Model dtype cant set


self.lin = nn.Linear(in_dim, hidden_dim).double()

and when try to use

out = self.lin(out)

Receive error:

RuntimeError: Expected object of scalar type Float but got scalar type Double for argument #2 ‘mat1’ in call to _th_addmm

out.float() fix it but i dont understand why self.lin on float type when i directly set double on it!?

Do you cast the overall model to a different type after setting self.lin to double by any chance?