votes up 1

Reduction fn (reduction_fn) must specify dst!

Package:
torch
github stars 50580
Exception Class:
ValueError

Raise code

            process_group_sync = res50_model_sync.layer1[0].bn1.process_group
            self.assertEqual(process_group_sync, process_group)

        def _run_reduction_test(
                self, tensor, expected_tensor, op, reduction_fn=dist.all_reduce, dst=None
        ):
            if reduction_fn != dist.all_reduce and dst is None:
                raise ValueError(f"Reduction fn {reduction_fn} must specify dst!")
            if dst is not None:
                reduction_fn(tensor, dst, op)
                # Only destination rank tensor is expected to have final result.
                if dist.get_rank() == dst:
                    self.assertEqual(tensor, expected_tensor)
            else:
                reduction_fn(tensor, op)
😲  Walkingbet is Android app that pays you real bitcoins for a walking. Withdrawable real money bonus is available now, hurry up! 🚶

NO FIXES YET

Just press the button and we will add solution
to this exception as soon as possible

* As many users press the button, the faster we create a fix

Add a possible fix

Please authorize to post fix