To Do List

To Do in jazzparser.taggers.ngram.tagger.NgramTaggerModel.generate_chord_sequence this isn't implemented yet for n-grams. It's not a high priority, but would be fun.

To Do in jazzparser.taggers.segmidi.chordclass.hmm.ChordClassHmm Make this inherit from jazzparser.utils.nltk.ngram.DictionaryHmmModel so that we can use a specialization of the jazzparser.utils.nltk.ngram.baumwelch.BaumWelchTrainer with it.

To Do in jazzparser.taggers.segmidi.chordclass.train.ChordClassBaumWelchTrainer 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.

To Do in jazzparser.utils.scripts.prepare_evaluation_options This is not used any more. Remove it, after checking it's definitely not used.

To Do in jptests.formalisms.music_halfspan.backoff.TestLfToStates.test_back_conversion update this test

To Do in jptests.formalisms.music_halfspan.rules.TestComposition Write these tests. I'm bored of writing dull tests right now.

To Do in midi.SysExEvent we should possibly terminate the sysex data on finding any 1-initial byte. Sysex messages are supposed to end with a 0xF7, but the data can only contain 0-initial bytes (7-bit values).

To Do in midi.check_midi add more possible problems to the checks.

To Do in midi.sequencer_portmidi This isn't working yet. It's a nice idea, but it's proving very difficult to get the timing to work out right. It seems that writing to the midi output stream takes a long time and holds up the playback, making a mess of the timings of events.