MusicRNN
constructor.
Path to the checkpoint directory.
(Optional) MusicRNNSpec
object. If undefined, will be loaded
from a config.json
file in the checkpoint directory.
Continues a provided quantized NoteSequence.
The sequence to continue. Must be quantized.
How many steps to continue.
(Optional) The softmax temperature to use when sampling from the logits. Argmax is used if not provided. Temperature can be any number value above 0, however, anything above 1.5 will essentially result in random results.
(Optional) Chord progression to use as conditioning. A chord progression param is an array of chords that are passed to the tonal package for parsing, so they're in that format. Example: ["G", "Em", "C", "D"]
Continues a provided quantized NoteSequence, and returns the computed probability distribution at each step.
The sequence to continue. Must be quantized.
How many steps to continue.
(Optional) The softmax temperature to use when sampling from the logits. Argmax is used if not provided.
(Optional) Chord progression to use as conditioning.
Loads variables from the checkpoint and instantiates the Encoder
and
Decoder
.
Returns true iff model is initialized.
Generated using TypeDoc
Main MusicRNN model class.
A MusicRNN is an LSTM-based language model for musical notes.