The cIRT package (CRAN, GitHub) is home to the implementation of Choice Item Response Theory (cIRT) described by Culpepper and Balamuta (2015). cIRT jointly models the accuracy of cognitive responses and item choices within a bayesian hierarchical framework.
We revisited the cIRT package and applied a series of modernizations alongside of a requested bug fix. In particular, we have updated all package dependencies and now depend on Rcpp to generate the registration for C++ functions we export with it.
Read More…
The RcppEnsmallen package brings to R a Header-Only C++ Mathematical Optimization Library for Armadillo. In particular, Ensmallen is a templated C++ mathematical optimization library (by the MLPACK team) that provides a simple set of abstractions for writing an objective function to optimize.
The development pace of the Ensmallen Development Team hasn’t stopped! In fact, within the last two weeks they issued another update. This update brought in a new SPSA optimization routine.
Read More…
The objective of sitmo (CRAN, GitHub) is to provide the means for performing parallel draws from a Parallel Pseudo Random Number Generator (PPRNG). To that end, the package is largely a bridge into the PPRNGs developed by the sitmo consultancy agency.
For this point-release, sitmo addresses issues that arose on CRAN. Specifically, we fix the use of parallelization flags in the Makevars{.win} files. Moreover, we address a minor compiler warning associated with a missing new line at the end of a file.
Read More…
The RcppEnsmallen package brings to R a Header-Only C++ Mathematical Optimization Library for Armadillo. In particular, Ensmallen is a templated C++ mathematical optimization library (by the MLPACK team) that provides a simple set of abstractions for writing an objective function to optimize.
Within this release, the Ensmallen Development Team added in many new optimization routines: FTML, SWATS, Padam, and Eve. In addition, the API changed for SGD-like optimizers to include the ability to reset the the parameters each time an optimization is called.
Read More…
The RcppEnsmallen package brings to R a Header-Only C++ Mathematical Optimization Library for Armadillo. In particular, Ensmallen is a templated C++ mathematical optimization library (by the MLPACK team) that provides a simple set of abstractions for writing an objective function to optimize.
First off, congratulations to the Ensmallen Development Team on their acceptance into Workshop on Systems for ML and Open Source Software at NIPS / NeurIPS held 7 December 2018 in Montréal, Canada.
Read More…
The RcppEnsmallen package brings to R a Header-Only C++ Mathematical Optimization Library for Armadillo. Previous packages, like RcppNumerical, enabled optimization routines for Eigen when combined with RcppEigen. In particular, Ensmallen is a templated C++ mathematical optimization library (by the MLPACK team) that provides a simple set of abstractions for writing an objective function to optimize. Provided within are various standard and cutting-edge optimizers that include full-batch gradient descent techniques, small-batch techniques, gradient-free optimizers, and constrained optimization.
Read More…
In this update, the uiucthemes package now has a gallery vignette that shows the sample skeleton of each theme supported by the package. Pictures of supported themes can be found next.
In addition, the package is now home to a Beamer template for the Market Information Lab (MIL) @ UIUC. The template was contributed by Jose Luis Rodriguez. If you are interested in contributing to the uiucthemes package, please feel free to open a PR at the project’s repository: https://github.
Read More…
The cetcolor package (CRAN, GitHub) is designed to bring to R the 56 colour maps created by Peter Kovesi, a part-time research fellow with the Geophysics and Image Analysis group within the Centre for Exploration Targeting in the School of Earth and Environment at The University of Western Australia. These colour maps are special as they avoid points of locally high colour contrast, which leads to the perception of false anomalies in your data when there are none.
Read More…
The searcher package (CRAN, GitHub) provides a search interface to look up terms on popular websites such as ‘Google’, ‘Bing’, ‘DuckDuckGo’, ‘ixquick’, ‘StackOverflow’, ‘GitHub’, and ‘BitBucket’. Upon calling a search_*() function, a browser window will open with the search results for the query.
This release brings API parity by ensuring all search functions have the rlang flag turned on by default to guarantee R-specific results. Previously, the main search engine functions lacked the rlang flag and could potential retrieve odd search results.
Read More…
The objective of sitmo (CRAN, GitHub) is to provide the means for performing parallel draws from a Parallel Pseudo Random Number Generator (PPRNG). To that end, the package is largely a bridge into the PPRNGs developed by the sitmo consultancy agency.
Within this release, sitmo now contains the Threefry engine (Salmon et al., 2011 doi:10.1145/2063384.2063405) and Vandercorput engine (van der Corput, 1935 an:0012.34705). We’ve also added a contributor Ralf Stubner who provided a patch to ensure the C++11 compatibility of the Threefry engine.
Read More…