Fix check for falsey inputs in run_summarization

[JohnGiorgi]

What does this PR do?

In the PyTorch version of run_summarization.py, there is a check that excludes examples where the source document and target summary are both None.

transformers/examples/pytorch/summarization/run_summarization.py

Lines 516 to 520 in ccc0897

	inputs, targets = [], []
	for i in range(len(examples[text_column])):
	if examples[text_column][i] is not None and examples[summary_column][i] is not None:
	inputs.append(examples[text_column][i])
	targets.append(examples[summary_column][i])

I think this should be relaxed to check for falsey inputs instead. I think this because some datasets, like MultiNews, contain examples with empty strings:

from datasets import load_dataset

multi_news = load_dataset("multi_news", split="validation")
assert not multi_news[4850]["document"]

and these are not caught by the is not None check.

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

@sgugger, @patil-suraj

[HuggingFaceDocBuilderDev]

The documentation is not available anymore as the PR was closed or merged.

[sgugger]

Thanks for your PR! You made me bug a little bit because the test is not relaxed but stronger ;-) But I see what you mean, and I think it's the right way to go!