Fix sagemaker DP/MP (#23681)

* Check for use_sagemaker_dp

* Add a check for is_sagemaker_mp when setting _n_gpu again. Should be last broken thing

* Try explicit check?

* Quality

src/transformers/trainer.py

@@ -3398,7 +3398,9 @@ class Trainer:
             tensors = nested_xla_mesh_reduce(tensors, name)
         elif is_sagemaker_mp_enabled():
             tensors = smp_gather(tensors)
-        elif self.args.parallel_mode == ParallelMode.DISTRIBUTED:
+        elif (self.args.distributed_state is not None and self.args.distributed_state.distributed_type != "NO") or (
+            self.args.distributed_state is None and self.local_rank != -1
+        ):
             tensors = distributed_concat(tensors)
         return tensors
 

src/transformers/training_args.py

@@ -1629,6 +1629,9 @@ class TrainingArguments:
             device = torch.device("cuda", local_rank)
             self._n_gpu = 1
             torch.cuda.set_device(device)
+        elif is_sagemaker_dp_enabled():
+            self.distributed_state = PartialState(_use_sagemaker_dp=True)
+            self._n_gpu = 1
         elif self.deepspeed:
             # Need to do similar for Accelerator init
             os.environ["ACCELERATE_USE_DEEPSPEED"] = "true"
@@ -1653,8 +1656,9 @@ class TrainingArguments:
         if is_torch_tpu_available():
             device = self.distributed_state.device
             self._n_gpu = 0
-        elif is_sagemaker_dp_enabled():
-            self._n_gpu = 1
+        elif is_sagemaker_dp_enabled() or is_sagemaker_mp_enabled():
+            # Already set _n_gpu
+            pass
         elif self.distributed_state.distributed_type == DistributedType.NO:
             if self.use_mps_device:
                 if not torch.backends.mps.is_available():