TensorFlow Hub is a library for the publication, discovery, and consumption of reusable parts of machine learning models. A module is a self-contained piece of a TensorFlow graph, along with its weights and assets, that can be reused across different tasks in a process known as transfer learning. Transfer learning train a model with a smaller dataset, improve generalization, and speed up training.

Easily construct prompts and associated logic for interacting with large language models (LLMs). tidyprompt introduces the concept of prompt wraps, which are building blocks that you can use to quickly turn a simple prompt into a complex one. Prompt wraps do not just modify the prompt text, but also add extraction and validation functions that will be applied to the response of the LLM. This ensures that the user gets the desired output. tidyprompt can add various features to prompts and their evaluation by LLMs, such as structured output, automatic feedback, retries, reasoning modes, autonomous R function calling, and R code generation and evaluation. It is designed to be compatible with any LLM provider that offers chat completion.

An interface to twitter-text', a JavaScript library which is responsible for determining the length/validity of a tweet and identifying/linking any URLs or special tags (e.g. mentions or hashtags) which may be present.

Computes a point pattern in R^2 or on a graph that is representative of a collection of many data patterns. The result is an approximate barycenter (also known as Fréchet mean or prototype) based on a transport-transform metric. Possible choices include Optimal SubPattern Assignment (OSPA) and Spike Time metrics. Details can be found in Müller, Schuhmacher and Mateu (2020) <doi:10.1007/s11222-020-09932-y>.

This package provides a tbl_ts class (the tsibble') for temporal data in an data- and model-oriented format. The tsibble provides tools to easily manipulate and analyse temporal data, such as filling in time gaps and aggregating over calendar periods.

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

Collection of functions that allow to export data frames to excel workbook.

Unobserved components time series model using the linear innovations state space representation (single source of error) with choice of error distributions and option for dynamic variance. Methods for estimation using automatic differentiation, automatic model selection and ensembling, prediction, filtering, simulation and backtesting. Based on the model described in Hyndman et al (2012) <doi:10.1198/jasa.2011.tm09771>.

Simulate genotypes for case-parent triads, case-control, and quantitative trait samples with realistic linkage diequilibrium structure and allele frequency distribution. For studies of epistasis one can simulate models that involve specific SNPs at specific sets of loci, which we will refer to as "pathways". TriadSim generates genotype data by resampling triad genotypes from existing data. The details of the method is described in the manuscript under preparation "Simulating Autosomal Genotypes with Realistic Linkage Disequilibrium and a Spiked in Genetic Effect" Shi, M., Umbach, D.M., Wise A.S., Weinberg, C.R.

Uses the Distorted Wave Born Approximation (DWBA) to compute the acoustic backward scattering, the geometry of the object is formed by a volumetric mesh, composed of tetrahedrons. This computation is done efficiently through an analytical 3D integration that allows for a solution which is expressed in terms of elementary functions for each tetrahedron. It is important to note that this method is only valid for objects whose acoustic properties, such as density and sound speed, do not vary significantly compared to the surrounding medium. (See Lavia, Cascallares and Gonzalez, J. D. (2023). TetraScatt model: Born approximation for the estimation of acoustic dispersion of fluid-like objects of arbitrary geometries. arXiv preprint <arXiv:2312.16721>).

This package provides tools for Topological Data Analysis. The package focuses on statistical analysis of persistent homology and density clustering. For that, this package provides an R interface for the efficient algorithms of the C++ libraries GUDHI <https://project.inria.fr/gudhi/software/>, Dionysus <https://www.mrzv.org/software/dionysus/>, and PHAT <https://bitbucket.org/phat-code/phat/>. This package also implements methods from Fasy et al. (2014) <doi:10.1214/14-AOS1252> and Chazal et al. (2015) <doi:10.20382/jocg.v6i2a8> for analyzing the statistical significance of persistent homology features.

Read General Transit Feed Specification (GTFS) zipfiles into a list of R dataframes. Perform validation of the data structure against the specification. Analyze the headways and frequencies at routes and stops. Create maps and perform spatial analysis on the routes and stops. Please see the GTFS documentation here for more detail: <https://gtfs.org/>.

