def setUp(self):
self.batch_size = 30
self.action_size = 3
self.q_values = np.random.normal(
size=(self.batch_size, self.action_size)).astype(np.float32)
self.qout = action_value.DiscreteActionValue(
chainer.Variable(self.q_values))
def __call__(self, x, test=False):
h = x
for l in self.conv_layers:
h = self.activation(l(h))
# Advantage
batch_size = x.shape[0]
ya = self.a_stream(h, test=test)
mean = F.reshape(F.sum(ya, axis=1) / self.n_actions, (batch_size, 1))
ya, mean = F.broadcast(ya, mean)
ya -= mean
# State value
ys = self.v_stream(h, test=test)
ya, ys = F.broadcast(ya, ys)
q = ya + ys
return action_value.DiscreteActionValue(q)
