This package provides modular, graph-based agents powered by large language models (LLMs) for intelligent task execution in R. Supports structured workflows for tasks such as forecasting, data visualization, feature engineering, data wrangling, data cleaning, SQL', code generation, weather reporting, and research-driven question answering. Each agent performs iterative reasoning: recommending steps, generating R code, executing, debugging, and explaining results. Includes built-in support for packages such as tidymodels', modeltime', plotly', ggplot2', and prophet'. Designed for analysts, developers, and teams building intelligent, reproducible AI workflows in R. Compatible with LLM providers such as OpenAI', Anthropic', Groq', and Ollama'. Inspired by the Python package langagent'.

Estimates marginal likelihood from a posterior sample using the method described in Wang et al. (2023) <doi:10.1093/sysbio/syad007>, which does not require evaluation of any additional points and requires only the log of the unnormalized posterior density for each sampled parameter vector.

Estimation of various extensions of the mixed models including latent class mixed models, joint latent class mixed models, mixed models for curvilinear outcomes, mixed models for multivariate longitudinal outcomes using a maximum likelihood estimation method (Proust-Lima, Philipps, Liquet (2017) <doi:10.18637/jss.v078.i02>).

Constructs tables of counts and proportions out of data sets with possibility to insert tables to Excel, Word, HTML, and PDF documents. Transforms tables to data suitable for modelling. Features Gibbs sampling based log-linear (NB2) and power analyses (original by Oleksandr Ocheredko <doi:10.35566/isdsa2019c5>) for tabulated data.

Fit different model forms to single-cohort litter decomposition data (mass remaining through time) using likelihood-based estimation. Models span simple empirical to process-motivated forms with differing numbers of free parameters. Provides parameter estimates, uncertainty, and tools for model comparison/selection. Based on Cornwell & Weedon (2013) <doi:10.1111/2041-210X.12138>.

Four measures of linkage disequilibrium are provided: the usual r^2 measure, the r^2_S measure (r^2 corrected by the structure sample), the r^2_V (r^2 corrected by the relatedness of genotyped individuals), the r^2_VS measure (r^2 corrected by both the relatedness of genotyped individuals and the structure of the sample).

This package provides functions for the implementation of a density goodness-of-fit test, based on piecewise approximation of the L2 distance.

Probabilistic record linkage without direct identifiers using only diagnosis codes. Method is detailed in: Hejblum, Weber, Liao, Palmer, Churchill, Szolovits, Murphy, Kohane & Cai (2019) <doi: 10.1038/sdata.2018.298> ; Zhang, Hejblum, Weber, Palmer, Churchill, Szolovits, Murphy, Liao, Kohane & Cai (2021) <doi: 10.1093/jamia/ocab187>.

This package provides a curated collection of Howard Phillips Lovecraft's complete stories, collected for the purpose of text analysis.

This package provides a wrapper built around the libLBFGS optimization library by Naoaki Okazaki. The lbfgs package implements both the Limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) and the Orthant-Wise Quasi-Newton Limited-Memory (OWL-QN) optimization algorithms. The L-BFGS algorithm solves the problem of minimizing an objective, given its gradient, by iteratively computing approximations of the inverse Hessian matrix. The OWL-QN algorithm finds the optimum of an objective plus the L1-norm of the problem's parameters. The package offers a fast and memory-efficient implementation of these optimization routines, which is particularly suited for high-dimensional problems.

This package provides tools for detecting and correcting sample mix-ups between two sets of measurements, such as between gene expression data on two tissues. This is a revised version of the lineup package, to be more general and not tied to the qtl package.

Fast and accurate inference of gene-environment associations (GEA) in genome-wide studies (Caye et al., 2019, <doi:10.1093/molbev/msz008>). We developed a least-squares estimation approach for confounder and effect sizes estimation that provides a unique framework for several categories of genomic data, not restricted to genotypes. The speed of the new algorithm is several times faster than the existing GEA approaches, then our previous version of the LFMM program present in the LEA package (Frichot and Francois, 2015, <doi:10.1111/2041-210X.12382>).

