{"Package":"mlr3","Title":"Machine Learning in R - Next Generation","Version":"0.3.0","Authors@R":"\nc(person(given = \"Michel\",\nfamily = \"Lang\",\nrole = c(\"cre\", \"aut\"),\nemail = \"michellang@gmail.com\",\ncomment = c(ORCID = \"0000-0001-9754-0393\")),\nperson(given = \"Bernd\",\nfamily = \"Bischl\",\nrole = \"aut\",\nemail = \"bernd_bischl@gmx.net\",\ncomment = c(ORCID = \"0000-0001-6002-6980\")),\nperson(given = \"Jakob\",\nfamily = \"Richter\",\nrole = \"aut\",\nemail = \"jakob1richter@gmail.com\",\ncomment = c(ORCID = \"0000-0003-4481-5554\")),\nperson(given = \"Patrick\",\nfamily = \"Schratz\",\nrole = \"aut\",\nemail = \"patrick.schratz@gmail.com\",\ncomment = c(ORCID = \"0000-0003-0748-6624\")),\nperson(given = \"Giuseppe\",\nfamily = \"Casalicchio\",\nrole = \"ctb\",\nemail = \"giuseppe.casalicchio@stat.uni-muenchen.de\",\ncomment = c(ORCID = \"0000-0001-5324-5966\")),\nperson(given = \"Stefan\",\nfamily = \"Coors\",\nrole = \"ctb\",\nemail = \"mail@stefancoors.de\",\ncomment = c(ORCID = \"0000-0002-7465-2146\")),\nperson(given = \"Quay\",\nfamily = \"Au\",\nrole = \"ctb\",\nemail = \"quayau@gmail.com\",\ncomment = c(ORCID = \"0000-0002-5252-8902\")),\nperson(given = \"Martin\",\nfamily = \"Binder\",\nrole = \"aut\",\nemail = \"mlr.developer@mb706.com\"))","Description":"Efficient, object-oriented programming on the\nbuilding blocks of machine learning. Provides 'R6' objects for tasks,\nlearners, resamplings, and measures. The package is geared towards\nscalability and larger datasets by supporting parallelization and\nout-of-memory data-backends like databases. While 'mlr3' focuses on\nthe core computational operations, add-on packages provide additional\nfunctionality.","License":"LGPL-3","URL":"https://mlr3.mlr-org.com, https://github.com/mlr-org/mlr3","BugReports":"https://github.com/mlr-org/mlr3/issues","Depends":"R (>= 3.1.0)","Imports":"R6 (>= 2.4.1), backports, checkmate (>= 2.0.0), data.table (>=\n1.12.8), digest, future.apply (>= 1.5.0), lgr (>= 0.3.4),\nmlbench, mlr3measures (>= 0.1.3), mlr3misc (>= 0.2.0), paradox\n(>= 0.2.0), uuid","Suggests":"Matrix, bibtex, callr, datasets, evaluate, future,\nfuture.callr, mlr3data, progressr, rpart, testthat","RdMacros":"mlr3misc","Encoding":"UTF-8","LazyData":"true","NeedsCompilation":"no","RoxygenNote":"7.1.0","Collate":"'mlr_reflections.R' 'BenchmarkResult.R' 'DataBackend.R'\n'DataBackendCbind.R' 'DataBackendDataTable.R'\n'DataBackendMatrix.R' 'DataBackendRbind.R'\n'DataBackendRename.R' 'Learner.R' 'LearnerClassif.R'\n'mlr_learners.R' 'LearnerClassifDebug.R'\n'LearnerClassifFeatureless.R' 'LearnerClassifRpart.R'\n'LearnerRegr.R' 'LearnerRegrFeatureless.R' 'LearnerRegrRpart.R'\n'Measure.R' 'MeasureClassif.R' 'mlr_measures.R'\n'MeasureClassifCosts.R' 'MeasureDebug.R' 'MeasureElapsedTime.R'\n'MeasureOOBError.R' 'MeasureRegr.R' 'MeasureSelectedFeatures.R'\n'MeasureSimple.R' 'Prediction.R' 'PredictionClassif.R'\n'PredictionRegr.R' 'ResampleResult.R' 'Resampling.R'\n'mlr_resamplings.R' 'ResamplingBootstrap.R' 'ResamplingCV.R'\n'ResamplingCustom.R' 'ResamplingHoldout.R'\n'ResamplingInsample.R' 'ResamplingRepeatedCV.R'\n'ResamplingSubsampling.R' 'Task.R' 'TaskSupervised.R'\n'TaskClassif.R' 'mlr_tasks.R' 'TaskClassif_german_credit.R'\n'TaskClassif_iris.R' 'TaskClassif_pima.R' 'TaskClassif_sonar.R'\n'TaskClassif_spam.R' 'TaskClassif_wine.R' 'TaskClassif_zoo.R'\n'TaskGenerator.R' 'mlr_task_generators.R'\n'TaskGenerator2DNormals.R' 'TaskGeneratorFriedman1.R'\n'TaskGeneratorSmiley.R' 'TaskGeneratorXor.R' 'TaskRegr.R'\n'TaskRegr_boston_housing.R' 'TaskRegr_mtcars.R'\n'TaskUnsupervised.R' 'Task_cbind.R' 'Task_rbind.R'\n'as_data_backend.R' 'assertions.R' 'auto_convert.R'\n'benchmark.R' 'benchmark_grid.R' 'default_measures.R'\n'deprecated.R' 'fix_factor_levels.R' 'helper.R'\n'mlr_coercions.R' 'mlr_sugar.R' 'predict.R' 'reexports.R'\n'resample.R' 'worker.R' 'zzz.R'","Packaged":"2020-06-02 08:56:41 UTC; michel","Author":"Michel Lang [cre, aut] (),\nBernd Bischl [aut] (),\nJakob Richter [aut] (),\nPatrick Schratz [aut] (),\nGiuseppe Casalicchio [ctb] (),\nStefan Coors [ctb] (),\nQuay Au [ctb] (),\nMartin Binder [aut]","Maintainer":"Michel Lang ","Repository":"CRAN","Date/Publication":"2020-06-02 12:00:02 UTC","repoType":"cran","tarballUrl":"ftp://cran.r-project.org/pub/R/src/contrib/mlr3_0.3.0.tar.gz","jsonAuthors":[{"name":"Michel Lang","email":"michellang@gmail.com","maintainer":true},{"name":"Bernd Bischl","email":"bernd_bischl@gmx.net","maintainer":false},{"name":"Jakob Richter","email":"jakob1richter@gmail.com","maintainer":false},{"name":"Patrick Schratz","email":"patrick.schratz@gmail.com","maintainer":false},{"name":"Giuseppe Casalicchio","email":"giuseppe.casalicchio@stat.uni-muenchen.de","maintainer":false},{"name":"Stefan Coors","email":"mail@stefancoors.de","maintainer":false},{"name":"Quay Au","email":"quayau@gmail.com","maintainer":false},{"name":"Martin Binder","email":"mlr.developer@mb706.com","maintainer":false}],"readme":"\n# mlr3

