learning Research 9(2008), 1871-1874.
http://www.csie.ntu.edu.tw/~cjlin/liblinear