[LLM Tests] Config support for LLM test suite / enable test suite in GHA #1324
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
dbogunowicz
changed the title
dbogunowicz
commented
Oct 17, 2023
dbogunowicz
commented
Oct 17, 2023
dbogunowicz
commented
Oct 17, 2023
tests/deepsparse/transformers/pipelines/test_text_generation.py
Outdated
Show resolved
Hide resolved
|
Tests are green yay! :) |
dbogunowicz
commented
Oct 17, 2023
tests/deepsparse/transformers/pipelines/test_text_generation.py
Outdated
Show resolved
Hide resolved
horheynm
reviewed
Oct 18, 2023
horheynm
reviewed
Oct 18, 2023
horheynm
reviewed
Oct 18, 2023
horheynm
reviewed
Oct 18, 2023
bfineran
reviewed
Oct 18, 2023
bfineran
suggested changes
Oct 19, 2023
tests/deepsparse/transformers/pipelines/test_text_generation.py
Outdated
Show resolved
Hide resolved
tests/deepsparse/transformers/pipelines/test_text_generation.py
Outdated
Show resolved
Hide resolved
tests/deepsparse/transformers/pipelines/test_text_generation.py
Outdated
Show resolved
Hide resolved
tests/deepsparse/transformers/pipelines/test_text_generation.py
Outdated
Show resolved
Hide resolved
tests/deepsparse/transformers/pipelines/test_text_generation.py
Outdated
Show resolved
Hide resolved
…euralmagic/deepsparse into feature/damian/final_llm_tests
bfineran
approved these changes
Oct 23, 2023
dbogunowicz
added a commit
that referenced
this pull request
Oct 25, 2023
Reverting my change from #1324 Back then, I was not aware that this property of the sparsezoo `Model` was an intended change (assumed it was a typo)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Configurable testing suite for LLMs.
Introduces two types of tests:
Unit tests - test the smaller functionalities of the LLM pipelines (
text_generationandchat). Those are small, self-contained lightweight tests that use the tinystories model. Those will run on every commit in GHA. It should be easy and simple to add new test cases here by any developer.Integration test for LLMs - test end-to-end all the relevant LLM pipelines. Those are integration tests that are configurable and by default will not run on every commit but with some lower frequency. So far we have tests for codegen and opt models. I have run a limited scope of these tests on my local machine, so I do not guarantee that they will instantaneously all pass on the first go once deployed. Codegen has been thoroughly tested by me, but opt, because of its size, not.
Along with this feature come small fixes to the LLM pipelines that were required to make tests work properly.