Class ChordClassBaumWelchTrainer

source code

object --+
         |
        ChordClassBaumWelchTrainer

Class with methods to retrain a chordclass model using the Baum-Welch EM algorithm.

Module options must be processed already - we do that in the ChordClassTaggerModel, not here.

To Do: Inherit from the jazzparser.utils.nltk.ngram.baumwelch.BaumWelchTrainer. Currently, the generic trainer duplicates a lot of this code, since it was based on it.

Instance Methods

[hide private]

__init__(self, model, options={})
x.__init__(...) initializes x; see help(type(x)) for signature source code

train(self, emissions, logger=None, save_callback=None)
Performs unsupervised training using Baum-Welch EM.

source code

update_model(self, model)
Replaces the distributions of the saved model with those of the given model and saves it.

source code

Inherited from object: __delattr__, __format__, __getattribute__, __hash__, __new__, __reduce__, __reduce_ex__, __repr__, __setattr__, __sizeof__, __str__, __subclasshook__

Class Variables

[hide private]

OPTIONS = [ModuleOption('max_iterations', filter= int, help_te...

Properties

[hide private]

Inherited from object: __class__

Method Details

[hide private]

init(self, model, options=`{}`)
(Constructor)

source code

x.__init__(...) initializes x; see help(type(x)) for signature

Overrides: object.__init__: (inherited documentation)

train(self, emissions, logger=None, save_callback=None)

source code

Performs unsupervised training using Baum-Welch EM.

This is performed on a model that has already been initialized. You might, for example, create such a model using jazzparser.taggers.segmidi.chordclass.hmm.ChordClassHmm.initialize_chord_classes.

This is based on the training procedure in NLTK for HMMs: nltk.tag.hmm.HiddenMarkovModelTrainer.train_unsupervised.

Parameters:

emissions (jazzparser.data.input.MidiTaggerTrainingBulkInput or list of jazzparser.data.input.Inputs) - training MIDI data
logger (logging.Logger) - a logger to send progress logging to

Class Variable Details

[hide private]

OPTIONS

Value:

[ModuleOption('max_iterations', filter= int, help_text= "Number of tra
ining iterations to give up after " "if we don't reach convergence bef
ore.", usage= "max_iterations=N, where N is an integer", default= 100)
, ModuleOption('convergence_logprob', filter= float, help_text= "Diffe
rence in overall log probability of the " "training data made by one i
teration after which we " "consider the training to have converged.", 
usage= "convergence_logprob=X, where X is a small floating " "point nu
mber (e.g. 1e-3)", default= 1e-3), ModuleOption('split', filter= int, 
...

Class ChordClassBaumWelchTrainer

__init__(self, model, options={}) (Constructor)

train(self, emissions, logger=None, save_callback=None)

OPTIONS

init(self, model, options=`{}`)
(Constructor)