Patrick Schratz

Patrick Schratz

18 packages on CRAN

3 packages on GitHub

mlr

cran
99.99th

Percentile

Interface to a large number of classification and regression techniques, including machine-readable parameter descriptions. There is also an experimental extension for survival analysis, clustering and general, example-specific cost-sensitive learning. Generic resampling, including cross-validation, bootstrapping and subsampling. Hyperparameter tuning with modern optimization techniques, for single- and multi-objective problems. Filter and wrapper methods for feature selection. Extension of basic learners with additional operations common in machine learning, also allowing for easy nested resampling. Most operations can be parallelized.

99.99th

Percentile

Extends 'mlr3' with filter methods for feature selection. Besides standalone filter methods built-in methods of any machine-learning algorithm are supported. Partial scoring of multivariate filter methods is supported.

99.99th

Percentile

Extends the mlr3 ML framework with spatio-temporal resampling methods to account for the presence of spatiotemporal autocorrelation (STAC) in predictor variables. STAC may cause highly biased performance estimates in cross-validation if ignored.

oddsratio

cran
99.99th

Percentile

Simplified odds ratio calculation of GAM(M)s & GLM(M)s. Provides structured output (data frame) of all predictors and their corresponding odds ratios and confident intervals for further analyses. It helps to avoid false references of predictors and increments by specifying these parameters in a list instead of using 'exp(coef(model))' (standard approach of odds ratio calculation for GLMs) which just returns a plain numeric output. For GAM(M)s, odds ratio calculation is highly simplified with this package since it takes care of the multiple 'predict()' calls of the chosen predictor while holding other predictors constant. Also, this package allows odds ratio calculation of percentage steps across the whole predictor distribution range for GAM(M)s. In both cases, confident intervals are returned additionally. Calculated odds ratio of GAM(M)s can be inserted into the smooth function plot.

FSelector

cran
99.99th

Percentile

Functions for selecting attributes from a given dataset. Attribute subset selection is the process of identifying and removing as much of the irrelevant and redundant information as possible.

99.99th

Percentile

'Rcpp' (free of 'Java'/'Weka') implementation of 'FSelector' entropy-based feature selection algorithms based on an MDL discretization (Fayyad U. M., Irani K. B.: Multi-Interval Discretization of Continuous-Valued Attributes for Classification Learning. In 13'th International Joint Conference on Uncertainly in Artificial Intelligence (IJCAI93), pages 1022-1029, Chambery, France, 1993.) <https://www.ijcai.org/Proceedings/93-2/Papers/022.pdf> with a sparse matrix support.

containerit

github
99.99th

Percentile

Package R sessions, scripts, workspace directories, and R Markdown documents together with all dependencies to execute them in Docker containers. This package is supported by the project Opening Reproducible Research (<https://o2r.info>).

iml

cran
99.99th

Percentile

Interpretability methods to analyze the behavior and predictions of any machine learning model. Implemented methods are: Feature importance described by Fisher et al. (2018) <arXiv:1801.01489>, accumulated local effects plots described by Apley (2018) <arXiv:1612.08468>, partial dependence plots described by Friedman (2001) <www.jstor.org/stable/2699986>, individual conditional expectation ('ice') plots described by Goldstein et al. (2013) <doi:10.1080/10618600.2014.907095>, local models (variant of 'lime') described by Ribeiro et. al (2016) <arXiv:1602.04938>, the Shapley Value described by Strumbelj et. al (2014) <doi:10.1007/s10115-013-0679-x>, feature interactions described by Friedman et. al <doi:10.1214/07-AOAS148> and tree surrogate models.

mapview

cran
99.99th

Percentile

Quickly and conveniently create interactive visualisations of spatial data with or without background maps. Attributes of displayed features are fully queryable via pop-up windows. Additional functionality includes methods to visualise true- and false-color raster images and bounding boxes.

mlr3

cran
99.99th

Percentile

Efficient, object-oriented programming on the building blocks of machine learning. Provides 'R6' objects for tasks, learners, resamplings, and measures. The package is geared towards scalability and larger datasets by supporting parallelization and out-of-memory data-backends like databases. While 'mlr3' focuses on the core computational operations, add-on packages provide additional functionality.

