Metadata-Version: 2.4
Name: abbeflow
Version: 0.0.1
Summary: A learned prior over the optical glass manifold for realistic glass selection in lens design.
Project-URL: Homepage, https://github.com/HarrisonKramer/abbeflow
Author: Kramer Harrison
License: MIT
License-File: LICENSE
Classifier: Development Status :: 1 - Planning
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering :: Physics
Requires-Python: >=3.10
Requires-Dist: numpy
Requires-Dist: torch
Provides-Extra: optiland
Requires-Dist: optiland; extra == 'optiland'
Provides-Extra: train
Requires-Dist: optiland; extra == 'train'
Description-Content-Type: text/markdown

# AbbeFlow

AbbeFlow will be a normalizing-flow-based learned prior over the optical glass catalog manifold. It is trained on ~1700 glasses from 6 major manufacturers, predicting Buchdahl dispersion coefficients, in order to keep glass choices on the realizable manifold during lens optimization.

**Status**: AbbeFlow is currently an early pre-release / work in progress. Version v0.0.1 is simply a name-reservation scaffold on PyPI.