\n\nPackage website: [release](https://mlr3.mlr-org.com/) |\n[dev](https://mlr3.mlr-org.com/dev)\n\nEfficient, object-oriented programming on the building blocks of machine\nlearning. Successor of [mlr](https://github.com/mlr-org/mlr).\n\n\n\n[![R CMD Check via\n{tic}](https://github.com/mlr-org/mlr3/workflows/R%20CMD%20Check%20via%20%7Btic%7D/badge.svg?branch=master)](https://github.com/mlr-org/mlr3/actions)\n[![DOI](https://joss.theoj.org/papers/10.21105/joss.01903/status.svg)](https://doi.org/10.21105/joss.01903)\n[![CRAN Status\nBadge](https://www.r-pkg.org/badges/version-ago/mlr3)](https://cran.r-project.org/package=mlr3)\n[![Cran\nChecks](https://cranchecks.info/badges/worst/mlr3)](https://cran.r-project.org/web/checks/check_results_mlr3.html)\n[![codecov](https://codecov.io/gh/mlr-org/mlr3/branch/master/graph/badge.svg)](https://codecov.io/gh/mlr-org/mlr3)\n[![StackOverflow](https://img.shields.io/badge/stackoverflow-mlr3-orange.svg)](https://stackoverflow.com/questions/tagged/mlr3)\n[![Dependencies](https://tinyverse.netlify.com/badge/mlr3)](https://cran.r-project.org/package=mlr3)\n\n\n## Resources (for users and developers)\n\n - We *started* writing a [book](https://mlr3book.mlr-org.com/). This\n should be the central entry point to the package.\n - The [mlr3gallery](https://mlr3gallery.mlr-org.com) has some case\n studies and demonstrates how frequently occurring problems can be\n solved. It is still in early days so stay tuned for more to come.\n - **Videos**: Currently 2 recorded talks exist from useR\\!2019. Both\n are somewhat outdated already but can be useful to get a first\n overview about the package ecosystem and design ideas.\n - [mlr3](https://www.youtube.com/watch?v=wsP2hiFnDQs)\n - [mlr3pipelines and\n mlr3tuning](https://www.youtube.com/watch?v=gEW5RxkbQuQ)\n - [mlr3 reference manual](https://mlr3.mlr-org.com/reference/)\n - **Cheatsheets**\n - [Overview of cheatsheets](https://cheatsheets.mlr-org.com)\n - [mlr3](https://cheatsheets.mlr-org.com/mlr3.pdf)\n - [mlr3tuning](https://cheatsheets.mlr-org.com/mlr3tuning.pdf)\n - [mlr3pipelines](https://cheatsheets.mlr-org.com/mlr3pipelines.pdf)\n - **Templates/Tutorials**\n - [mlr3-learndrake](https://github.com/mlr-org/mlr3-learndrake):\n Shows how to use mlr3 with\n [drake](https://docs.ropensci.org/drake/) for reproducible ML\n workflow automation.\n - [List of extension\n packages](https://github.com/mlr-org/mlr3/wiki/Extension-Packages)\n - [mlr-outreach](https://github.com/mlr-org/mlr-outreach) contains\n public talks and slides resources.\n - Our [blog](https://mlr-org.com/) about *mlr* and *mlr3*. (We are not\n the most frequent bloggers ;) )\n - [Wiki](https://github.com/mlr-org/mlr3/wiki): Contains mainly\n information for developers.\n\n## Installation\n\nInstall the last release from CRAN:\n\n``` r\ninstall.packages(\"mlr3\")\n```\n\nInstall the development version from GitHub:\n\n``` r\nremotes::install_github(\"mlr-org/mlr3\")\n```\n\n## Example\n\n### Constructing Learners and Tasks\n\n``` r\nlibrary(mlr3)\n\n# create learning task\ntask_iris <- TaskClassif$new(id = \"iris\", backend = iris, target = \"Species\")\ntask_iris\n```\n\n ## (150 x 5)\n ## * Target: Species\n ## * Properties: multiclass\n ## * Features (4):\n ## - dbl (4): Petal.Length, Petal.Width, Sepal.Length, Sepal.Width\n\n``` r\n# load learner and set hyperparameter\nlearner <- lrn(\"classif.rpart\", cp = .01)\n```\n\n### Basic train + predict\n\n``` r\n# train/test split\ntrain_set <- sample(task_iris$nrow, 0.8 * task_iris$nrow)\ntest_set <- setdiff(seq_len(task_iris$nrow), train_set)\n\n# train the model\nlearner$train(task_iris, row_ids = train_set)\n\n# predict data\nprediction <- learner$predict(task_iris, row_ids = test_set)\n\n# calculate performance\nprediction$confusion\n```\n\n ## truth\n ## response setosa versicolor virginica\n ## setosa 11 0 0\n ## versicolor 0 12 1\n ## virginica 0 0 6\n\n``` r\nmeasure <- msr(\"classif.acc\")\nprediction$score(measure)\n```\n\n ## classif.acc \n ## 0.9666667\n\n### Resample\n\n``` r\n# automatic resampling\nresampling <- rsmp(\"cv\", folds = 3L)\nrr <- resample(task_iris, learner, resampling)\nrr$score(measure)\n```\n\n ## task task_id learner learner_id resampling\n ## 1: iris classif.rpart \n ## 2: iris classif.rpart \n ## 3: iris classif.rpart \n ## resampling_id iteration prediction classif.acc\n ## 1: cv 1 0.92\n ## 2: cv 2 0.92\n ## 3: cv 3 0.94\n\n``` r\nrr$aggregate(measure)\n```\n\n ## classif.acc \n ## 0.9266667\n\n## Why a rewrite?