This package provides a Text mining toolkit for Chinese, which includes facilities for Chinese string processing, Chinese NLP supporting, encoding detecting and converting. Moreover, it provides some functions to support tm package in Chinese.

This package provides a crawler for programmatically navigating THREDDS Data Server (<https://www.unidata.ucar.edu/software/tds/>) catalogs, and access dataset metadata and resources.

Compositional data consisting of three-parts can be color mapped with a ternary color scale. Such a scale is provided by the tricolore packages with options for discrete and continuous colors, mean-centering and scaling. See Jonas Schöley (2021) "The centered ternary balance scheme. A technique to visualize surfaces of unbalanced three-part compositions" <doi:10.4054/DemRes.2021.44.19>, Jonas Schöley, Frans Willekens (2017) "Visualizing compositional data on the Lexis surface" <doi:10.4054/DemRes.2017.36.21>, and Ilya Kashnitsky, Jonas Schöley (2018) "Regional population structures at a glance" <doi:10.1016/S0140-6736(18)31194-2>.

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>.

This package contains some auxiliary functions.

Support functions and datasets to facilitate the analysis of linguistic data. The current focus is on the calculation of corpus-linguistic dispersion measures as described in Gries (2021) <doi:10.1007/978-3-030-46216-1_5> and Soenning (2025) <doi:10.3366/cor.2025.0326>. The most commonly used parts-based indices are implemented, including different formulas and modifications that are found in the literature, with the additional option to obtain frequency-adjusted scores. Dispersion scores can be computed based on individual count variables or a term-document matrix.

The ToxCast Data Analysis Pipeline ('tcpl') is an R package that manages, curve-fits, plots, and stores ToxCast data to populate its linked MySQL database, invitrodb'. The package was developed for the chemical screening data curated by the US EPA's Toxicity Forecaster (ToxCast) program, but tcpl can be used to support diverse chemical screening efforts.

This package implements a decomposition of the two-way fixed effects instrumental variable estimator into all possible Wald difference-in-differences estimators. Provides functions to summarize the contribution of different cohort comparisons to the overall two-way fixed effects instrumental variable estimate, with or without controls. The method is described in Miyaji (2024) <doi:10.48550/arXiv.2405.16467>.

It performs the smoothing approach provided by penalized least squares for univariate and bivariate time series, as proposed by Guerrero (2007) and Gerrero et al. (2017). This allows to estimate the time series trend by controlling the amount of resulting (joint) smoothness. --- Guerrero, V.M (2007) <DOI:10.1016/j.spl.2007.03.006>. Guerrero, V.M; Islas-Camargo, A. and Ramirez-Ramirez, L.L. (2017) <DOI:10.1080/03610926.2015.1133826>.

Data collected on movement behavior is often in the form of time- stamped latitude/longitude coordinates sampled from the underlying movement behavior. These data can be compressed into a set of segments via the Top- Down Time Ratio Segmentation method described in Meratnia and de By (2004) <doi:10.1007/978-3-540-24741-8_44> which, with some loss of information, can both reduce the size of the data as well as provide corrective smoothing mechanisms to help reduce the impact of measurement error. This is an improvement on the well-known Douglas-Peucker algorithm for segmentation that operates not on the basis of perpendicular distances. Top-Down Time Ratio segmentation allows for disparate sampling time intervals by calculating the distance between locations and segments with respect to time. Provided a trajectory with timestamps, tdtr() returns a set of straight- line segments that can represent the full trajectory. McCool, Lugtig, and Schouten (2022) <doi:10.1007/s11116-022-10328-2> describe this method as implemented here in more detail.

Propensity score matching for non-binary treatments.

Interface to TensorFlow Probability', a Python library built on TensorFlow that makes it easy to combine probabilistic models and deep learning on modern hardware ('TPU', GPU'). TensorFlow Probability includes a wide selection of probability distributions and bijectors, probabilistic layers, variational inference, Markov chain Monte Carlo, and optimizers such as Nelder-Mead, BFGS, and SGLD.

This package provides a teal_data class as a unified data model for teal applications focusing on reproducibility and relational data.

Includes: (i) tests and visualisations that can help the modeller explore time series components and perform decomposition; (ii) modelling shortcuts, such as functions to construct lagmatrices and seasonal dummy variables of various forms; (iii) an implementation of the Theta method; (iv) tools to facilitate the design of the forecasting process, such as ABC-XYZ analyses; and (v) "quality of life" functions, such as treating time series for trailing and leading values.

This package creates a framework to store and apply display metadata to Analysis Results Datasets (ARDs). The use of tfrmt allows users to define table format and styling without the data, and later apply the format to the data.

We focus on the diagnostic ability assessment of medical tests when the outcome of interest is the status (alive or dead) of the subjects at a certain time-point t. This binary status is determined by right-censored times to event and it is unknown for those subjects censored before t. Here we provide three methods (unknown status exclusion, imputation of censored times and using time-dependent ROC curves) to evaluate the diagnostic ability of binary and continuous tests in this context. Two references for the methods used here are Skaltsa et al. (2010) <doi:10.1002/bimj.200900294> and Heagerty et al. (2000) <doi:10.1111/j.0006-341x.2000.00337.x>.

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 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>.

Calculates the number of true positives and false positives from a dataset formatted for Jackknife alternative free-response receiver operating characteristic which is used for statistical analysis which is explained in the book Chakraborty DP (2017), "Observer Performance Methods for Diagnostic Imaging - Foundations, Modeling, and Applications with R-Based Examples", Taylor-Francis <https://www.crcpress.com/9781482214840>.

Computation of effects under linear, logistic and Poisson regression models with transformed variables. Logarithm and power transformations are allowed. Effects can be displayed both numerically and graphically in both the original and the transformed space of the variables. The methods are described in Barrera-Gomez and Basagana (2015) <doi:10.1097/EDE.0000000000000247>.

Travel Time API <https://docs.traveltime.com/api/overview/introduction> helps users find locations by journey time rather than using â as the crow fliesâ distance. Time-based searching gives users more opportunities for personalisation and delivers a more relevant search.

This package provides access to datasets, models and preprocessing facilities for deep learning with images. Integrates seamlessly with the torch package and it's API borrows heavily from PyTorch vision package.

With the objective of including data from RSS feeds into your analysis, tidyRSS parses RSS, Atom and JSON feeds and returns a tidy data frame.

Topological data analytic methods in machine learning rely on vectorizations of the persistence diagrams that encode persistent homology, as surveyed by Ali &al (2000) <doi:10.48550/arXiv.2212.09703>. Persistent homology can be computed using TDA and ripserr and vectorized using TDAvec'. The Tidymodels package collection modularizes machine learning in R for straightforward extensibility; see Kuhn & Silge (2022, ISBN:978-1-4920-9644-3). These recipe steps and dials tuners make efficient algorithms for computing and vectorizing persistence diagrams available for Tidymodels workflows.