Change local_rank to 0 to make test pass on 1 GPU machine.

tests/unit/test_common_pytorch_utils.py

@@ -46,10 +46,10 @@ def test_get_device_all_gpus():
 
 @pytest.mark.gpu
 def test_get_device_local_rank():
-    device, gpus = get_device(local_rank=1)
+    device, gpus = get_device(local_rank=0)
     assert isinstance(device, torch.device)
     assert device.type == "cuda"
-    assert device.index == 1
+    assert device.index == 0
     assert gpus == 1
 
 
@@ -121,4 +121,3 @@ def test_move_to_device_gpu(model):
         assert isinstance(model_cuda_same_gpu, DataParallel)
     else:
         assert isinstance(model_cuda_same_gpu, Sequential)
-