make.predictions: Make predictions on a test set

In zellerlab/siamcat: Statistical Inference of Associations between Microbial Communities And host phenoTypes

Make predictions on a test set

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	integer, 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

Examples

data(siamcat_example)

# Simple example
siamcat_example <- train.model(siamcat_example, method='lasso')
siamcat.pred <- make.predictions(siamcat_example)

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

zellerlab/siamcat documentation built on Feb. 1, 2024, 2:21 a.m.