glmpynet 0.4.16

A scikit-learn compatible wrapper for glmnet-based logistic regression.

glmpynet

glmnetpp-based Logistic Regression for Scikit-Learn

glmpynet is a Python package that provides a scikit-learn-compatible LogisticRegression API powered by the high-performance glmnetpp C++ library, focusing on regularized logistic regression for binary classification. The initial version uses glmnetpp’s default settings (sourced from glmnet’s R documentation or online resources) for core methods (fit, predict), with parameters like C and penalty planned for later releases.

This project bridges the computational speed of glmnetpp with the ease-of-use of the Python data science ecosystem, acting as a drop-in replacement for sklearn.linear_model.LogisticRegression.

Key Features

• High Performance: Leverages the optimized glmnetpp C++ backend for fitting models, suitable for large datasets.
• Scikit-learn Compatible: Implements fit and predict methods, enabling seamless integration with sklearn pipelines and tools.
• Robust Development: Built with Bazel and Conda for reproducible builds, with comprehensive glmnetpp testing for reliability.

Installation

Once released, install glmpynet via pip:

pip install glmpynet

Quick Start

Using glmpynet is as simple as any scikit-learn estimator.

import numpy as np
from glmpynet import LogisticRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

# 1. Generate synthetic data
X, y = np.random.rand(100, 10), np.random.randint(0, 2, 100)
X_train, X_test, y_train, y_test = train_test_split(X, y)

# 2. Instantiate and fit the model
model = LogisticRegression()  # Uses glmnetpp defaults
model.fit(X_train, y_train)

# 3. Make predictions
y_pred = model.predict(X_test)

# 4. Evaluate the model
accuracy = accuracy_score(y_test, y_pred)
print(f"Model Accuracy: {accuracy:.2f}")

Project Status

glmpynet is in early development, focusing on drafting Sphinx documentation, validating the glmnetpp environment with all tests, and implementing a minimal LogisticRegression with default settings. Future releases will add parameters (e.g., C, penalty) and advanced features like multi-class support. See ROADMAP.md for details.

Contributing

Contributions are welcome! See CONTRIBUTING.md for guidelines on reporting bugs, suggesting features, or submitting pull requests.

License

This project is distributed under the MIT License.