SLOPE

SLOPE is a C++ library for Sorted L-One Penalized Estimation (SLOPE). Its main purpose is to serve as a backend for R and Python packages, but it can also be used as a standalone library in the off-chance that you want to fit your models entirely through C++.

Public API

SLOPE provides a clean public API through the include/ directory with minimal dependencies. The public API exposes three main components:

• Core SLOPE functionality - Main classes for fitting SLOPE models (Slope, SlopeFit, SlopePath)
• Cross-validation - Hyperparameter tuning with k-fold cross-validation
• Proximal operators - Direct access to the sorted L1 penalty (SortedL1Norm)

To use the public API, simply include:

#include <slope.h> // Main public API header

Getting Started

First, we define our model. Let's use logistic regression, by setting the loss to "logistic".

#include <slope.h>

slope::Slope model;
model.setLoss("logistic");

Next, we set the data matrix x and the response vector y. Here we use some toy data.

Eigen::MatrixXd x(3, 2);
Eigen::VectorXd y(3);

x << 1.1, 2.3, 0.2, 1.5, 0.5, 0.2;
y << 0, 1, 0;

Finally, we call the path() method to fit the full SLOPE path.

auto res = model.path(x, y);

Now we can retrieve the coefficients by calling res.getCoefs().

Dependencies

Building

• A C++17 compiler
• CMake 3.15 or later
• Eigen 3.4 or later

Documentation

• Doxygen

Testing

• Catch2

Contributing

Please see CONTRIBUTING.md for guidelines on contributing to this project.