DRAGON Evaluation
Evaluation method for the DRAGON (Diagnostic Report Analysis: General Optimization of NLP) challenge.
If you are using DRAGON resources, please cite the following article:
J. S. Bosma, K. Dercksen, L. Builtjes, R. André, C, Roest, S. J. Fransen, C. R. Noordman, M. Navarro-Padilla, J. Lefkes, N. Alves, M. J. J. de Grauw, L. van Eekelen, J. M. A. Spronck, M. Schuurmans, A. Saha, J. J. Twilt, W. Aswolinskiy, W. Hendrix, B. de Wilde, D. Geijs, J. Veltman, D. Yakar, M. de Rooij, F. Ciompi, A. Hering, J. Geerdink, and H. Huisman on behalf of the DRAGON consortium. The DRAGON benchmark for clinical NLP. npj Digital Medicine 8, 289 (2025). https://doi.org/10.1038/s41746-025-01626-x
Download the citation file for your reference manager: BibTeX | RIS
Installation
A pre-built Docker container with the DRAGON evaluation method is available:
docker pull joeranbosma/dragon_eval:latest
The DRAGON evaluation method can be pip-installed:
pip install dragon_eval
Or, alternatively, it can be installed from source:
pip install git+https://github.com/DIAGNijmegen/dragon_eval
The evaluation method was tested with Python 3.10. See requirements.txt for a full list of exact package versions.
Usage
The Docker container can be used to evaluate the synthetic datasets as specified in evaluate.sh. To evaluate the synthetic tasks, place the predictions to evaluate in the test-predictions folder and run ./evaluate.sh.
The DRAGON evaluation method can also be used from the command line (if installed with pip):
python -m dragon_eval --ground-truth-path="ground-truth" --predictions-path=test-predictions --output-file=metrics.json --folds 0 1 2 3 4 --tasks 000 001 002 003 004 005 006 007
The command above should work when executed from the dragon_eval folder, which needs to be cloned locally for the ground truth and prediction files to be present. Change the paths above when executing the command from a different place or storing the files in a different place. The tasks and folds to evaluate can be changed with the respective parameters.
Managed By
Diagnostic Image Analysis Group, Radboud University Medical Center, Nijmegen, The Netherlands
Contact Information
Joeran Bosma: Joeran.Bosma@radboudumc.nl
Metadata
Release files for dragon-eval 0.2.11
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| dragon_eval-0.2.11.tar.gz | 21.5 kB | Without provenance | Details for dragon_eval-0.2.11.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| dragon_eval-0.2.11-py3-none-any.whl | Python 3 | none | any | Details for dragon_eval-0.2.11-py3-none-any.whl |
Total release size: 43.8 kB
Release files / dragon_eval-0.2.11.tar.gz
| Download URL | dragon_eval-0.2.11.tar.gz |
|---|---|
| Size | 21.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
eacd09cdba89ede7da81f775fde98bdf688e5b809c0cc4a6415062c4b86ae70e
|
|
BLAKE2b-256 checksum How to use checksums |
82ffd184795e26faab1dbda78e916fb6372c3b8bc5d6c08fa254a7ce3cc6ec66
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.10.16
|
Release files / dragon_eval-0.2.11-py3-none-any.whl
| Download URL | dragon_eval-0.2.11-py3-none-any.whl |
|---|---|
| Size | 22.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
f7adf819326aef83841d417b2a7b332bbfa0073b94cb48032bac7747c3269b4d
|
|
BLAKE2b-256 checksum How to use checksums |
71cc58027a61264a85aac19d2ebba05c78f5eef26bbb108eed7aa593fda2d402
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.10.16
|
Security for dragon-eval 0.2.11
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.