This repository contains notebooks used for pipeline development, training, and evaluation for MatchMiner-AI. Not all code will run out of the box, since some notebooks depend on an institutional dataset that cannot be shared because it contains protected health information (PHI). A Zip file containing shareable synthetic datasets is provided.
To run full pipeline inference on your own notes using the models trained on synthetic data, refer to notebook run_on_your_notes_and_trials.ipynb. That notebook will require slight edits to pull your input notes and may need edits to work with your own GPU setup (if any).
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
HF Inference API was unable to determine this model's library.