python-glmnet-fix 2.2.2

Python wrapper for glmnet

This is a maintained fork of civisanalytics/python-glmnet, which was archived in 2024. The original package on PyPI (glmnet) can no longer be installed with modern Python toolchains due to build system issues.

Changes from the original

This fork applies minimal fixes to restore installability:

• Fixed invalid python_requires specifier (>=3.6.* → >=3.6)

• Added pyproject.toml with proper build dependencies (setuptools<60, numpy<2) to support pip build isolation

No changes were made to the library code itself. All credit goes to the original authors at Civis Analytics.

• Original repository: https://github.com/civisanalytics/python-glmnet (archived)

• Fork repository: https://github.com/randrover/python-glmnet

• License: GPLv2 (same as original)

Installation

pip install python-glmnet-fix

A Fortran compiler is required for building from source. For Mac users, brew install gcc will take care of this requirement.

Usage

The import name remains glmnet, same as the original:

from glmnet import LogitNet, ElasticNet

# Regularized Logistic Regression
m = LogitNet()
m = m.fit(x, y)
p = m.predict_proba(x)

# Regularized Linear Regression
m = ElasticNet()
m = m.fit(x, y)
p = m.predict(x)

The API follows the conventions of Scikit-Learn, so it is expected to work with tools from that ecosystem. Both models accept dense or sparse arrays.

Dependencies

• numpy

• scikit-learn

• scipy