crf_options {crfsuite} | R Documentation |
Conditional Random Fields parameters
crf_options(method = c("lbfgs", "l2sgd", "averaged-perceptron", "passive-aggressive", "arow"))
method |
character string with the type of training method. Either one of:
|
a list with elements
method: The training method
type: The type of graphical model which is always set crf1d: Linear-chain (first-order Markov) CRF
params: A data.frame with fields arg, arg_default and description indicating the possible hyperparameters of the algorithm, the default values and the description
default: A list of default values which can be used to pass on to the options
argument of crf
# L-BFGS with L1/L2 regularization opts <- crf_options("lbfgs") str(opts) # SGD with L2-regularization crf_options("l2sgd") # Averaged Perceptron crf_options("averaged-perceptron") # Passive Aggressive crf_options("passive-aggressive") # Adaptive Regularization of Weights (AROW) crf_options("arow")