Reconstructs animal tracks from magnetometer, accelerometer, depth and optional speed data. Designed primarily using data from Wildlife Computers Daily Diary tags deployed on northern fur seals.

Simplifies access to Tunisian government open data from <https://data.gov.tn/fr/>. Queries datasets by theme, author, or keywords, retrieves metadata, and gets structured results ready for analysis; all through the official CKAN API.

The Taylor Russell model is a widely used method for assessing test validity in personnel selection tasks. The three functions in this package extend this model in a number of notable ways. TR() estimates test validity for a single selection test via the original Taylor Russell model. It extends this model by allowing users greater flexibility in argument choice. For example, users can specify any three of the four parameters (base rate, selection ratio, criterion validity, and positive predictive value) of the Taylor Russell model and estimate the remaining parameter (see the help file for examples). The TaylorRussell() function generalizes the original Taylor Russell model to allow for multiple selection tests (predictors). To our knowledge, this is the first generalization of the Taylor Russell model to allow for three or more selection tests (it is also the first to correctly handle models with two selection tests). TRDemo() is a shiny program for illustrating the underlying logic of the Taylor Russell model. Taylor, HC and Russell, JT (1939) "The relationship of validity coefficients to the practical effectiveness of tests in selection: Discussion and tables" <doi:10.1037/h0057079>.

Download daily interest rates from the US Treasury XML feed. Leveraging <https://home.treasury.gov/treasury-daily-interest-rate-xml-feed>, this package serves as a wrapper, facilitating the retrieval of daily treasury rates across various categories, including par yield curves, treasury bills, long-term rates, and real yield curves.

The goal of TailID is to detect sensitive points in the tail of a dataset using techniques from Extreme Value Theory (EVT). It utilizes the Generalized Pareto Distribution (GPD) for assessing tail behavior and detecting inconsistent points with the Identical Distribution hypothesis of the tail. For more details see Manau (2025)<doi:10.4230/LIPIcs.ECRTS.2025.20>.

Compute age-adjusted rates by direct and indirect methods and other epidemiological indicators in a tidy way, wrapping functions from the epitools package.

Integrates several popular high-dimensional methods based on Linear Discriminant Analysis (LDA) and provides a comprehensive and user-friendly toolbox for linear, semi-parametric and tensor-variate classification as mentioned in Yuqing Pan, Qing Mai and Xin Zhang (2019) <arXiv:1904.03469>. Functions are included for covariate adjustment, model fitting, cross validation and prediction.

Efficient method for fitting nonparametric matrix trace regression model. The detailed description can be found in C. Lee, L. Li, H. Zhang, and M. Wang (2021). Nonparametric Trace Regression via Sign Series Representation. <arXiv:2105.01783>. The method employs the aggregation of structured sign series for trace regression (ASSIST) algorithm.

Factor and autoregressive models for matrix and tensor valued time series. We provide functions for estimation, simulation and prediction. The models are discussed in Li et al (2021) <doi:10.48550/arXiv.2110.00928>, Chen et al (2020) <DOI:10.1080/01621459.2021.1912757>, Chen et al (2020) <DOI:10.1016/j.jeconom.2020.07.015>, and Xiao et al (2020) <doi:10.48550/arXiv.2006.02611>.

Token-Oriented Object Notation (TOON) is a compact, human-readable serialization format designed for passing structured data to Large Language Models with significantly reduced token usage. It's intended for LLM input as a lossless, drop-in representation of JSON data.

Create rich and fully interactive timeline visualizations. Timelines can be included in Shiny apps or R markdown documents. timevis includes an extensive API to manipulate a timeline after creation, and supports getting data out of the visualization into R. Based on the vis.js Timeline JavaScript library.

Return the first four moments, estimation of parameters and sample of the TSMSN distributions (Skew Normal, Skew t, Skew Slash or Skew Contaminated Normal).