There is no warning or error when convert 'int' to tensor

1

There is no warning or error when I tried to cast an ‘int’ to tensor, though it is not correct, there is no warning there and may cause “trouble”.

torch.Tensor(8)
Out[9]: 
tensor([8.9326e+36, 4.5598e-41, 6.1690e-30, 3.0837e-41, 3.3631e-44, 0.0000e+00,
        7.3986e+31, 4.5559e-41])
torch.Tensor(8.0)
Traceback (most recent call last):
  File "/home/lii/Programs/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py", line 2961, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-10-bdc841c73802>", line 1, in <module>
    torch.Tensor(8.0)
TypeError: new(): data must be a sequence (got float)
2

The constructors have long been deprecated. Don’t use them anymore!
torch.tensor will do what you want. And if you want the equivalent of torch.Tensor(8), you could use the far better named torch.empty(8) instead.

Best regards

Thomas