def test_weighted_sequence_mean_semantics(self):
inputs = np.array([[1, 1, 1], [1, 1, 0]], dtype=np.float32)
weights1 = np.array([1, 1, 1], dtype=np.float32)
layer = metrics._WeightedSequenceMean()
full_signature = (signature(inputs), signature(weights1))
layer.init(full_signature)
mean1 = layer((inputs, weights1))
np.testing.assert_allclose(mean1, 0.5)
weights2 = np.array([1, 1, 0], dtype=np.float32)
mean2 = layer((inputs, weights2))
np.testing.assert_allclose(mean2, 1.0)
def test_weighted_sequence_mean_semantics(self):
layer = metrics._WeightedSequenceMean()
sample_input = np.ones((2, 3))
sample_weights = np.ones((3, ))
full_signature = shapes.signature([sample_input, sample_weights])
layer.init(full_signature)
x = np.array([[1., 1., 1.], [1., 1., 0.]])
weights = np.array([1., 1., 1.])
mean = layer((x, weights))
np.testing.assert_allclose(mean, 0.5)
weights = np.array([1., 1., 0.])
mean = layer((x, weights))
np.testing.assert_allclose(mean, 1.)