Metadata-Version: 2.4
Name: dtaudit
Version: 0.0.1
Summary: Diagnostic audits for drug-target affinity (DTA) benchmarks (name reserved; first release forthcoming)
Author-email: danielLuo2 <31luoxudong@gmail.com>
License-Expression: MIT
Keywords: drug-target affinity,DTA,benchmark,evaluation,bioinformatics
Classifier: Development Status :: 1 - Planning
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.9
Description-Content-Type: text/markdown

# dtaudit

Diagnostic audits for drug-target affinity (DTA) benchmarks.

`dtaudit` takes a table of test-set predictions (drug, target, label,
prediction) and reports what pooled MSE and concordance index hide: how a model
compares with trivial drug / target main-effect baselines, whether it captures
drug-target interactions beyond those main effects, whether its predictions
have collapsed toward a constant, and how censored affinity labels affect the
evaluation.

**This release only reserves the package name and contains no functionality.**
The first functional release is in preparation; please do not depend on this
version.
