mrf.nll {CRF}

Calculate MRF negative log-likelihood

Description

Calculate the negative log-likelihood of MRF model

Usage

mrf.nll(par, crf, instances, infer.method = infer.chain, ...)

Arguments

par

The parameter vector of CRF

crf

The CRF

instances

The training data matrix of MRF model

infer.method

The inference method used to compute the likelihood

...

Extra parameters need by the inference method

Details

This function calculates the negative log-likelihood of MRF model as well as the gradient. This function is intended to be called by optimization algorithm in training process. Before calling this function, the MRF sufficient statistics must be calculated and stored in object par.stat of CRF.

In the training data matrix instances, each row is an instance and each column corresponds a node in CRF.

Value

This function will return the value of MRF negative log-likilihood.

See Also

mrf.stat, mrf.update, train.mrf