BaseTrainer with `find_unused_parameters=True` when using DistributedDataParallel() (#14323)

ultralytics/engine/trainer.py

@@ -266,7 +266,7 @@ class BaseTrainer:
         self.amp = bool(self.amp)  # as boolean
         self.scaler = torch.cuda.amp.GradScaler(enabled=self.amp)
         if world_size > 1:
-            self.model = nn.parallel.DistributedDataParallel(self.model, device_ids=[RANK])
+            self.model = nn.parallel.DistributedDataParallel(self.model, device_ids=[RANK], find_unused_parameters=True)
 
         # Check imgsz
         gs = max(int(self.model.stride.max() if hasattr(self.model, "stride") else 32), 32)  # grid size (max stride)