Added distil_whisper.

[MaanavD]

Should work! Added distil whisper to torchbench. Local time is within normal CI requirements.

[xuzhao9]

Could you please address the following test error?

======================================================================
ERROR: test_hf_distil_whisper_train_cpu (__main__.TestBenchmark.test_hf_distil_whisper_train_cpu)
----------------------------------------------------------------------
components._impl.workers.subprocess_rpc.ChildTraceException: Traceback (most recent call last):
  File "/runner/_work/benchmark/benchmark/components/_impl/workers/subprocess_rpc.py", line 482, in _run_block
    exec(  # noqa: P204
  File "<subprocess-worker>", line 2, in <module>
  File "/runner/_work/benchmark/benchmark/torchbenchmark/util/model.py", line 309, in invoke
    return self._invoke_staged_train_test(num_batch=self.num_batch)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/runner/_work/benchmark/benchmark/torchbenchmark/util/model.py", line 299, in _invoke_staged_train_test
    losses = self.forward()
             ^^^^^^^^^^^^^^
  File "/runner/_work/benchmark/benchmark/torchbenchmark/util/framework/huggingface/model_factory.py", line 153, in forward
    outputs = self.model(**self.example_inputs)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/runner/miniconda3/envs/pr-ci-a100/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1511, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/runner/miniconda3/envs/pr-ci-a100/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1520, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: WhisperForAudioClassification.forward() got an unexpected keyword argument 'input_ids'

    working_dir: /tmp/tmpmzqebwgh
    stdout:
        [2023-12-05] 03:15:45.738260: TIMER_SUBPROCESS_BEGIN_EXEC
        [2023-12-05] 03:15:45.740672: TIMER_SUBPROCESS_FAILED
        [2023-12-05] 03:15:45.740737: TIMER_SUBPROCESS_FINISHED
        [2023-12-05] 03:15:45.740759: TIMER_SUBPROCESS_BEGIN_READ

    stderr:


The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/runner/_work/benchmark/benchmark/test.py", line 78, in train_fn
    task.invoke()
  File "/runner/_work/benchmark/benchmark/torchbenchmark/__init__.py", line 399, in invoke
    self.worker.run("""
  File "/runner/_work/benchmark/benchmark/components/_impl/workers/subprocess_worker.py", line 158, in run
    self._run(snippet)
  File "/runner/_work/benchmark/benchmark/components/_impl/workers/subprocess_worker.py", line 323, in _run
    subprocess_rpc.SerializedException.raise_from(
  File "/runner/_work/benchmark/benchmark/components/_impl/workers/subprocess_rpc.py", line 458, in raise_from
    raise e from ChildTraceException(traceback_str)
TypeError: WhisperForAudioClassification.forward() got an unexpected keyword argument 'input_ids'

[xuzhao9]

The CI error are pre-existing errors from the trunk, should be safe to land this.

[facebook-github-bot]

@xuzhao9 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

[facebook-github-bot]

@xuzhao9 merged this pull request in d61b74a.