train.mrf {CRF}R Documentation

Train MRF model

Description

Train the MRF model to estimate the parameters

Usage

train.mrf(
  crf,
  instances,
  nll = mrf.nll,
  infer.method = infer.chain,
  ...,
  trace = 0
)

Arguments

crf

The CRF

instances

The training data matrix of CRF model

nll

The function to calculate negative log likelihood

infer.method

The inference method used to compute the likelihood

...

Extra parameters need by the inference method

trace

Non-negative integer to control the tracing informtion of the optimization process

Details

This function trains the Markov Random Fields (MRF) model, which is a simple variant of CRF model.

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

Value

This function will directly modify the CRF and return the same CRF.

See Also

mrf.update, mrf.stat, mrf.nll, make.crf


[Package CRF version 0.4-3 Index]