General Package for Meta-Analysis

User-friendly general package providing standard methods for meta-analysis and supporting Schwarzer, Carpenter, and Rücker , "Meta-Analysis with R" (2015): - common effect and random effects meta-analysis; - several plots (forest, funnel, Galbraith / radial, L'Abbe, Baujat, bubble); - three-level meta-analysis model; - generalised linear mixed model; - logistic regression with penalised likelihood for rare events; - Hartung-Knapp method for random effects model; - Kenward-Roger method for random effects model; - prediction interval and density of the prediction distribution; - expected proportion of comparable studies with clinically important benefit or harm; - statistical tests for funnel plot asymmetry; - trim-and-fill method to evaluate bias in meta-analysis; - meta-regression; - cumulative meta-analysis and leave-one-out meta-analysis; - import data from 'RevMan 5'; - produce forest plot summarising several (subgroup) meta-analyses.

Dimension Reduction, Regression and Discrimination for Chemometrics

Data exploration and prediction with focus on high dimensional data and chemometrics. The package was initially designed about partial least squares regression and discrimination models and variants, in particular locally weighted PLS models (LWPLS). Then, it has been expanded to many other methods for analyzing high dimensional data. The name 'rchemo' comes from the fact that the package is orientated to chemometrics, but most of the provided methods are fully generic to other domains. Functions such as transform(), predict(), coef() and summary() are available. Tuning the predictive models is facilitated by generic functions gridscore() (validation dataset) and gridcv() (cross-validation). Faster versions are also available for models based on latent variables (LVs) (gridscorelv() and gridcvlv()) and ridge regularization (gridscorelb() and gridcvlb()).

A Byte-Pair-Encoding (BPE) Tokenizer for OpenAI's Large Language Models

A thin wrapper around the tiktoken-rs crate, allowing to encode text into Byte-Pair-Encoding (BPE) tokens and decode tokens back to text. This is useful to understand how Large Language Models (LLMs) perceive text.

Secure and Intuitive Access to 'Plug' Interface

Provides a secure and user-friendly interface to interact with the 'Plug' < https://plugbytpf.com.br> 'API'. It enables developers to store and manage tokens securely using the 'keyring' package, retrieve data from 'API' endpoints with the 'httr2' package, and handle large datasets with chunked data fetching. Designed for simplicity and security, the package facilitates seamless integration with 'Plug' ecosystem.

Goodness-of-Fit Tests for Capture-Recapture Models

Performs goodness-of-fit tests for capture-recapture models as described by Gimenez et al. (2018) . Also contains several functions to process capture-recapture data.

Approximate Bayesian Regularization for Parsimonious Estimates

Approximate Bayesian regularization using Gaussian approximations. The input is a vector of estimates and a Gaussian error covariance matrix of the key parameters. Bayesian shrinkage is then applied to obtain parsimonious solutions. The method is described on Karimova, van Erp, Leenders, and Mulder (2025) . Gibbs samplers are used for model fitting. The shrinkage priors that are supported are Gaussian (ridge) priors, Laplace (lasso) priors (Park and Casella, 2008 ), and horseshoe priors (Carvalho, et al., 2010; ). These priors include an option for grouped regularization of different subsets of parameters (Meier et al., 2008; ). F priors are used for the penalty parameters lambda^2 (Mulder and Pericchi, 2018 ). This correspond to half-Cauchy priors on lambda (Carvalho, Polson, Scott, 2010 ).

Bayesian Model to Estimate Population Trends from Counts Series

Infers the trends of one or several animal populations over time from series of counts. It does so by accounting for count precision (provided or inferred based on expert knowledge, e.g. guesstimates), smoothing the population rate of increase over time, and accounting for the maximum demographic potential of species. Inference is carried out in a Bayesian framework. This work is part of the FRB-CESAB working group AfroBioDrivers < https://www.fondationbiodiversite.fr/en/the-frb-in-action/programs-and-projects/le-cesab/afrobiodrivers/>.

Fast Network Modularity and Roles Computation by Simulated Annealing (Rgraph C Library Wrapper for R)

Provides functions to compute the modularity and modularity-related roles in networks. It is a wrapper around the rgraph library (Guimera & Amaral, 2005, ).

Spatial Data Download and Utility Functions

A suite of conversion functions to create internally standardized spatial polygons data frames. Utility functions use these data sets to return values such as country, state, time zone, watershed, etc. associated with a set of longitude/latitude pairs. (They also make cool maps.)

Improved Methods for Constructing Prediction Intervals for Network Meta-Analysis

Improved methods to construct prediction intervals for network meta-analysis. The parametric bootstrap and Kenward-Roger-type adjustment by Noma et al. (2022) are implementable.