infer.conditional {CRF} | R Documentation |
Conditional inference method
Description
Computing the partition function and marginal probabilities
Usage
infer.conditional(crf, clamped, infer.method, ...)
Arguments
crf |
The CRF |
clamped |
The vector of fixed values for clamped nodes, 0 for unfixed nodes |
infer.method |
The inference method to solve the clamped CRF |
... |
The parameters for |
Details
Conditional inference (takes another inference method as input)
Value
This function will return a list with components:
node.bel |
Node belief. It is a matrix with |
edge.bel |
Edge belief. It is a list of matrices. The size of list is |
logZ |
The logarithmic value of CRF normalization factor Z. |
Examples
library(CRF)
data(Small)
i <- infer.conditional(Small$crf, c(0,1,0,0), infer.exact)
[Package CRF version 0.4-3 Index]