SDOH Coding Model #538
Merged
SDOH Coding Model #538
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
plandes
approved these changes
Sep 9, 2025
dalloliogm
pushed a commit
to dalloliogm/PyHealth
that referenced
this pull request
Nov 26, 2025
* add sentence level sdoh multi-label classification model and test * revert test mask * sdoh: llama download HF api key; response parse test; fix inf test * doc * just added more details to the docs * readthedocs updates --------- Co-authored-by: Paul Landes <landes@mailc.net> Co-authored-by: John Wu <johnwu3@sunlab-serv-03.cs.illinois.edu>
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Outdated suggestions cannot be applied.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 pull request introduces a new social determinants of health (SDoH) sentence classifier based on a fine-tuned large language model, adds its dependencies, and provides a corresponding test. It also adds a utility for reproducible random seed setting in tests. The main changes are grouped below:
New SDoH Classifier Implementation:
SdohClassifierinpyhealth/models/sdoh.py, which uses a fine-tuned Llama model with PEFT adapters to predict SDoH labels from clinical sentences. Includes prompt engineering, model loading, and response parsing logic.Dependency Management:
peftandaccelerateto thepyproject.tomldependencies to support the new classifier's model loading and inference.Testing and Reproducibility:
TestSdohintests/core/test_sdoh.pyto validate classifier predictions with a sample sentence, ensuring correct label extraction.set_random_seedutility intests/base.pyfor deterministic test runs, including CUDA and cuDNN configuration for reproducibility.