Fix Runtime Error: Expected tensor for argument #1 'indices' to have scalar type Long.

[cuicaihao]

The following cells will cause runtime issues:

input_ids = torch.tensor(padded)
attention_mask = torch.tensor(attention_mask)
with torch.no_grad():
    last_hidden_states = model(input_ids, attention_mask=attention_mask)` 

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
<ipython-input-24-17a214174eeb> in <module>
      4 attention_mask = torch.tensor(attention_mask)
      5 with torch.no_grad():
----> 6     last_hidden_states = model(input_ids, attention_mask=attention_mask)
.... ... 
RuntimeError: Expected tensor for argument #1 'indices' to have scalar type Long; but got torch.IntTensor instead (while checking arguments for embedding)

It can be solved by changing the input_ids and attention_mask type to torch.int64

input_ids = torch.tensor(padded).to(torch.int64)
attention_mask = torch.tensor(attention_mask).to(torch.int64)