This package provides a classification tree method that uses Uncorrelated Linear Discriminant Analysis (ULDA) for variable selection, split determination, and model fitting in terminal nodes. It automatically handles missing values and offers visualization tools. For more details, see Wang (2024) <doi:10.48550/arXiv.2410.23147>.

Implementation based on Zhang, Jie & Huang, Kun (2014) <doi:10.4137/CIN.S14021> Normalized ImQCM: An Algorithm for Detecting Weak Quasi-Cliques in Weighted Graph with Applications in Gene Co-Expression Module Discovery in Cancers. Cancer informatics, 13, CIN-S14021.

This package provides a Low Rank Correction Variational Bayesian algorithm for high-dimensional multi-source heterogeneous quantile linear models. More details have been written up in a paper submitted to the journal Statistics in Medicine, and the details of variational Bayesian methods can be found in Ray and Szabo (2021) <doi:10.1080/01621459.2020.1847121>. It simultaneously performs parameter estimation and variable selection. The algorithm supports two model settings: (1) local models, where variable selection is only applied to homogeneous coefficients, and (2) global models, where variable selection is also performed on heterogeneous coefficients. Two forms of parameter estimation are output: one is the standard variational Bayesian estimation, and the other is the variational Bayesian estimation corrected with low-rank adjustment.

This package provides a LaTeX Letter class for rmarkdown', using the pandoc-letter template adapted for use with markdown'.

Logic Forest is an ensemble machine learning method that identifies important and interpretable combinations of binary predictors using logic regression trees to model complex relationships with an outcome. Wolf, B.J., Slate, E.H., Hill, E.G. (2010) <doi:10.1093/bioinformatics/btq354>.

This package provides a method for detecting multiple change points in high-dimensional time series, targeting dense or spatially clustered signals. See Li et al. (2023) "L2 Inference for Change Points in High-Dimensional Time Series via a Two-Way MOSUM". arXiv preprint <arXiv:2208.13074>.

Lights Out is a puzzle game consisting of a grid of lights that are either on or off. Pressing any light will toggle it and its adjacent lights. The goal of the game is to switch all the lights off. This package provides an interface to play the game on different board sizes, both through the command line or with a visual application. Puzzles can also be solved using the automatic solver included. View a demo online at <https://daattali.com/shiny/lightsout/>.

Fit the log binomial regression model (LBM) by Exact method. Limited parameter space of LBM causes trouble to find admissible estimates and fail to converge when MLE is close to or on the boundary of space. Exact method utilizes the property of boundary vectors to re-parametrize the model without losing any information, and fits the model on the standard fitting algorithm with no convergence issues.

This package provides a framework for integrating Large Language Models (LLMs) with R programming through workflow automation. Built on the ReAct (Reasoning and Acting) architecture, enables bi-directional communication between LLMs and R environments. Features include automated code generation and execution, intelligent error handling with retry mechanisms, persistent session management, structured JSON output validation, and context-aware conversation management.

This package provides a variety of ordination and community analyses useful in analysis of data sets in community ecology. Includes many of the common ordination methods, with graphical routines to facilitate their interpretation, as well as several novel analyses.

This package implements Cumulative Sum (CUSUM) control charts specifically designed for monitoring processes following a Gamma distribution. Provides functions to estimate distribution parameters, simulate control limits, and apply cautious learning schemes for adaptive thresholding. It supports upward and downward monitoring with guaranteed performance evaluated via Monte Carlo simulations. It is useful for quality control applications in industries where data follows a Gamma distribution. Methods are based on Madrid-Alvarez et al. (2024) <doi:10.1002/qre.3464> and Madrid-Alvarez et al. (2024) <doi:10.1080/08982112.2024.2440368>.

This package provides methods of developing linear time series modelling. Methods are given for loglikelihood computation, forecasting and simulation.