Metadata-Version: 2.3
Name: fibermc
Version: 0.0.4
Summary: A Jax-based differentiable Monte Carlo estimator with applications to differentiable simulation, computational geometry, and topology optimization.
Project-URL: Homepage, https://github.com/PrincetonLIPS/fibers-standalone/tree/main
Project-URL: Issues, https://github.com/PrincetonLIPS/fibers-standalone/issues
Author-email: Nick Richardson <njkrichardson@princeton.edu>
License: MIT
Keywords: computational geometry,implicit differentiation,jax,optimization,topology optimization
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.8
Requires-Dist: chex>=0.1.87
Requires-Dist: jax>=0.4.30
Requires-Dist: jaxlib>=0.4.30
Requires-Dist: ml-dtypes>=0.5.0
Requires-Dist: numpy>=2.0.0
Requires-Dist: opt-einsum>=3.4.0
Requires-Dist: scipy>=1.13.1
Provides-Extra: all
Requires-Dist: jaxopt>=0.8.3; extra == 'all'
Requires-Dist: shapely>=2.0.6; extra == 'all'
Provides-Extra: jaxopt
Requires-Dist: jaxopt>=0.8.3; extra == 'jaxopt'
Provides-Extra: shapely
Requires-Dist: shapely>=2.0.6; extra == 'shapely'
Description-Content-Type: text/markdown

# Fiber Monte Carlo 

Fiber Monte Carlo (FMC) is a differentiable variant of the [simple Monte Carlo](https://en.wikipedia.org/wiki/Monte_Carlo_method) estimator designed with 
low-dimensional geometric-oriented applications in mind. The methodological and theoretical aspects of FMC are outlined in the accompanying [paper](https://openreview.net/pdf?id=sP1tCl2QBk), but this Python package contains implementations of a variety of general-purpose estimators with FMC as the underlying method, as well as utilities specific applications like computational geometry, differentiable rendering and topology optimization. 
