Cross entropy error function

Share:

Description

The function calculates the cross entropy error from the original and estimate parameters.

Usage

1
crossEntropyError(original, estimate)

Arguments

original

The original data matrix.

estimate

The calculated data matrix.

Details

This function can be used for the darch.errorFunction parameter of the darch function, but is only a valid error function if used in combination with the softmaxUnit activation function! It is not a valid value for the parameter rbm.errorFunction.

Value

A list with the name of the error function in the first entry and the error value in the second entry.

References

Rubinstein, R.Y., Kroese, D.P. (2004). The Cross-Entropy Method: A Unified Approach to Combinatorial Optimization, Monte-Carlo Simulation, and Machine Learning, Springer-Verlag, New York.

See Also

Other error functions: mseError, rmseError

Examples

1
2
3
4
5
## Not run: 
data(iris)
model <- darch(Species ~ ., iris, darch.errorFunction = "crossEntropyError")

## End(Not run)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.