Predict Method for Graph-SVM Fits

Share:

Description

Obtains predictions from a fitted graphSVM object.

Usage

1
2
  ## S3 method for class 'graphSVM'
predict(object, newdata, type = "response", ...)

Arguments

object

a fitted object of class inheriting from 'graphSVM'

newdata

a matrix with variables to predict

type

response gives the predictions class gives the predicted classes.

...

currently ignored.

Value

the predictions.

Author(s)

Marc Johannes JohannesMarc@gmail.com

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
## Not run: 
library(pathClass)
data(example_data)
matched <- matchMatrices(x=x, adjacency=adjacency.matrix, mapping=mapping)
dk <- calc.diffusionKernel(L=matched$adjacency, is.adjacency=TRUE, beta=0) # beta should be tuned
fit <- fit.graph.svm(matched$x[1:5,], y[1:5], DEBUG=TRUE, mapping=matched$mapping, 
diffusionKernel=dk)
predict(fit, newdata=matched$x[6:10,])

## End(Not run)