make.predictions: Make predictions on a test set

In KonradZych/SIAMCAT: Statistical Inference of Associations between Microbial Communities And host phenoTypes

Description

This function takes a siamcat-class-object containing a model trained by train.model and performs predictions on a given test-set.

Usage

make.predictions(siamcat, siamcat.holdout = NULL,
    normalize.holdout = TRUE, verbose = 1)

Arguments

siamcat	object of class siamcat-class
siamcat.holdout	optional, object of class siamcat-class on which to make predictions, defaults to NULL
normalize.holdout	boolean, should the holdout features be normalized with a frozen normalization (see normalize.features) using the normalization parameters in siamcat?, defaults to TRUE
verbose	control output: 0 for no output at all, 1 for only information about progress and success, 2 for normal level of information and 3 for full debug information, defaults to 1

Details

This functions uses the model in the model_list-slot of the siamcat object to make predictions on a given test set. The test set can either consist of the test instances in the cross- validation, saved in the data_split-slot of the same siamcat object, or a completely external feature set, given in the form of another siamcat object (siamcat.holdout).

Value

object of class siamcat-class with the slot pred_matrix filled or a matrix containing the predictions for the holdout set

Examples

    data(siamcat_example)
    # Simple example
    siamcat.pred <- make.predictions(siamcat_example)

    # Predictions on a holdout-set
    ## Not run: pred.mat <- make.predictions(siamcat.trained, siamcat.holdout,
    normalize.holdout=TRUE)
## End(Not run)

KonradZych/SIAMCAT documentation built on May 17, 2019, 6:20 p.m.