Implementations several algorithms for kernel k-means. The default 'OTQT' algorithm is a fast alternative to standard implementations of kernel k-means, particularly in cases with many clusters. For a small number of clusters, the implemented 'MacQueen' method typically performs the fastest. For more details and performance evaluations, see Berlinski and Maitra (2025) &lt;<a href="https://doi.org/10.1002%2Fsam.70032" target="_top">doi:10.1002/sam.70032</a>&gt;.

Download, cache, and manage social contact survey data from the social contact data community on Zenodo (&lt;<a href="https://zenodo.org/communities/social_contact_data" target="_top">https://zenodo.org/communities/social_contact_data</a>&gt;) for use in infectious disease modelling. Provides functions to list available surveys, download survey files with automatic caching, and retrieve citations. Contact survey data describe who contacts whom in a population and are used to parameterise age-structured transmission models, for example via the 'socialmixr' package. The surveys available include those from the POLYMOD study (Mossong et al. (2008) &lt;<a href="https://doi.org/10.1371%2Fjournal.pmed.0050074" target="_top">doi:10.1371/journal.pmed.0050074</a>&gt;) and other social contact data shared on Zenodo.

Create self-contained SVG information cards with embedded 'Google Fonts', shields-style badges, and custom logos. Cards are fully portable SVG files ideal for dashboards, reports, and web applications. Includes functions to export cards to PNG format and display them in 'R Markdown' and 'Quarto' documents.

Provides R bindings to the 'Automerge' Conflict-free Replicated Data Type ('CRDT') library. 'Automerge' enables automatic merging of concurrent changes without conflicts, making it ideal for distributed systems, collaborative applications, and offline-first architectures. The approach of local-first software was proposed in Kleppmann, M., Wiggins, A., van Hardenberg, P., McGranaghan, M. (2019) &lt;<a href="https://doi.org/10.1145%2F3359591.3359737" target="_top">doi:10.1145/3359591.3359737</a>&gt;. This package supports all 'Automerge' data types (maps, lists, text, counters) and provides both low-level and high-level synchronization protocols for seamless interoperability with 'JavaScript' and other 'Automerge' implementations.

Provides scalable generalized linear and mixed effects models tailored for sequence count data analysis (e.g., analysis of 16S or RNA-seq data). Uses Dirichlet-multinomial sampling to quantify uncertainty in relative abundance or relative expression conditioned on observed count data. Implements scale models as a generalization of normalizations which account for uncertainty in scale (e.g., total abundances) as described in Nixon et al. (2025) &lt;<a href="https://doi.org/10.1186%2Fs13059-025-03609-3" target="_top">doi:10.1186/s13059-025-03609-3</a>&gt; and McGovern et al. (2025) &lt;<a href="https://doi.org/10.1101%2F2025.08.05.668734" target="_top">doi:10.1101/2025.08.05.668734</a>&gt;.

Renders plots to a temporary image using the ragg graphics device and returns knitr::include_graphics() output. Optionally saves the image to a specified path. This helps ensure consistent appearance across interactive sessions, saved files, and knitted documents. For more details see Pedersen and Shemanarev (2025) &lt;<a href="https://doi.org/10.32614%2FCRAN.package.ragg" target="_top">doi:10.32614/CRAN.package.ragg</a>&gt;.

Provides drop-in replacements for 'purrr' and 'furrr' mapping functions with built-in fault tolerance, automatic checkpointing, and seamless recovery capabilities. When long-running computations are interrupted due to errors, system crashes, or other failures, simply re-run the same code to automatically resume from the last checkpoint. Ideal for large-scale data processing, API calls, web scraping, and other time-intensive operations where reliability is critical. For 'purrr' methodology, see Wickham and Henry (2023) &lt;<a href="https://purrr.tidyverse.org/" target="_top">https://purrr.tidyverse.org/</a>&gt;.

https://cran.r-project.org/package=pmlsp

https://cran.r-project.org/package=pixr

Provides a unified interface for interacting with Large Language Models (LLMs) through various providers including OpenAI &lt;<a href="https://platform.openai.com/docs/api-reference" target="_top">https://platform.openai.com/docs/api-reference</a>&gt;, Ollama &lt;<a href="https://ollama.com/" target="_top">https://ollama.com/</a>&gt;, and other OpenAI-compatible APIs. Features include automatic connection testing, max_tokens limit auto-adjustment, structured JSON responses with schema validation, interactive JSON schema generation, prompt templating, and comprehensive diagnostics.

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.

Provides a higher-level interface to the 'torch' package for defining, training, and tuning neural networks. This package supports feedforward (multi-layer perceptron) and recurrent neural networks (RNN (Recurrent Neural Networks), LSTM (Long Short-Term Memory), GRU (Gated Recurrent Unit)), and also reduces boilerplate code while enabling seamless integration with 'torch'. The methods to train neural networks from this package also bridges to titanic ML frameworks in R namely 'tidymodels' ecosystem, enabling the 'parsnip' model specifications, workflows, recipes, and tuning tools.

Implementation of the FASSTER (Forecasting with Additive Switching of Seasonality, Trend, and Exogenous Regressors) model for forecasting time series with multiple seasonal patterns. The model combines state space methodology with a switching component in the observation equation to allow flexible modeling of complex seasonal patterns, including time-varying effects and multiple seasonalities.

Simulate population dynamics from realistically complex matrix population models in a plug-and-play fashion. Supports aspatial and spatially implicit models with one or more species and time-varying covariates, stochasticity, density dependence, additions or removals of individuals, interspecific interactions, and metapopulations.

Alternative R Frontend — a modern R console written in Rust - eitsupi/arf

Explore the QueryChat open-source project, math foundations of machine learning, and fresh tutorials on MLOps, data engineering, and AI tools in this curated roundup.

A repo for sharing data visualisation projects. Contribute to jemarnold/data-vis-threads development by creating an account on GitHub.