def test_get_item_backward(self):
x = Variable(np.array([[1, 2, 3], [4, 5, 6]]))
y = F.get_item(x, 1)
y.backward()
expecteds = Variable(np.array([[0, 0, 0], [1, 1, 1]]))
for result, expected in zip(x.grad.data, expecteds.data):
assert all(result == expected)
def test_forward1(self):
x_data = np.arange(12).reshape((2, 2, 3))
x = Variable(x_data)
y = F.get_item(x, 0)
self.assertTrue(array_allclose(y.data, x_data[0]))
def test_backward2(self):
x_data = np.arange(12).reshape(4, 3)
slices = slice(1, 3)
f = lambda x: F.get_item(x, slices)
gradient_check(f, x_data)
def test_backward1(self):
x_data = np.array([[1, 2, 3], [4, 5, 6]])
slices = 1
f = lambda x: F.get_item(x, slices)
gradient_check(f, x_data)
def test_forward3(self):
x_data = np.arange(12).reshape((2, 2, 3))
x = Variable(x_data)
y = F.get_item(x, (Ellipsis, 2))
self.assertTrue(array_allclose(y.data, x_data[..., 2]))
def test_forward2(self):
x_data = np.arange(12).reshape((2, 2, 3))
x = Variable(x_data)
y = F.get_item(x, (0, 0, slice(0, 2, 1)))
self.assertTrue(array_allclose(y.data, x_data[0, 0, 0:2:1]))
def test_backward(self):
x = Variable(np.array([[1, 2, 3], [4, 5, 6]]))
ind = [1, 1, 0]
y = get_item(x, ind)
y.backward()
assert_equal(x.grad.data, [[1, 1, 1], [2, 2, 2]])
def test_forward(self):
x = Variable(np.array([[1, 2, 3], [4, 5, 6]]))
ind = [1, 1, 0]
y = get_item(x, ind)
assert_equal(y.data, [[4, 5, 6], [4, 5, 6], [1, 2, 3]])
def test_backward2(self):
x_data = np.arange(12).reshape(4, 3)
slices = slice(1, 3)
y_grad = np.ones(x_data[slices].shape)
f = lambda x: F.get_item(x, slices)
check_backward(f, x_data, y_grad=y_grad)
def test_backward1(self):
x_data = np.array([[1, 2, 3], [4, 5, 6]])
slices = 1
y_grad = np.ones(x_data[slices].shape)
f = lambda x: F.get_item(x, slices)
check_backward(f, x_data, y_grad=y_grad)
def test_get_item_forward(self):
x = Variable(np.array([[1, 2, 3], [4, 5, 6]]))
y = F.get_item(x, 1)
excepted = Variable(np.array([4, 5, 6]))
assert all(y.data == excepted.data)
