Skip to content

return overall_score from MTBenchBranch.judge_answers() #138

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.

Sign up for GitHub

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

Merged
danmcp merged 2 commits into from
Sep 28, 2024
Merged

return overall_score from MTBenchBranch.judge_answers() #138

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
219bca1
return overall_score from MTBenchBranch.generate_judgement()
alimaredia Sep 26, 2024
b22b40b
update flag for basic workflow tests
alimaredia Sep 27, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .github/workflows/e2e-nvidia-t4-x1.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -142,7 +142,7 @@ jobs:
working-directory: ./instructlab
run: |
. venv/bin/activate
./scripts/basic-workflow-tests.sh -m
./scripts/basic-workflow-tests.sh -msq

stop-runner:
name: Stop external EC2 runner
Expand Down
6 changes: 4 additions & 2 deletions src/instructlab/eval/mt_bench.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -260,10 +260,12 @@ def judge_answers(
serving_gpus Number of gpus allocated for serving. Used to tune with max_workers=auto. None indicates to use value specified in constructor.

Returns:
overall_score overall score from the evaluation
qa_pairs Question and answer pairs (with scores) from the evaluation
error_rate percentage of questions dropped due to errors during evaluation

@danmcp danmcp Sep 27, 2024

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

If you do make another change to this commit, this is also missing for the mt_bench case above

"""
logger.debug(locals())
_, qa_pairs, _, error_rate = mt_bench_judgment.generate_judgment(
overall_score, qa_pairs, _, error_rate = mt_bench_judgment.generate_judgment(
self.model_name,
self.judge_model_name,
server_url,
Expand All @@ -275,4 +277,4 @@ def judge_answers(
bench_name="mt_bench_branch",
merge_system_user_message=self.merge_system_user_message,
)
return qa_pairs, error_rate
return overall_score, qa_pairs, error_rate
6 changes: 5 additions & 1 deletion tests/test_branch_judge_answers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,11 @@
"../taxonomy",
"main",
)
qa_pairs, error_rate = mt_bench_branch.judge_answers("http://localhost:8000/v1")
overall_score, qa_pairs, error_rate = mt_bench_branch.judge_answers(
"http://localhost:8000/v1"
)

print(f"Overall Score: {overall_score}")
print(f"Error Rate: {error_rate}")
print(f"QA Pair 0:")
pprint.pprint(qa_pairs[0])
Expand Down