def check_forward(self, diag_data, non_diag_data):
diag = chainer.Variable(diag_data)
non_diag = chainer.Variable(non_diag_data)
y = lower_triangular_matrix(diag, non_diag)
correct_y = numpy.zeros((self.batch_size, self.n, self.n),
dtype=numpy.float32)
tril_rows, tril_cols = numpy.tril_indices(self.n, -1)
correct_y[:, tril_rows, tril_cols] = cuda.to_cpu(non_diag_data)
diag_rows, diag_cols = numpy.diag_indices(self.n)
correct_y[:, diag_rows, diag_cols] = cuda.to_cpu(diag_data)
gradient_check.assert_allclose(correct_y, cuda.to_cpu(y.data))
def __call__(self, state, test=False):
h = self.hidden_layers(state, test=test)
v = self.v(h)
mu = self.mu(h)
if self.scale_mu:
mu = scale_by_tanh(mu, high=self.action_space.high,
low=self.action_space.low)
mat_diag = F.exp(self.mat_diag(h))
if hasattr(self, 'mat_non_diag'):
mat_non_diag = self.mat_non_diag(h)
tril = lower_triangular_matrix(mat_diag, mat_non_diag)
mat = F.batch_matmul(tril, tril, transb=True)
else:
mat = F.expand_dims(mat_diag ** 2, axis=2)
return QuadraticActionValue(
mu, mat, v, min_action=self.action_space.low,
max_action=self.action_space.high)
def __call__(self, s):
if self._use_batch_norm:
h = self._L_bn0(s)
else:
h = s
h = self._linear_L1(h)
if self._use_batch_norm:
h = self._L_bn1(h)
else:
pass
h = F.relu(h)
h = self._linear_L2(h)
if self._use_batch_norm:
h = self._L_bn2(h)
else:
pass
h = F.relu(h)
diag = F.exp(self._linear_L3_diag(h))
rest = self._linear_L3_rest(h)
return lower_triangular_matrix(diag, rest)
def _L_matrix(self, x):
diag = F.exp(self._linear_L_diag(x))
rest = self._linear_L_rest(x)
return lower_triangular_matrix(diag, rest)
