def test_LinearSoftmax(self):
model = Seq()
model.add(Linear(2, 1))
model.add(Softmax())
data = np.array([2., 3.])
out = model.forward(data)
self.assertEqual(out, 1.)
def test_LinearSigmoid(self):
model = Seq()
model.add(Linear(2, 1, initialize='ones'))
model.add(Sigmoid())
data = np.array([2., 3.])
out = model.forward(data)
self.assertEqual(round(out, 2), 1.)
def test_LinearLayerNumericalGradientCheck(self):
x = np.random.rand(3)
model = Seq()
model.add(Linear(3, 2, initialize='ones'))
num_grad = numerical_gradient.calc(model.forward, x)
deriv_grad = model.backward(np.array([1, 1]))
num_grad = np.sum(num_grad, axis=0)
numerical_gradient.assert_are_similar(deriv_grad, num_grad)
def test_Linear(self):
model = Seq()
model.add(Linear(2, 1, initialize='ones'))
data = np.array([2., 2.])
y = model.forward(data)
self.assertEqual(y, np.array([5]))