HwyBLSTM Class

Bidirection LSTM with Highway Layers - LSTM++

Based on model proposed by Yee-King et al. 1

class spiegelib.estimator.HwyBLSTM(input_shape, num_outputs, lstm_size=128, highway_layers=6, **kwargs)

Bases: spiegelib.estimator.tf_estimator_base.TFEstimatorBase

Parameters

  • input_shape (tuple) – Shape of matrix that will be passed to model input

  • num_outputs (int) – Number of outputs the model has

  • kwargs – optional keyword arguments to pass to spiegelib.estimator.TFEstimatorBase

build_model()

Construct LSTM++ Model

1

Yee-King, Matthew John, Leon Fedden, and Mark d’Inverno. “Automatic programming of VST sound synthesizers using deep networks and other techniques.” IEEE Transactions on Emerging Topics in Computational Intelligence 2.2 (2018): 150-159.