Medhelm Epic #3787
Open
Medhelm Epic #3787
Conversation
This file contains hidden or 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
yifanmai
requested changes
Aug 7, 2025
| openpyxl~=3.1 | ||
| python-docx~=1.1 | ||
| transformers~=4.45,<4.50 | ||
| evaluation-instruments @ git+https://github.com/epic-open-source/evaluation-instruments.git@1c4637e84fe4dc54f6695e438f3baca6b2cd4573 |
Collaborator
There was a problem hiding this comment.
PyPI packages cannot depend on packages outside PyPI. You should instead provide instructions to users to manually install this this package, either by printing the installation command in an error message, or by documenting it in ReadTheDocs.
| from helm.benchmark.annotation.model_as_judge import AnnotatorModelInfo, LLMAsJuryAnnotator | ||
| from helm.clients.auto_client import AutoClient | ||
|
|
||
| from evaluation_instruments import prep |
Collaborator
There was a problem hiding this comment.
Display an error if this is not installed:
from helm.common.optional_dependencies import OptionalDependencyNotInstalled
try:
from evaluation_instruments import prep
import evaluation_instruments.instruments.pdsqi_9.pdsqi_prompt as pdsqi
except ModuleNotFoundError as e:
# Provide manual instructions for installing evaluation-instruments from GitHub
# because PyPI does not allow installing dependencies directly from GitHub.
raise OptionalDependencyNotInstalled(
f"Optional dependency {e.name} is not installed. "
"Please run `evaluation-instruments @ git+https://github.com/epic-open-source/evaluation-instruments.git@1c4637e84fe4dc54f6695e438f3baca6b2cd4573` to install it."
) from e # noqa: E501| def get_note_summary_spec(config_path: Optional[str] = None) -> RunSpec: | ||
| if config_path is None: | ||
| package = "helm.benchmark.scenarios" | ||
| config_path = str(pkg_resources.files(package).joinpath("note_summary_scenario.yaml")) |
Collaborator
There was a problem hiding this comment.
You need to add *.yaml to the manifest, or this file will not actually get included in the package.
Line 3 in 89001e7
|
|
||
| return instances | ||
|
|
||
| def read_file(self, file_path: str) -> List[str]: |
|
|
||
|
|
||
| @dataclass | ||
| class Rubric: |
Collaborator
There was a problem hiding this comment.
I did not look at the rubric logic too closely, but let me know if there's anything you want me to check.
Collaborator
|
Ping - last activity was two weeks ago. |
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.
This PR adds a new scenario,
NoteSummaryScenario, developed in partnership with Epic Systems. The scenario focuses on generating clinical note summaries for the Emergency Medicine specialty, reflecting real-world medical documentation needs.To assess the quality of the model-generated summaries, we adopt the "LLM as a judge" evaluation framework based on PDSQI-9, a rubric co-developed by UW Madison and Epic Systems for systematic evaluation that doesn't require a gold standard response to compare against.