Hi there,
I am having issue understand the following implementation of ConvLSTM. I don’t really understand what input_size + hidden_size is? Also the 4 * hidden_size value for output? conv_lstm = ConvLSTMCell(256, self.mem_size) tells us that 256 and self.mem_size are input_size and hidden_size. Also, why are we multiplying hidden_size by 4 for output? Can some one more experience explain to me what is going on within the convolution? Thank you.
self.mem_size=256
self.conv_lstm = ConvLSTMCell(256, self.mem_size)
import torch
from torch import nn
from torch.autograd import Variable
class ConvLSTMCell(nn.Module):
def __init__(self, input_size, hidden_size, kernel_size=3, stride=1, padding=1):
super(ConvLSTMCell, self).__init__()
self.input_size = input_size
self.hidden_size = hidden_size
self.Gates = nn.Conv2d(input_size + hidden_size, 4 * hidden_size, kernel_size=kernel_size,
stride=stride, padding=padding)
for params in self.Gates.parameters():
params.requires_grad=True
torch.nn.init.xavier_normal_(self.Gates.weight)
torch.nn.init.constant_(self.Gates.bias, 0)
def forward(self, input_, prev_state):
batch_size = input_.data.size()[0]
spatial_size = input_.data.size()[2:]
if prev_state is None:
state_size = [batch_size, self.hidden_size] + list(spatial_size)
if(torch.cuda.is_available()):
prev_state = (Variable(torch.zeros(state_size).cuda(),requires_grad=True),Variable(torch.zeros(state_size).cuda(),requires_grad=True))
else:
prev_state = (Variable(torch.zeros(state_size),requires_grad=True),Variable(torch.zeros(state_size),requires_grad=True))
prev_hidden, prev_cell = prev_state
stacked_inputs = torch.cat((input_, prev_hidden), 1)
gates = self.Gates(stacked_inputs)
in_gate, remember_gate, out_gate, cell_gate = gates.chunk(4, 1)
in_gate = torch.sigmoid(in_gate)
remember_gate = torch.sigmoid(remember_gate)
out_gate = torch.sigmoid(out_gate)
cell_gate = torch.tanh(cell_gate)
cell = (remember_gate * prev_cell) + (in_gate * cell_gate)
hidden = out_gate * torch.tanh(cell)
return hidden, cell