SALSA.jlhttp://mloss.orgUpdates and additions to SALSA.jlenMon, 28 Sep 2015 17:28:56 -0000SALSA.jl 0.0.5<html><p>SALSA (Software lab for Advanced machine Learning with Stochastic Algorithms) is a native Julia implementation of the well-known stochastic algorithms for Regularized Empirical Risk minimization and sparse linear modelling. It aims at bridging a gap between sophisticated machine learning concepts and user-friendly environment with API which guides the user through many intermediate selection steps. The API features low- and high-level routines for both an experienced and a novice user. SALSA is managed and versioned by the package manager embedded in Julia. </p> <p>The SALSA software package can be used as a framework and a library. As a framework it embraces all steps of creating a valid ML model, such as cross-validation and hyperparameter tuning. As a library it can be used for embedding separate core algorithmic and preprocessing routines into other frameworks as well. Nice interoperability between the Julia eco-system and other languages (such as C/C++, Python etc.) ensures seamless and quick integration. </p></html>Vilen JumutcMon, 28 Sep 2015 17:28:56 -0000 vector machinesonline learningstochastic gradient descentmachine learningoptimizationalgorithmssupervised learningsparse learningsparsityregularizationpredictiononline multiclass classificationsparse datastatistical learningmachine learning toolboxstochastic optimization