Add support for Torch ORT to Transformer based Tasks

[SeanNaren]

What does this PR do?

Adds Torch ORT support to Transformer based Flash Tasks!

Before submitting

• Was this discussed/approved via a Github issue? (no need for typos and docs improvements)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure your PR does only one thing, instead of bundling different changes together?
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests? [not needed for typos/docs]
• Did you verify new and existing tests pass locally with your changes?
• If you made a notable change (that affects users), did you update the CHANGELOG?

PR review

• Is this pull request ready for review? (if not, please submit in draft mode)

Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

Did you have fun?

Make sure you had fun coding 🙃

[codecov]

Codecov Report

Merging #667 (f1723e7) into master (4e89a37) will decrease coverage by 0.70%.
The diff coverage is 76.31%.

@@            Coverage Diff             @@
##           master     #667      +/-   ##
==========================================
- Coverage   90.01%   89.30%   -0.71%     
==========================================
  Files         185      186       +1     
  Lines        9664     9696      +32     
==========================================
- Hits         8699     8659      -40     
- Misses        965     1037      +72     

Flag	Coverage Δ
unittests	89.30% <76.31%> (-0.71%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
flash/text/seq2seq/question_answering/model.py	76.47% <ø> (ø)
flash/text/seq2seq/summarization/model.py	80.00% <ø> (ø)
flash/text/seq2seq/translation/model.py	76.19% <ø> (ø)
flash/text/ort_callback.py	53.33% <53.33%> (ø)
flash/text/seq2seq/core/model.py	76.62% <87.50%> (+1.26%)	⬆️
flash/text/classification/model.py	92.30% <92.85%> (-0.88%)	⬇️
flash/core/utilities/imports.py	89.18% <100.00%> (+0.09%)	⬆️
flash/graph/classification/model.py	37.70% <0.00%> (-62.30%)	⬇️
flash/graph/data.py	45.00% <0.00%> (-50.00%)	⬇️
flash/graph/classification/cli.py	42.85% <0.00%> (-50.00%)	⬇️
... and 2 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 4e89a37...f1723e7. Read the comment docs.

[ethanwharris]

Awesome, looks really neat 😃 Do we need to add torch ORT to the text requirements if it will be enabled by default? Also, I guess the docs section needs to be added to the summarization task too?

[SeanNaren]

Awesome, looks really neat 😃 Do we need to add torch ORT to the text requirements if it will be enabled by default? Also, I guess the docs section needs to be added to the summarization task too?

I just realised that the only way to test is with an NVIDIA GPU or AMD GPU. Do we test Flash at all on GPU? cc @ethanwharris

[ethanwharris]

Awesome, LGTM 😃