RNNCell

neuralpy.layers.recurrent.RNNCell(input_size, hidden_size, bias=True, non_linearity='tanh', name=None)

danger

RNNCell Layer is unstable and buggy, not ready for any real use

Applies an Elman RNN cell with tanh or ReLU non-linearity.

To learn more about RNNCell, please check pytorch documentation

Supported Arguments:

• input_size: (Integer) The number of expected features in the input
• hidden_size: (Integer) The number of features in the hidden state
• bias=True: (Boolean) If true then uses the bias,Defaults to true
• non_linearity=tanh: (String) The non-linearity to use.Default tanh
• name=None: (String) Name of the layer, if not provided then automatically calculates a unique name for the layer

Example Code

import torch

from neuralpy.layers.recurrent import RNNCell

rnn = RNNCell(10, 20)

input = torch.randn(6, 3, 10)

hx = torch.randn(3, 20)

output = []

for i in range(6):

hx = rnn(input[i], hx)

output.append(hx)