\n\n[mlr](https://github.com/mlr-org/mlr) was first released to\n[CRAN](https://cran.r-project.org/package=mlr) in 2013. Its core design\nand architecture date back even further. The addition of many features\nhas led to a [feature\ncreep](https://en.wikipedia.org/wiki/Feature_creep) which makes\n[mlr](https://github.com/mlr-org/mlr) hard to maintain and hard to\nextend. We also think that while mlr was nicely extensible in some parts\n(learners, measures, etc.), other parts were less easy to extend from\nthe outside. Also, many helpful R libraries did not exist at the time\n[mlr](https://github.com/mlr-org/mlr) was created, and their inclusion\nwould result in non-trivial API changes.\n\n## Design principles\n\n - Only the basic building blocks for machine learning are implemented\n in this package.\n - Focus on computation here. No visualization or other stuff. That can\n go in extra packages.\n - Overcome the limitations of R’s [S3\n classes](https://adv-r.hadley.nz/s3.html) with the help of\n [R6](https://cran.r-project.org/package=R6).\n - Embrace [R6](https://cran.r-project.org/package=R6) for a clean\n OO-design, object state-changes and reference semantics. This might\n be less “traditional R”, but seems to fit `mlr` nicely.\n - Embrace\n [`data.table`](https://cran.r-project.org/package=data.table) for\n fast and convenient data frame computations.\n - Combine `data.table` and `R6`, for this we will make heavy use of\n list columns in data.tables.\n - Defensive programming and type safety. All user input is checked\n with [`checkmate`](https://cran.r-project.org/package=checkmate).\n Return types are documented, and mechanisms popular in base R which\n “simplify” the result unpredictably (e.g., `sapply()` or `drop`\n argument in `[.data.frame`) are avoided.\n - Be light on dependencies. `mlr3` requires the following packages at\n runtime:\n - [`future.apply`](https://cran.r-project.org/package=future.apply):\n Resampling and benchmarking is parallelized with the\n [`future`](https://cran.r-project.org/package=future)\n abstraction interfacing many parallel backends.\n - [`backports`](https://cran.r-project.org/package=backports):\n Ensures backward compatibility with older R releases. Developed\n by members of the `mlr` team. No recursive dependencies.\n - [`checkmate`](https://cran.r-project.org/package=checkmate):\n Fast argument checks. Developed by members of the `mlr` team. No\n extra recursive dependencies.\n - [`mlr3misc`](https://cran.r-project.org/package=mlr3misc):\n Miscellaneous functions used in multiple mlr3 [extension\n packages](https://github.com/mlr-org/mlr3/wiki/Extension-Packages).\n Developed by the `mlr` team. No extra recursive dependencies.\n - [`paradox`](https://cran.r-project.org/package=paradox):\n Descriptions for parameters and parameter sets. Developed by the\n `mlr` team. No extra recursive dependencies.\n - [`R6`](https://cran.r-project.org/package=R6): Reference class\n objects. No recursive dependencies.\n - [`data.table`](https://cran.r-project.org/package=data.table):\n Extension of R’s `data.frame`. No recursive dependencies.\n - [`digest`](https://cran.r-project.org/package=digest): Hash\n digests. No recursive dependencies.\n - [`uuid`](https://cran.r-project.org/package=uuid): Create unique\n string identifiers. No recursive dependencies.\n - [`lgr`](https://cran.r-project.org/package=lgr): Logging\n facility. No extra recursive dependencies.\n - [`mlr3measures`](https://cran.r-project.org/package=mlr3measures):\n Performance measures. No extra recursive dependencies.\n - [`mlbench`](https://cran.r-project.org/package=mlbench): A\n collection of machine learning data sets. No dependencies.\n - [Reflections](https://en.wikipedia.org/wiki/Reflection_%28computer_programming%29):\n Objects are queryable for properties and capabilities, allowing you\n to program on them.\n - Additional functionality that comes with extra dependencies:\n - To capture output, warnings and exceptions,\n [`evaluate`](https://cran.r-project.org/package=evaluate) and\n [`callr`](https://cran.r-project.org/package=callr) can be used.\n\n## Extension Packages\n\n

