def bottle_neck(self, hidden_state, lstm_cells, x_data, train):
x = chainer.Variable(x_data, volatile=not train)
a1 = self.l1_a(F.concat((x, hidden_state["h1"])))
x1 = self.l1_x(F.concat((x, hidden_state["h1"], lstm_cells["c1"])))
c1, h1 = peephole_lstm(lstm_cells["c1"], a1, x1)
h1 = gradient_clip(h1, 10.0)
a2 = self.l2_a(F.concat((x, hidden_state["h2"], h1)))
x2 = self.l2_x(F.concat((x, hidden_state["h2"], h1, lstm_cells["c2"])))
c2, h2 = peephole_lstm(lstm_cells["c2"], a2, x2)
h2 = gradient_clip(h2, 10.0)
a3 = self.l2_a(F.concat((x, hidden_state["h3"], h2)))
x3 = self.l2_x(F.concat((x, hidden_state["h3"], h2, lstm_cells["c3"])))
c3, h3 = peephole_lstm(lstm_cells["c3"], a3, x3)
h3 = gradient_clip(h3, 10.0)
y = self.l4(F.concat((h1, h2, h3)))
y = gradient_clip(y, 100.0)
n = int((y.data.shape[1] - 1) / 6)
gi, b = gauss_bernoulli_params(n, y)
hidden_state = {"h1": h1, "h2": h2, "h3": h3}
if train:
lstm_cells = {"c1": c1, "c2": c2, "c3": c3}
return gi, b, hidden_state, lstm_cells
def check_backward(self, f):
x = chainer.Variable(f(self.x))
y = gradient_clip(x, self.a, self.b)
y.creator.forward((x.data,))
y.grad = f(self.y_grad)
y.backward()
assert_allclose(y.data, x.data)
assert_allclose(x.grad, f(self.x_grad_ab))
y = gradient_clip(x, self.a)
y.creator.forward((x.data,))
y.grad = f(self.y_grad)
y.backward()
assert_allclose(y.data, x.data)
assert_allclose(x.grad, f(self.x_grad_a))
def bottle_neck(self, hidden_state, lstm_cells, x_data, train):
x = chainer.Variable(x_data, volatile=not train)
h1_in = self.l1_first(x) + self.l1_recur(hidden_state['h1'])
c1, h1 = F.lstm(lstm_cells['c1'], h1_in)
h1 = gradient_clip(h1, 10.0)
h2_in = self.l2_first(x) + self.l2_recur(hidden_state['h2']) + self.l2_input(h1)
c2, h2 = F.lstm(lstm_cells['c2'], h2_in)
h2 = gradient_clip(h2, 10.0)
h3_in = self.l3_first(x) + self.l3_recur(hidden_state['h3']) + self.l3_input(h2)
c3, h3 = F.lstm(lstm_cells['c3'], h3_in)
h3 = gradient_clip(h3, 10.0)
y = self.l4(F.concat((h1, h2, h3)))
y = gradient_clip(y, 100.0)
n = int((y.data.shape[1] - 1) / 6)
gi, b = gauss_bernoulli_params(n, y)
hidden_state = {'h1': h1, 'h2': h2, 'h3': h3}
if train:
lstm_cells = {'c1': c1, 'c2': c2, 'c3': c3}
return gi, b, hidden_state, lstm_cells
