Warning that make an error the GUI

on the step 3 there a little warning when i click on the button train modele

UserWarning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance.

the GUI catch this warning as an error on the output section but train well.
During 20min i was thinking there an error i was lookin at on the deps, cuda version, ect...

DEBUG:matplotlib:CONFIGDIR=C:\Users\londa\.matplotlib DEBUG:matplotlib:interactive is False DEBUG:matplotlib:platform is win32 INFO:torch.nn.parallel.distributed:Reducer buckets have been rebuilt in this iteration. INFO:gazo-tst-3:Train Epoch: 1 [73%] INFO:gazo-tst-3:[200, 0.0001] INFO:gazo-tst-3:loss_disc=3.803, loss_gen=3.319, loss_fm=10.956,loss_mel=20.974, loss_kl=2.616 INFO:gazo-tst-3:====> Epoch: 1 [2025-08-09 19:19:19] | (0:02:23.575950) INFO:gazo-tst-3:Train Epoch: 2 [45%] INFO:gazo-tst-3:[400, 9.99875e-05] INFO:gazo-tst-3:loss_disc=3.507, loss_gen=3.726, loss_fm=14.578,loss_mel=20.790, loss_kl=2.431 INFO:gazo-tst-3:====> Epoch: 2 [2025-08-09 19:21:15] | (0:01:56.007228) INFO:gazo-tst-3:Train Epoch: 3 [18%]

Reproduce the issue from the GUI's step 3 by clicking the train model button and observing the output section while training continues. Verify that the PyTorch UserWarning is not presented as an error, while genuine training errors remain visible; the reported training log shows that training continues successfully.