Suppose I create a char level RNN. For the sake of simplicity, I’ll convert every letter into a onehot vector of length 26. Then, I created the RNN class as :
class RNN(nn.Module):
def __init__(self, input_size, hidden_size, output_size):
super(RNN, self).__init__()
self.hidden_size = hidden_size
self.i2h = nn.Linear(input_size + hidden_size, hidden_size)
self.i2o = nn.Linear(input_size + hidden_size, output_size)
self.softmax = nn.LogSoftmax(dim=1)
I’m skipping the other functions for now. My question is what does the first argument in the constructor function, input_size
specify? Since every character is a vector of length 26, should the input_size be set to 26?