runtime fit with std output fitted-pipeline throws error

Command

python -m d3m.runtime fit -p tests/data/pipelines/random-forest-classifier.yml -r tests/data/problems/iris_problem_1/problemDoc.json -i tests/data/datasets/iris_dataset_1/datasetDoc.json -s -

Error

Traceback (most recent call last):
  File "/home/lin/anaconda3/envs/d3m/lib/python3.6/runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "/home/lin/anaconda3/envs/d3m/lib/python3.6/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/home/lin/PycharmProjects/d3m/d3m/runtime.py", line 2780, in <module>
    main(sys.argv)
  File "/home/lin/PycharmProjects/d3m/d3m/runtime.py", line 2776, in main
    handler(arguments, parser)
  File "/home/lin/PycharmProjects/d3m/d3m/runtime.py", line 2312, in handler
    arguments.runtime_handler(arguments, pipeline_resolver=pipeline_resolver, meta_parser=meta_parser, dataset_resolver=dataset_resolver, problem_resolver=problem_resolver)
  File "/home/lin/PycharmProjects/d3m/d3m/runtime.py", line 1706, in _fit
    pickle.dump(fitted_pipeline, arguments.save)
TypeError: write() argument must be str, not bytes

More Detail

When output to a file, no error was raised