GRU network
Next to LSTMs, another popular recurrent neural network variant is the Gated Recurrent Unit, or GRU. It's appeal is in its simplicity: GRU cells require less computation than LSTM cells while often matching them in performance.
The code you are provided with is the RNN model definition that you coded previously. Your task is to adapt it such that it produces a GRU network instead. torch
and torch.nn as nn
have already been imported for you.
This exercise is part of the course
Intermediate Deep Learning with PyTorch
Exercise instructions
- Update the RNN model definition in order to obtain a GRU network; assign the GRU layer to
self.gru
.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
class Net(nn.Module):
def __init__(self):
super().__init__()
# Define RNN layer
self.rnn = nn.RNN(
input_size=1,
hidden_size=32,
num_layers=2,
batch_first=True,
)
self.fc = nn.Linear(32, 1)
def forward(self, x):
h0 = torch.zeros(2, x.size(0), 32)
out, _ = self.rnn(x, h0)
out = self.fc(out[:, -1, :])
return out