{"id":13423656,"url":"https://github.com/liquidSVM/liquidSVM","last_synced_at":"2025-03-15T17:32:09.258Z","repository":{"id":57731034,"uuid":"89102345","full_name":"liquidSVM/liquidSVM","owner":"liquidSVM","description":"Support vector machines (SVMs) and related kernel-based learning algorithms are a well-known class of machine learning algorithms, for non-parametric classification and regression. liquidSVM is an implementation of SVMs whose key features are: fully integrated hyper-parameter selection, extreme speed on both small and large data sets, full flexibility for experts, and inclusion of a variety of different learning scenarios: multi-class classification, ROC, and Neyman-Pearson learning, and least-squares, quantile, and expectile regression.","archived":false,"fork":false,"pushed_at":"2020-02-20T15:10:12.000Z","size":4574,"stargazers_count":67,"open_issues_count":17,"forks_count":9,"subscribers_count":5,"default_branch":"master","last_synced_at":"2025-03-10T09:22:06.003Z","etag":null,"topics":["apache-spark","c-plus-plus","classification","expectile-regression","machine-learning","matlab","ml","octave","python","quantile-regression","r","r-package","regression","rstats","svm"],"latest_commit_sha":null,"homepage":null,"language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"agpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/liquidSVM.png","metadata":{"files":{"readme":"README.md","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":"2017-04-22T22:16:26.000Z","updated_at":"2025-01-21T12:26:10.000Z","dependencies_parsed_at":"2022-09-26T22:01:43.672Z","dependency_job_id":null,"html_url":"https://github.com/liquidSVM/liquidSVM","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/liquidSVM%2FliquidSVM","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liquidSVM%2FliquidSVM/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liquidSVM%2FliquidSVM/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liquidSVM%2FliquidSVM/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/liquidSVM","download_url":"https://codeload.github.com/liquidSVM/liquidSVM/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243767109,"owners_count":20344873,"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":["apache-spark","c-plus-plus","classification","expectile-regression","machine-learning","matlab","ml","octave","python","quantile-regression","r","r-package","regression","rstats","svm"],"created_at":"2024-07-31T00:00:39.943Z","updated_at":"2025-03-15T17:32:06.308Z","avatar_url":"https://github.com/liquidSVM.png","language":"C++","readme":"\n## General Information\n\nSupport vector machines (SVMs) and related kernel-based learning algorithms are\na well-known class of machine learning algorithms, for non-parametric\nclassification and regression. **liquidSVM** is an implementation of\nSVMs whose key features are:\n\n* fully integrated hyper-parameter selection,\n* extreme speed on both small and large data sets,\n* Bindings for [R](#R), [Python](#python), [MATLAB / Octave](#matlab-octave), [Java](#java), and [Spark](#spark),\n* full flexibility for experts, and\n* inclusion of a variety of different learning scenarios:\n    - multi-class classification, ROC, and Neyman-Pearson learning,\n    - least-squares, quantile, and expectile regression.\n\n\nFor questions and comments just contact us via\n[mail](http://www.uni-stuttgart.de/cgi-bin/mail.cgi?liquidSVM=mathematik.uni-stuttgart.de).\nThere you also can ask to be registerd to our mailing list.\n\nliquidSVM is licensed under [AGPL 3.0](http://www.gnu.org/licenses/agpl-3.0.html). In case you need another license, please contact [me](http://www.isa.uni-stuttgart.de/Steinwart/).\n\n## Command Line interface\n\n[Installation instructions](http://www.isa.uni-stuttgart.de/software/install.txt) for the command line versions.\n\n|                                      |                                                                                    |\n|--------------------------------------|------------------------------------------------------------------------------------|\n| Terminal version for Linux/OS X      | [liquidSVM.tar.gz](http://www.isa.uni-stuttgart.de/software/liquidSVM.tar.gz)      |\n| Terminal version for Windows (64bit) | avx2: [liquidSVM.zip](http://www.isa.uni-stuttgart.de/software/avx2/liquidSVM.zip) |\n|                                      | avx:  [liquidSVM.zip](http://www.isa.uni-stuttgart.de/software/avx/liquidSVM.zip)  |\n|                                      | sse2: [liquidSVM.zip](http://www.isa.uni-stuttgart.de/software/sse2/liquidSVM.zip) |\n| Previous versions                    | [v1.1](v1.1) (June 2016), [v1.0](v1.0) (January 2016)                              |\n\nOn Linux and Mac on the terminal `liquidSVM` can be used in the following way:\n```bash\nwget www.isa.uni-stuttgart.de/software/liquidSVM.tar.gz\ntar xzf liquidSVM.tar.gz\ncd liquidSVM\nmake all\nscripts/mc-svm.sh banana-mc 1 2\n```\n\n## R\n\nRead the [demo vignette](http://www.isa.uni-stuttgart.de/software/R/demo.html) for a tutorial on installing liquidSVM-package and how to use it and the [documentation vignette](http://www.isa.uni-stuttgart.de/software/R/documentation.html) for more advanced installation options and usage.