next up previous contents
Next: SdpLearner Up: MedmixLearner Previous: get   Contents


Train a MED Gaussian mixture classifier given a MedmixLearner object, training examples and labels, and a list of options. The returned trained classifier can be used to predict labels for test examples.

learned = train(medmixObj, gramObj, labelsObj, optList)
medmixObj - MedmixLearner object
gramObj - training Gram object. Can include test data in last rows/cols for transductive applications.
labelsObj - training Labels object. Must contain corresponding train labels only.
optList - list of NAME,VALUE option pairs

'Seed' - value for random number generator seed. Use [] for clock. (default=unchanged)
'Missing' - struct of missing data entries (default=[])

Darrin Lewis 2006-05-08