You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
pytorch version: 0.4.1
when evaluate the model, eval batch size is 1, if max_update is None: max_update = op_ids.eq(self.update_id).sum(-1).max().item()
the value of max_update might be zero. So run the following code: v = torch.zeros(1, max_update, self.hidden_size, device=input_ids.device) , An error message may be given: RuntimeError: sizes must be non-negative
The text was updated successfully, but these errors were encountered:
pytorch version: 0.4.1
when evaluate the model, eval batch size is 1,
if max_update is None: max_update = op_ids.eq(self.update_id).sum(-1).max().item()
the value of
max_update
might be zero. So run the following code:v = torch.zeros(1, max_update, self.hidden_size, device=input_ids.device)
, An error message may be given:RuntimeError: sizes must be non-negative
The text was updated successfully, but these errors were encountered: