{"id":18439459,"url":"https://github.com/idiap/ml3","last_synced_at":"2025-10-23T22:42:49.988Z","repository":{"id":11222474,"uuid":"13612670","full_name":"idiap/ML3","owner":"idiap","description":"ML3 classifier (Multiclass Latent Locally Linear Support Vector Machines)","archived":false,"fork":false,"pushed_at":"2015-04-10T13:55:15.000Z","size":1359,"stargazers_count":5,"open_issues_count":0,"forks_count":5,"subscribers_count":8,"default_branch":"master","last_synced_at":"2025-03-06T03:04:12.275Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/idiap.png","metadata":{"files":{"readme":"README.rst","changelog":null,"contributing":null,"funding":null,"license":"COPYING","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2013-10-16T07:40:29.000Z","updated_at":"2016-10-31T12:17:24.000Z","dependencies_parsed_at":"2022-09-03T21:30:20.255Z","dependency_job_id":null,"html_url":"https://github.com/idiap/ML3","commit_stats":null,"previous_names":[],"tags_count":2,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/idiap%2FML3","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/idiap%2FML3/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/idiap%2FML3/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/idiap%2FML3/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/idiap","download_url":"https://codeload.github.com/idiap/ML3/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247732689,"owners_count":20986903,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-06T06:24:51.070Z","updated_at":"2025-10-23T22:42:44.920Z","avatar_url":"https://github.com/idiap.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"Multiclass Latent Locally Linear SVM\n==========================================\n\n| Copyright (c) 2013 Idiap Research Institute, http://www.idiap.ch/\n| Written by `Marco Fornoni \u003chttp://fornoni.github.io/\u003e`_ \u003cmarco.fornoni@alumni.epfl.ch\u003e\n|\n| Idiap Research Institute,\n| Centre du Parc, P.O. Box 592,\n| Rue Marconi 19,\n| 1920 Martigny, Switzerland\n| Telephone: +41 27 721 77 57\n| Fax: +41 27 721 77 12\n\nThis file is part of the ML3 Software.\n\nML3 is free software: you can redistribute it and/or modify\nit under the terms of the GNU General Public License version 3 as\npublished by the Free Software Foundation.\n\nML3 is distributed in the hope that it will be useful,\nbut WITHOUT ANY WARRANTY; without even the implied warranty of\nMERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\nGNU General Public License for more details.\n\nYou should have received a copy of the GNU General Public License\nalong with ML3. If not, see \u003chttp://www.gnu.org/licenses/\u003e.\n\n\nAbout\n-----\nKernelized Support Vector Machines (SVM) have gained the status of off-the-shelf \nclassifiers, able to deliver state of the art performance on almost any problem. \nStill, their practical use is constrained by their computational and memory \ncomplexity, which grows super-linearly with the number of training samples. \nIn order to retain the low training and testing complexity of linear classifiers \nand the exibility of non linear ones, a growing, promising alternative is \nrepresented by methods that learn non-linear classifiers through local combinations \nof linear ones.\n\nThe `Multiclass Latent Locally Linear SVM \u003chttp://publications.idiap.ch/downloads/papers/2013/Fornoni_ACML2013_2013.pdf\u003e`_ \n(ML3) can learn complex decision functions, traditionally given by kernels, through \nthe use of locally linear decision functions. Differently from kernel classifiers, \nML3 makes use of a set of linear models that are locally linearly combined to form \na non-linear decision boundary in the input space. Thanks to the latent \nformulation, the combination coefficients are modeled as latent variables and \nefficiently estimated using an analytic solution.\n\nML3 has potential applications on large-scale problems, requiring powerful \nclassifiers and efficient learning methods, whose training complexity with \nrespect to the number of samples is not super-linear.\n\n\nUsage\n-----\nThis is a mixed C++ and MATLAB (c) implementation of the ML3 \nalgorithm, with the main algorithm being implemented in a mex file. \nIt is develped under Ubuntu 12.10, Matlab R2013a and it makes use\nof the `Eigen 3.1 library \u003chttp://eigen.tuxfamily.org\u003e`_.\nConfigurations differing from the above are not officially supported.\n\nIn order to use the software you need to:\n\n1. Install the Eigen 3.1 library, using:\n    `$ sudo apt-get install libeigen3-dev`\n\n2. Compile ML3 for your architecture, using \n    `$ make`\n\n3. From MATLAB, instantiate the ML3 algorithm using \n    `algo=ML3();`\n\n4. Train the algorithm using\n    `model=algo.train(features,labels);`\n\n5. Test the algorithm using \n    `[dec_values,predict_labels,accuracy,confusion]=algo.test(features,labels,model);`\n\n\nCite ML3\n--------\nIf you find this software useful, please cite::\n\n  @INPROCEEDINGS{Fornoni_ACML2013_2013,\n         author = {Fornoni, Marco and Caputo, Barbara and Orabona, Francesco},\n         editor = {Ong, Cheng Soon and Ho, Tu-Bao},\n       keywords = {Latent SVM, Locally Linear Support Vector Machines, multiclass classification},\n       projects = {Idiap},\n          title = {Multiclass Latent Locally Linear Support Vector Machines},\n      booktitle = {JMLR W\\\u0026CP, Volume 29: Asian Conference on Machine Learning},\n           year = {2013},\n          pages = {229-244},\n       location = {Canberra, Australia},\n           issn = {1938-7228},\n            url = {http://jmlr.org/proceedings/papers/v29/},\n            pdf = {http://publications.idiap.ch/downloads/papers/2013/Fornoni_ACML2013_2013.pdf}\n  }\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fidiap%2Fml3","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fidiap%2Fml3","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fidiap%2Fml3/lists"}