99.99th

Percentile

Implements methods for feature selection with 'mlr3', e.g. random search and sequential selection. Various termination criteria can be set and combined. The class 'AutoFSelector' provides a convenient way to perform nested resampling in combination with 'mlr3'.

99.99th

Percentile

Recommended Learners for 'mlr3'. Extends 'mlr3' and 'mlr3proba' with interfaces to essential machine learning packages on CRAN. This includes, but is not limited to: (penalized) linear and logistic regression, linear and quadratic discriminant analysis, k-nearest neighbors, naive Bayes, support vector machines, and gradient boosting.

mlr3misc

cran
99.99th

Percentile

Frequently used helper functions and assertions used in 'mlr3' and its companion packages. Comes with helper functions for functional programming, for printing, to work with 'data.table', as well as some generally useful 'R6' classes. This package also supersedes the package 'BBmisc'.

mlr3verse

cran
99.99th

Percentile

The 'mlr3' package family is a set of packages for machine-learning purposes built in a modular fashion. This wrapper package is aimed to simplify the installation and loading of the core 'mlr3' packages. Get more information about the 'mlr3' project at <https://mlr3book.mlr-org.com/>.

mlr3viz

cran
99.99th

Percentile

Provides visualizations for 'mlr3' objects such as tasks, predictions, resample results or benchmark results via the autoplot() generic of 'ggplot2'. The returned 'ggplot' objects are intended to provide sensible defaults, yet can easily be customized to create camera-ready figures. Visualizations include barplots, boxplots, histograms, ROC curves, and Precision-Recall curves.

99.99th

Percentile

Unified parallelization framework for multiple back-end, designed for internal package and interactive usage. The main operation is parallel mapping over lists. Supports 'local', 'multicore', 'mpi' and 'BatchJobs' mode. Allows tagging of the parallel operation with a level name that can be later selected by the user to switch on parallel execution for exactly this operation.

RSAGA

cran
99.99th

Percentile

Provides access to geocomputing and terrain analysis functions of the geographical information system (GIS) 'SAGA' (System for Automated Geoscientific Analyses) from within R by running the command line version of SAGA. This package furthermore provides several R functions for handling ASCII grids, including a flexible framework for applying local functions (including predict methods of fitted models) and focal functions to multiple grids. SAGA GIS is available under GPLv2 / LGPLv2 licence from <http://sourceforge.net/projects/saga-gis/>.

skimr

cran
99.99th

Percentile

A simple to use summary function that can be used buttskyle96@gmail.comwith pipes and displays nicely in the console. The default summary statistics may be modified by the user as can the default formatting. Support for data frames and vectors is included, and users can implement their own skim methods for specific object types as described in a vignette. Default summaries include support for inline spark graphs. Instructions for managing these on specific operating systems are given in the "Using skimr" vignette and the README.

sperrorest

cran
99.99th

Percentile

Implements spatial error estimation and permutation-based variable importance measures for predictive models using spatial cross-validation and spatial block bootstrap.

RQGIS

github
99.99th

Percentile

Establishes an interface between R and 'QGIS', i.e. it allows the user to access 'QGIS' functionalities from the R console. It achieves this by using the 'QGIS' Python API via the command line. Hence, RQGIS extends R's statistical power by the incredible vast geo-functionality of 'QGIS' (including also 'GDAL', 'SAGA'- and 'GRASS'-GIS among other third-party providers). This in turn creates a powerful environment for advanced and innovative (geo-)statistical geocomputing. 'QGIS' is licensed under GPL version 2 or greater and is available from <http://www.qgis.org/en/site/>.

tic

github
99.99th

Percentile

Provides a way to describe common build and deployment workflows for R-based projects: packages, websites (e.g. blogdown, pkgdown), or data processing (e.g. research compendia). The recipe is described independent of the continuous integration tool used for processing the workflow (e.g. 'Travis CI' or 'AppVeyor'). This package has been peer-reviewed by rOpenSci (v. 0.3.0.9004).