Microsoft Corporation

Microsoft Corporation

9 packages on CRAN

doParallel

cran
99.99th

Percentile

Provides a parallel backend for the %dopar% function using the parallel package.

doSNOW

cran
99.99th

Percentile

Provides a parallel backend for the %dopar% function using the snow package of Tierney, Rossini, Li, and Sevcikova.

interpret

cran
99.99th

Percentile

Package for training interpretable machine learning models. Historically, the most interpretable machine learning models were not very accurate, and the most accurate models were not very interpretable. Microsoft Research has developed an algorithm called the Explainable Boosting Machine (EBM) which has both high accuracy and interpretable characteristics. EBM uses machine learning techniques like bagging and boosting to breathe new life into traditional GAMs (Generalized Additive Models). This makes them as accurate as random forests and gradient boosted trees, and also enhances their intelligibility and editability. Details on the EBM algorithm can be found in the paper by Rich Caruana, Yin Lou, Johannes Gehrke, Paul Koch, Marc Sturm, and Noemie Elhadad (2015, <doi:10.1145/2783258.2788613>).

lightgbm

cran
99.99th

Percentile

Tree based algorithms can be improved by introducing boosting frameworks. 'LightGBM' is one such framework, based on Ke, Guolin et al. (2017) <https://papers.nips.cc/paper/6907-lightgbm-a-highly-efficient-gradient-boosting-decision>. This package offers an R interface to work with it. It is designed to be distributed and efficient with the following advantages: 1. Faster training speed and higher efficiency. 2. Lower memory usage. 3. Better accuracy. 4. Parallel learning supported. 5. Capable of handling large-scale data. In recognition of these advantages, 'LightGBM' has been widely-used in many winning solutions of machine learning competitions. Comparison experiments on public datasets suggest that 'LightGBM' can outperform existing boosting frameworks on both efficiency and accuracy, with significantly lower memory consumption. In addition, parallel experiments suggest that in certain circumstances, 'LightGBM' can achieve a linear speed-up in training time by using multiple machines.

lmSubsets

cran
99.99th

Percentile

Exact and approximation algorithms for variable-subset selection in ordinary linear regression models. Either compute all submodels with the lowest residual sum of squares, or determine the single-best submodel according to a pre-determined statistical criterion. Hofmann et al. (2020) <10.18637/jss.v093.i03>.

miniCRAN

cran
99.99th

Percentile

Makes it possible to create an internally consistent repository consisting of selected packages from CRAN-like repositories. The user specifies a set of desired packages, and 'miniCRAN' recursively reads the dependency tree for these packages, then downloads only this subset. The user can then install packages from this repository directly, rather than from CRAN. This is useful in production settings, e.g. server behind a firewall, or remote locations with slow (or zero) Internet access.

monaco

cran
99.99th

Percentile

A HTML widget rendering the 'Monaco' editor. The 'Monaco' editor is the code editor which powers 'VS Code'. It is particularly well developed for 'JavaScript'. In addition to the built-in features of the 'Monaco' editor, the widget allows to prettify multiple languages, to view the 'HTML' rendering of 'Markdown' code, and to view and resize 'SVG' images.

onnx

cran
99.99th

Percentile

R Interface to 'ONNX' - Open Neural Network Exchange <https://onnx.ai/>. 'ONNX' provides an open source format for machine learning models. It defines an extensible computation graph model, as well as definitions of built-in operators and standard data types.

99.99th

Percentile

A 'Shiny' app including the 'Monaco' editor. The 'Monaco' editor is the code editor which powers 'VS Code'. It is particularly well developed for 'JavaScript'. In addition to the 'Monaco' editor features, the app provides prettifiers and minifiers for multiple languages, 'SCSS' and 'TypeScript' compilers, code checking for 'C' and 'C++' (requires 'cppcheck').