Fix torch.distributed.error on single gpu

[xyao-nv]

Fixes #

Changes proposed in this pull request:

• In single GPU mode, no dist training is initialized
• When dist training is not initialized, torch.distributed.barrier() shall be skipped

Before submitting

• I've read and followed all steps in the Making a pull request
  section of the CONTRIBUTING docs.
• I've updated or added any relevant docstrings.
• If this PR fixes a bug, I've added a test that will fail without my fix.
• If this PR adds a new feature, I've added tests that sufficiently cover my new functionality.

[youliangtan]

LGTM. Thanks for the quick fix.