
Free Your Numbers
Showing 5 of 25 projects. Click any project card for scope, mentors, and proposal studio.
Mentors: Student: beingamanforever
This project implements an anova classdef for the GNU Octave statistics package, providing a unified object-oriented interface for one-way, two-way, and N-way analysis of variance. The anova class delegates core computation to the existing anova1, anova2, and anovan functions, preserving their tested logic while exposing a clean, MATLAB-compatible OOP API with properties (Coefficients, anova table, fitted values, residuals, diagnostics) and methods (summary, multcompare, plotDiagnostics, predict, getEffectSizes). The class stores model state, enabling interactive post-hoc workflows without refitting. Beyond the classdef, the project delivers targeted performance improvements to the underlying functions: vectorizing loop-heavy group statistics in anova1 with accumarray, vectorizing anova2 SSE computation, reducing memory in anovan interaction column construction, and adding NaN handling to anova2. Deliverables: (1) anova classdef with constructor, properties, and core methods, (2) summary and display methods with formatted ANOVA table output, (3) multcompare method for post-hoc pairwise comparisons, (4) plotDiagnostics method for residual and leverage plots, (5) performance optimizations to anova1, anova2, anovan internals, (6) NaN handling for anova2, (7) comprehensive BISTs and Texinfo documentation, (8) benchmark suite. All computation reuses existing functions with no architectural rewrites.
Mentors: Student: Avanish Salunke
The GNU Octave Statistics package currently lacks an object-oriented framework for linear regression, missing the deep statistical diagnostics and memory-efficient deployment options found in MATLAB. This project aims to implement the LinearModel and CompactLinearModel classes to cleanly separate heavy model development from lightweight predictions. It will also modernize the fitlm function to support table inputs and Wilkinson formulas, and introduce a highly efficient stepwiselm function for automated model selection. The classes will include essential methods, extensive BIST testing, and comprehensive documentation, greatly improving Octave's statistical capabilities and MATLAB compatibility.
Mentors: Student: Thomas Kolman
I propose to work on the following project: porting the Chebfun package to GNU Octave. Chebfun is a numerical computation package for MATLAB that provides a framework for representing functions in a symbolic manner using ”Chebyshev technology”, a term coined by the package authors to describe the use of Chebyshev polynomials to represent functions up to machine precision. GNU Octave is supposed to be mostly compatible with MATLAB, so Chebfun should ideally work without change on the former, but there are some slight differences and unimplemented features in Octave that make this difficult. Primarily, the main issue is that Octave’s support for object-oriented programming (classdef) is not as feature complete and as stable as MATLAB’s, which is a major problem since Chebfun is written in classdef. The goal of this project is to port Chebfun to GNU Octave, and to make it work as a dropin replacement for the MATLAB version. Since fixing classdef is not a feasible/recommended goal for a GSoC project, I will instead focus primarily on making the necessary changes to the Chebfun codebase to make it work in Octave. Along the way, I will compile new information about the status of classdef in Octave, and I will share my findings with the Octave developers, and perhaps aid in some changes to the Octave codebase itself.
Mentors: Student: Swayam Shah
The GNU Octave Statistics package lacks class definitions for knnsearch and rangesearch, and its KDTree method is currently disabled due to poor performance (GitHub issue #151). This project aims to implement KDTreeSearcher, ExhaustiveSearcher, and hnswSearcher classes, along with the createns helper function, to enhance clustering capabilities. The KDTree implementation will be optimized using a compiled Octave (.oct) library for faster index construction and queries. Each searcher class will include efficient knnsearch and rangesearch methods. Deliverables include optimized implementations, extensive testing, benchmarks, and comprehensive documentation, improving performance, usability, and MATLAB compatibility.
Mentors: Student: Ruchika Sonagote
This proposal aims to significantly extend the classification capabilities of the GNU Octave statistics package by implementing additional methods for the existing ClassificationKNN class and introducing new classification classdefs, namely ClassificationGAM and ClassificationDiscriminant. These enhancements will bridge the gap between GNU Octave and MATLAB in terms of advanced data analysis and machine learning functionalities, fostering a richer environment for scientific computing.