Pytorch RuntimeError: expected scalar type Long but found Int

I try launch test, but get this error:

*Traceback (most recent call last):
  File "<ipython-input-59-5947c660ace7>", line 26, in test_ClassNLLCriterionUnstable
    torch_layer_output_var = torch_layer(torch.log(layer_input_var),
  File "C:\Program Files\Python38\lib\site-packages\torch\nn\modules\module.py", line 550, in __call__
    result = self.forward(*input, **kwargs)
  File "C:\Program Files\Python38\lib\site-packages\torch\nn\modules\loss.py", line 205, in forward
    return F.nll_loss(input, target, weight=self.weight, ignore_index=self.ignore_index, reduction=self.reduction)
  File "C:\Program Files\Python38\lib\site-packages\torch\nn\functional.py", line 2115, in nll_loss
    ret = torch._C._nn.nll_loss(input, target, weight, _Reduction.get_enum(reduction), ignore_index)
RuntimeError: expected scalar type Long but found Int*

Test code:

class TestLayers(unittest.TestCase):
     def test_ClassNLLCriterionUnstable(self):
        np.random.seed(42)
        torch.manual_seed(42)

        batch_size, n_in = 2, 4
        for _ in range(100):
            # layers initialization
            torch_layer = torch.nn.NLLLoss()
            custom_layer = ClassNLLCriterionUnstable()

            layer_input = np.random.uniform(0, 1, (batch_size, n_in)).astype(np.float32)
            layer_input /= layer_input.sum(axis=-1, keepdims=True)
            layer_input = layer_input.clip(custom_layer.EPS, 1. - custom_layer.EPS)  # unifies input
            target_labels = np.random.choice(n_in, batch_size)
            target = np.zeros((batch_size, n_in), np.float32)
            target[np.arange(batch_size), target_labels] = 1  # one-hot encoding

            # 1. check layer output
            custom_layer_output = custom_layer.updateOutput(layer_input, target)
            layer_input_var = Variable(torch.from_numpy(layer_input), requires_grad=True)
            torch_layer_output_var = torch_layer(torch.log(layer_input_var), 
                                                 Variable(torch.from_numpy(target_labels), requires_grad=False))
            self.assertTrue(np.allclose(torch_layer_output_var.data.numpy(), custom_layer_output, atol=1e-6))

ClassNLLCriterionUnstable code:

class ClassNLLCriterionUnstable(Criterion):
    EPS = 1e-15
    def __init__(self):
        a = super(ClassNLLCriterionUnstable, self)
        super(ClassNLLCriterionUnstable, self).__init__()

    def updateOutput(self, input, target):  
        # Use this trick to avoid numerical errors
        input_clamp = np.clip(input, self.EPS, 1 - self.EPS)
         # Your code goes here. ################################################
        return self.output

    def updateGradInput(self, input, target):

        # Use this trick to avoid numerical errors
        input_clamp = np.clip(input, self.EPS, 1 - self.EPS)

        # Your code goes here. ################################################
        return self.gradInput

    def __repr__(self):
        return "ClassNLLCriterionUnstable"

I don’t know how to fix it. I am new to machine learning, I wanted to implement the negative loglikehood criterion and compare it with the pytorch.