One hundred and fifty-three new packages made it to CRAN in January. Here are my “Top 40” picks in eight categories: Computational Methods, Data, Machine Learning, Medicine, Science, Statistics, Utilities, and Visualization.
ecmwfr v1.1.0: Provides a programmatic interface to the European Centre for Medium-Range Weather Forecasts’ public and restricted dataset web services ECMWF, as well as Copernicus’s Climate Data Store CDS, allowing users to download weather forecasts and climate data. There are vignettes for both CDS and ECMWFR.
nhdR v0.5.1: Provides tools for working with the National Hydrography Dataset, with functions for querying, downloading, and networking both the NHD and NHDPlus datasets. There are vignettes for Creating Simple Maps and Quering Flow Information, as well as an example.
analysisPipelines v1.0.0: Implements an R interface that enables data scientists to compose inter-operable pipelines between R, Spark, and Python for data manipulation, exploratory analysis, modeling, and reporting. There are vignettes for Python Functions, R data frames, Spark data frames, Interoperable Pipelines, Meta-pipelines, Streaming Analysis Pipelines, and Using Pipelines with Spark.
foto v1.0.0: Implements the Fourier Transform Textural Ordination method, which uses a principal component analysis on radially averaged, two-dimensional Fourier spectra to characterize image texture. See the vignette for details.
ruimtehol v0.1.2: Wraps the StarSpace library, allowing users to calculate word, sentence, article, document, webpage, link, and entity embeddings. The techniques are explained in detail in Wu et al. (2017). See the vignette for more information.
zoomgrid v1.0.0: Implements a grid search algorithm with zoom to help solve difficult optimization problems where there are many local optima inside the domain of the target function. Look here for information.
bayesCT v0.99.0: Provides functions to simulate and analyze Bayesian adaptive clinical trials, incorporating historical data and allowing for early stopping. There is an Introduction, and vignettes for Binomial and Normal outcomes.
BioMedR v1.1.1: Provides tools for calculating 293 chemical descriptors and 14 kinds of chemical fingerprints, 9920 protein descriptors based on protein sequences, more than 6000 DNA/RNA descriptors from nucleotide sequences, and six types of interaction descriptors. There is a very informative vignette.
dr4pl v1.1.8: Models the relationship between dose levels and responses in a pharmacological experiment using the 4 Parameter Logistic model, and provides bounds that prevent parameter estimates from diverging. See Gadagkar and Call (2015) and Ritz et al. (2015) for background information, and the vignette for examples.
GMMAT v1.0.3: Provides functions to perform association tests using generalized linear mixed models (GLMMs) in genome-wide association studies (GWAS) and sequencing association studies. See Chen et al. (2016) and Chen et al. (2019) for background information, and the vignette for an introduction to the package.
ethnobotanyR v0.1.4: Implements functions to calculate common quantitative ethnobotany indices to assess the cultural significance. See Tardio and Pardo-de-Santayana (2008) for background information, and the vignette for information on the package.
apcf v0.1.2: Implements the adapted pair correlation function, which transfers the concept of the pair correlation function from point patterns to patterns of objects of finite size and irregular shape. This is a re-implementation of the method suggested by Nuske et al. (2009). See the vignette for details.
concurve v1.0.1: Provides functions to compute confidence (compatibility/consonance) intervals for various statistical tests, along with their corresponding P-values and S-values. Consonance functions allow modelers to determine what effect sizes are compatible with the test model at various compatibility levels. For details, see Poole (1987), Schweder and Hjort (2002), Singh, Xie, and Strawderman (2007), and Amrhein, Trafimow and Greenland (2018). See the vignette for examples.
IMaGES v0.1: Provides functions to implement Independent Multiple-sample Greedy Equivalence Search (IMaGES), a causal inference algorithm for creating aggregate graphs and structural equation modeling data for one or more datasets. See Ramsey et. al (2010) for background information. There is a vignette.
mimi v0.1.0: Implements functions to estimate main effects and interactions in mixed data sets with missing values. Estimation is done through a convex program where main effects are assumed sparse and the interactions low-rank. See Geneviève et al. (2018).
pcLasso v1.1: Implements a method for fitting the entire regularization path of the principal components lasso for linear and logistic regression models. See Tay, Friedman, and Tibshirani (2014) for details and the vignette for an Introduction.
qrandom v1.1: Implements an API to the ANU Quantum Random Number Generator, provided by the Australian National University, that generates true random numbers in real-time by measuring the quantum fluctuations of the vacuum. The quantum Random Number Generator is based on the papers by Symul et al. (2011) and Haw, et al. (2015). Look here for live random numbers.
surveysd v1.0.0: Provides functions to calculate point estimates and their standard errors in complex household surveys using bootstrap replicates. A comprehensive description of the methodology can be found here.
logger v0.1: Provides a flexible and extensible way of formatting and delivering log messages with low overhead. There is an Introduction and vignettes on The Anatomy of a Log Request, Format Customization, Migration, Benchmarks, Logging, and Extensions.