HMM_classify: Classify Gaze Data

In maltelueken/gazeHMM: Classify Gaze Data With a Hidden Markov Model

Description

Classifies gaze data into eye movement events by using a hidden Markov model from the package depmixS4.

Usage

HMM_classify(
  data,
  nstates,
  respstart,
  trstart,
  instart,
  sf = c(10, 10),
  fit.control = em.control(maxit = 5000, random.start = F)
)

Arguments

data	Data frame with preprocessed gaze samples as returned by preprocess. The object must contain the variables vel, acc, and angle.
nstates	Number of states in the hidden Markov model.
respstart	Starting values for the response model in the hidden Markov model.
trstart	Starting values for the transition model in the hidden Markov model.
instart	Starting values for the initial state model in the hidden Markov model.
sf	Vector of length two indicating by which factor velocity and acceleration data will be divided.
fit.control	List of settings for the EM algorithm as returned by em.control.

Value

An object of class depmix.fitted.

maltelueken/gazeHMM documentation built on Oct. 6, 2020, 11:10 a.m.