Transport theory has seen much success in many fields of statistics and machine learning. We provide a variety of algorithms to compute Wasserstein distance, barycenter, and others. See Peyré and Cuturi (2019) <doi:10.1561/2200000073> for the general exposition to the study of computational optimal transport.

This package provides R Markdown output formats to use Tufte styles for PDF and HTML output.

The goal of this package will be to provide a simple interface for automatic machine learning that fits the tidymodels framework. The intention is to work for regression and classification problems with a simple verb framework.

Compute arbitrary non-parametric bootstrap statistics on data in tidy data frames.

This package provides tools for computing various vector summaries of persistence diagrams studied in Topological Data Analysis. For improved computational efficiency, all code for the vector summaries is written in C++ using the Rcpp and RcppArmadillo packages.

Data frames with time information are subset and flagged with period information. Data frames with times are dealt as timeDF objects and periods are represented as periodDF objects.

The strength of evidence provided by epidemiological and observational studies is inherently limited by the potential for unmeasured confounding. We focus on three key quantities: the observed bound of the confidence interval closest to the null, the relationship between an unmeasured confounder and the outcome, for example a plausible residual effect size for an unmeasured continuous or binary confounder, and the relationship between an unmeasured confounder and the exposure, for example a realistic mean difference or prevalence difference for this hypothetical confounder between exposure groups. Building on the methods put forth by Cornfield et al. (1959), Bross (1966), Schlesselman (1978), Rosenbaum & Rubin (1983), Lin et al. (1998), Lash et al. (2009), Rosenbaum (1986), Cinelli & Hazlett (2020), VanderWeele & Ding (2017), and Ding & VanderWeele (2016), we can use these quantities to assess how an unmeasured confounder may tip our result to insignificance.

Our method introduces mathematically well-defined measures for tightness of branches in a hierarchical tree. Statistical significance of the findings is determined, for all branches of the tree, by performing permutation tests, optionally with generalized Pareto p-value estimation.

This package implements the truncated harmonic mean estimator (THAMES) of the reciprocal marginal likelihood using posterior samples and unnormalized log posterior values via reciprocal importance sampling. Metodiev, Perrot-Dockès, Ouadah, Irons, Latouche, & Raftery (2024). Bayesian Analysis. <doi:10.1214/24-BA1422>.

Trelliscope is a scalable, flexible, interactive approach to visualizing data (Hafen, 2013 <doi:10.1109/LDAV.2013.6675164>). This package provides methods that make it easy to create a Trelliscope display specification for TrelliscopeJS. High-level functions are provided for creating displays from within tidyverse or ggplot2 workflows. Low-level functions are also provided for creating new interfaces.

Here we provide tools for the computation and factorization of high-dimensional tensor products that are formed by smaller matrices. The methods are based on properties of Kronecker products (Searle 1982, p. 265, ISBN-10: 0470009616). We evaluated this methodology by benchmark testing and illustrated its use in Gaussian Linear Models ('Lopez-Cruz et al., 2024') <doi:10.1093/g3journal/jkae001>.

Builds tables with customizable rows. Users can specify the type of data to use for each row, as well as how to handle missing data and the types of comparison tests to run on the table columns.

Uses indicator species scores across binary partitions of a sample set to detect congruence in taxon-specific changes of abundance and occurrence frequency along an environmental gradient as evidence of an ecological community threshold. Relevant references include Baker and King (2010) <doi:10.1111/j.2041-210X.2009.00007.x>, King and Baker (2010) <doi:10.1899/09-144.1>, and Baker and King (2013) <doi:10.1899/12-142.1>.

Formula-based user-interfaces to specific transformation models implemented in package mlt (<DOI:10.32614/CRAN.package.mlt>, <DOI:10.32614/CRAN.package.mlt.docreg>). Available models include Cox models, some parametric survival models (Weibull, etc.), models for ordered categorical variables, normal and non-normal (Box-Cox type) linear models, and continuous outcome logistic regression (Lohse et al., 2017, <DOI:10.12688/f1000research.12934.1>). The underlying theory is described in Hothorn et al. (2018) <DOI:10.1111/sjos.12291>. An extension to transformation models for clustered data is provided (Barbanti and Hothorn, 2022, <DOI:10.1093/biostatistics/kxac048>) and a tutorial explains applications in survival analysis (Siegfried et al., 2025, <DOI:10.48550/arXiv.2402.06428>). Multivariate conditional transformation models (Klein et al, 2022, <DOI:10.1111/sjos.12501>) and shift-scale transformation models (Siegfried et al, 2023, <DOI:10.1080/00031305.2023.2203177>) can be fitted as well. The package contains an implementation of a doubly robust score test, described in Kook et al. (2024, <DOI:10.1080/01621459.2024.2395588>).

Tensor-train is a compact representation for higher-order tensors. Some algorithms for performing tensor-train decomposition are available such as TT-SVD, TT-WOPT, and TT-Cross. For the details of the algorithms, see I. V. Oseledets (2011) <doi:10.1137/090752286>, Yuan Longao, et al (2017) <doi:10.48550/arXiv.1709.02641>, I. V. Oseledets (2010) <doi:10.1016/j.laa.2009.07.024>.

Detection of outliers in time series following the Chen and Liu (1993) <DOI:10.2307/2290724> procedure. Innovational outliers, additive outliers, level shifts, temporary changes and seasonal level shifts are considered.

Converting text to numerical features requires specifically created procedures, which are implemented as steps according to the recipes package. These steps allows for tokenization, filtering, counting (tf and tfidf) and feature hashing.

Tidy standardized mean differences ('SMDs'). tidysmd uses the smd package to calculate standardized mean differences for variables in a data frame, returning the results in a tidy format.

This package provides a toolbox for comparing two data frames. This package is defunct. I recommend you use the "versus" package instead.

Bindings for the Tabula <https://tabula.technology/> Java library, which can extract tables from PDF files. This tool can reduce time and effort in data extraction processes in fields like investigative journalism. It allows for automatic and manual table extraction, the latter facilitated through a Shiny interface, enabling manual areas selection\ with a computer mouse for data retrieval.

This package implements the Temporal Forest algorithm for feature selection in high-dimensional longitudinal data. The method combines time-aware network construction via weighted gene co-expression network analysis (WGCNA), module-based feature screening, and stability selection using tree-based models. This package provides tools for reproducible longitudinal analysis, closely following the methodology described in Shao, Moore, and Ramirez (2025) <https://github.com/SisiShao/TemporalForest>.

Generates stochastic time series and genealogies associated with a population dynamics model. Times series are simulated using the Gillespie exact and approximate algorithms and a new algorithm we introduce that uses both approaches to optimize the time execution of the simulations. Genealogies are simulated from a trajectory using a backwards-in-time based approach. Methods are described in Danesh G et al (2022) <doi:10.1111/2041-210X.14038>.

Sometimes you need to split your data and work on the two chunks independently before bringing them back together. Taber allows you to do that with its two functions.

The textrank algorithm is an extension of the Pagerank algorithm for text. The algorithm allows to summarize text by calculating how sentences are related to one another. This is done by looking at overlapping terminology used in sentences in order to set up links between sentences. The resulting sentence network is next plugged into the Pagerank algorithm which identifies the most important sentences in your text and ranks them. In a similar way textrank can also be used to extract keywords. A word network is constructed by looking if words are following one another. On top of that network the Pagerank algorithm is applied to extract relevant words after which relevant words which are following one another are combined to get keywords. More information can be found in the paper from Mihalcea, Rada & Tarau, Paul (2004) <https://www.aclweb.org/anthology/W04-3252/>.