Evaluation harness for the MedEval triage agent. Scores accuracy, safety, hallucination, cost, and latency.
Project description
medeval-harness
Evaluation harness for the MedEval triage agent.
Scores a running MedEval agent against a 50-case ESI dataset on:
- Exact and adjacent ESI level accuracy
- Under-triage and over-triage rates (the safety metrics)
- Hallucination rate (LLM-extracted facts unsupported by complaint text)
- Decision-path consistency
- Cost per case and per-evaluation
- Latency (p50 / p95)
Install
pip install medeval-harness
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
medeval_harness-0.1.0.tar.gz
(16.4 kB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file medeval_harness-0.1.0.tar.gz.
File metadata
- Download URL: medeval_harness-0.1.0.tar.gz
- Upload date:
- Size: 16.4 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.4
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
3debe19f34388a2e26956416828a66b4715bfea06fe89f59014919ac94e61c7b
|
|
| MD5 |
aa336ca15c156f355e50b947db22f3e2
|
|
| BLAKE2b-256 |
92afeff1cbada98e86e4b3115036b83f55b77fdf4b42129869dd58c8b916b0ce
|
File details
Details for the file medeval_harness-0.1.0-py3-none-any.whl.
File metadata
- Download URL: medeval_harness-0.1.0-py3-none-any.whl
- Upload date:
- Size: 16.3 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.4
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
23ccb0f9331c0c8843aba300c223d1bd4eb635bfb81dbfe7a903dc4c2a1ead4c
|
|
| MD5 |
8eb536b9fe674646839b7adf29f08d5f
|
|
| BLAKE2b-256 |
2ed105ce2d0411d461b9970292a21bccedce12b80d24ba14e3be344ac1abec64
|