\n\nConsult the\n[wiki](https://github.com/mlr-org/mlr3/wiki/Extension-Packages) for\nshort descriptions and links to the respective repositories.\n\n## Contributing to mlr3\n\nThis R package is licensed under the\n[LGPL-3](https://www.gnu.org/licenses/lgpl-3.0.en.html). If you\nencounter problems using this software (lack of documentation,\nmisleading or wrong documentation, unexpected behaviour, bugs, …) or\njust want to suggest features, please open an issue in the [issue\ntracker](https://github.com/mlr-org/mlr3/issues). Pull requests are\nwelcome and will be included at the discretion of the maintainers.\n\nPlease consult the [wiki](https://github.com/mlr-org/mlr3/wiki/) for a\n[style guide](https://github.com/mlr-org/mlr3/wiki/Style-Guide), a\n[roxygen guide](https://github.com/mlr-org/mlr3/wiki/Roxygen-Guide) and\na [pull request\nguide](https://github.com/mlr-org/mlr3/wiki/PR-Guidelines).\n\n## Citing mlr3\n\nIf you use mlr3, please cite our [JOSS\narticle](https://doi.org/10.21105/joss.01903):\n\n @Article{mlr3,\n title = {{mlr3}: A modern object-oriented machine learning framework in {R}},\n author = {Michel Lang and Martin Binder and Jakob Richter and Patrick Schratz and Florian Pfisterer and Stefan Coors and Quay Au and Giuseppe Casalicchio and Lars Kotthoff and Bernd Bischl},\n journal = {Journal of Open Source Software},\n year = {2019},\n month = {dec},\n doi = {10.21105/joss.01903},\n url = {https://joss.theoj.org/papers/10.21105/joss.01903},\n }","jobInfo":{"package":"mlr3","version":"0.3.0","parsingStatus":"success","parserVersion":1,"parsedAt":"2020-07-22T13:29:53+0000"}}

Description

Arguments

Dictionary

Meta Information

See Also