RNN¶
- class
dragon.nn.
RNN
(
input_size,
hidden_size,
nonlinearity='relu',
num_layers=1,
bidirectional=False,
dropout=0,
**kwargs
)[source]¶ Apply a multi-layer Elman RNN. [Elman, 1990].
The data format of inputs should be \((T, N, C)\):
t, n, c = 8, 2, 4 m = dragon.nn.RNN(8, 16) x = dragon.constant([t, n, c], 'float32') y = m(x)
__init__¶
RNN.
__init__
(
input_size,
hidden_size,
nonlinearity='relu',
num_layers=1,
bidirectional=False,
dropout=0,
**kwargs
)[source]¶Create a
RNN
module.- Parameters:
- input_size (int) – The dimension of input.
- hidden_size (int) – The dimension of hidden state.
- nonlinearity ({'tanh', 'relu'}, optional) – The nonlinearity.
- num_layers (int, optional, default=1) – The number of recurrent layers.
- bidirectional (bool, optional, default=False) – Whether to create a bidirectional rnn.
- dropout (number, optional, default=0) – The dropout ratio.