Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[sharktank] Evaluation - Add Perplexity test #233

Merged
merged 38 commits into from
Oct 16, 2024
Merged

Conversation

archana-ramalingam
Copy link
Collaborator

Add Perplexity test for LLM evaluation

Copy link
Contributor

@rsuderman rsuderman left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just some minor optional changes.

sharktank/sharktank/utils/load_llm.py Outdated Show resolved Hide resolved
sharktank/sharktank/utils/tokenizer.py Show resolved Hide resolved
sharktank/tests/evaluate/perplexity.py Outdated Show resolved Hide resolved
@archana-ramalingam archana-ramalingam merged commit e30d0af into main Oct 16, 2024
6 of 9 checks passed
@archana-ramalingam archana-ramalingam deleted the perplexity-test branch October 16, 2024 06:46
Comment on lines +18 to +21
llama_8b_f16_gguf_path = "/data/extra/models/llama3.1_8B/llama8b_f16.gguf"
llama_8b_f16_tokenizer_path = (
"/data/extra/models/llama3.1_8B/tokenizer_config.json"
)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Unit tests failed on this PR and after merge:


Any files required to run a test should be either

  • included in the repository (if small enough)
  • downloaded (and cached) on demand as part of the test
  • downloaded (and cached) ahead of time via a script

As this is, this test will only run on a machine that has already run some unknown, undocumented setup steps.

Comment on lines +59 to +60
- name: Run perplexity test
run: pytest sharktank/tests/evaluate/perplexity_test.py
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If more tests are going in this category, we could use pytest marks or some other filtering to pick up the list of tests to run. As this is now, the new tests are running in multiple workflows (on every commit and here on a nightly schedule).

@@ -7,6 +7,7 @@ onnx==1.15.0
huggingface-hub==0.22.2
transformers==4.40.0
sentencepiece==0.2.0
datasets==3.0.0
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We should steer towards putting requirements in the subproject requirements files instead of this top level file, especially if this is a test-only requirement

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants