calc.loss: Calculate total loss of model (Squared hinge loss).

In textreg: n-Gram Text Regression, aka Concise Comparative Summarization

Description

Calculate the loss for a model in predicting the -1/+1 labeling. If new text and labeling given, then calc loss on the new text and labeling. This can be useful for cross validation and train-test splits.

Usage

calc.loss(model.blob, new.text = NULL, new.labeling = NULL,
  loss = c("square.hinge", "square", "hinge"))

Arguments

model.blob	The model returned from textreg
new.text	New text (string or tm Corpus) to predict labeling for
new.labeling	Labeling to go with new text.
loss	Type of loss to calc for.

Value

Three numbers: total loss, loss from prediction, loss from penalty term

Examples

data( testCorpora )
res = textreg( c( "", "", "A", "A" ), c( -1, -1, 1, 1 ), C=1, Lq=1,
          convergence.threshold=0.00000001, verbosity=0 )
calc.loss( res )
calc.loss( res, new.text=c("A B C A"), new.labeling=c(1) )

textreg documentation built on May 2, 2019, 8:34 a.m.