def test_random_normal_tf_value(self):
output_shape = tf.shape([2, 2])
self.assertTrue(
tf.reduce_all(
tf.equal(tf.shape(random_normal_like(self.test_tf)),
output_shape)),
'Output tensor shape should be same as input')
def test_random_normal_torch_value(self):
output_shape = (2, 2)
self.assertTrue((random_normal_like(self.test_torch).size() == output_shape),
'Output tensor shape should be same as input')
def test_random_normal_torch_type(self):
self.assertIsInstance(random_normal_like(self.test_torch), torch.Tensor, 'Output must be torch.Tensor')
def test_random_normal_tf_type(self):
self.assertIsInstance(random_normal_like(self.test_tf), tf.Tensor, 'Output type must be tf.Tensor')
def test_random_normal_np_value(self):
self.assertTrue((random_normal_like(self.test_np).shape == (2, 2)),
'Output array shape should be same as input')
def test_random_normal_np_type(self):
self.assertIsInstance(random_normal_like(self.test_np), np.ndarray, 'Output must be NumPy Array')