\n\nAn easy usage is:\n```r\ninstall.packages(\"liquidSVM\")\nlibrary(liquidSVM)\nbanana \u003c- liquidData('banana-mc')\nmodel \u003c- mcSVM( Y~. , banana$train, display=1, threads=2)\nresult \u003c- test(model, banana$test)\nerrors(result)\n```\n\n## Python\n\nRead the [demo notebook](http://www.isa.uni-stuttgart.de/software/python/demo.html) for a tutorial on installing liquidSVM-package and how to use it and the [homepage](bindings/python/) for more advanced installation options and usage.\n\nTo install use:\n```bash\npip install --user liquidSVM\n```\nand then in Python you can use it e.g. like:\n```python\nfrom liquidSVM import *\nbanana = LiquidData('banana-mc')\nmodel = mcSVM(banana.train, display=1, threads=2)\nresult, err = model.test(banana.test)\n```\n\n\n## MATLAB / Octave\n\nThe [MATLAB bindings](bindings/matlab/) are currently getting a better interface,\nand this is a preview version.\n\n\u003e It does currently not work on Windows.\n\nFor installation download the Toolbox\n[liquidSVM.mltbx](http://www.isa.uni-stuttgart.de/software/matlab/liquidSVM.mltbx)\nand install it in MATLAB by double clicking it.\nTo compile and add paths issue:\n```matlab\nmakeliquidSVM native\n```\nThen you can use it like:\n```matlab\nbanana = liquidData('banana-mc');\nmodel = svm_mc(banana.train, 'DISPLAY', 1, 'THREADS', 2);\n[result, err] = model.test(banana.test);\n```\n\nMost of the code also works in `Octave`\nif you use [liquidSVM-octave.zip](http://www.isa.uni-stuttgart.de/software/matlab/liquidSVM-octave.zip).\n\n\n## Java\nThe main homepage is [here](bindings/java/).\nFor installation download [liquidSVM-java.zip](http://www.isa.uni-stuttgart.de/software/java/liquidSVM-java.zip) and unzip it.\nThe classes are all in package `de.uni_stuttgart.isa.liquidsvm` and an easy example is:\n```java\nLiquidData banana = new LiquidData(\"banana-mc\");\nSVM model = new MC(banana.train, new Config().display(1).threads(2));\nResultAndErrors result = model.test(banana.test);\n```\nIf this is implemented in the file `Example.java` this can be compiled and run using\n```bash\n# if you want to compile the JNI-native library:\nmake lib\n# compile your Java-Code\njavac -classpath liquidSVM.jar Example.java\n# and run it\njava -Djava.library.path=. -cp .:liquidSVM.jar Example\n```\n\n## Spark\nThis is a preview version, see [Spark](bindings/spark/) for more details.\nDownload [liquidSVM-spark.zip](http://www.isa.uni-stuttgart.de/software/spark/liquidSVM-spark.zip) and unzip it.\nAssume you have `Spark` installed in `$SPARK_HOME` you can issue:\n```bash\nmake lib\nexport LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH\n$SPARK_HOME/bin/spark-submit \\\n  --class de.uni_stuttgart.isa.liquidsvm.spark.App \\\n  liquidSVM-spark.jar banana-mc\n```\nIf you have configured `Spark` to be used on a cluster with `Hadoop` use:\n```bash\nhdfs dfs -put data/covtype-full.train.csv data/covtype-full.test.csv .\nmake lib\n$SPARK_HOME/bin/spark-submit --files ../libliquidsvm.so \\\n  --conf spark.executor.extraLibraryPath=. \\\n  --conf spark.driver.extraLibraryPath=. \\\n  --class de.uni_stuttgart.isa.liquidsvm.spark.App \\\n  --num-executors 14 liquidSVM-spark.jar covtype-full\n```\n\n\n\nExtra Datasets for the Demo\n---------------------------\n\n[covertype data set with 35.090 training and 34.910 test samples](http://www.isa.uni-stuttgart.de/software/covtype.35000.zip)\n\n[covertype data set with 522.909 training and 58.103 test samples](http://www.isa.uni-stuttgart.de/software/covtype.zip)\n\nBoth datasets were compiled from [LIBSVM's version of the covertype dataset](https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html), which\nin turn was taken from the [UCI repository](http://mlr.cs.umass.edu/ml/datasets/Covertype) and preprocessed as in [\\[RC02a\\].](https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/ref.html#RC02a)\nCopyright for this dataset is by Jock A. Blackard and Colorado State University.\n\nCitation\n--------\n\nIf you use liquidSVM, please cite it as:\n\n\u003e I. Steinwart and P. Thomann.\n\u003e *liquidSVM: A fast and versatile SVM package.*\n\u003e [*ArXiv e-prints 1702.06899*](http://arxiv.org/abs/1702.06899), February 2017.\n","funding_links":[],"categories":["C++","Machine Learning"],"sub_categories":["Kernel Methods"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FliquidSVM%2FliquidSVM","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FliquidSVM%2FliquidSVM","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FliquidSVM%2FliquidSVM/lists"}