This package provides R bindings to llama.cpp for running Large Language Models ('LLMs') locally with optional Vulkan GPU acceleration via ggmlR'. Supports model loading, text generation, tokenization', token-to-piece conversion, embeddings (single and batch), encoder-decoder inference, low-level batch management, chat templates, LoRA adapters, explicit backend/device selection, multi-GPU split, and NUMA optimization. Includes a high-level ragnar'-compatible embedding provider ('embed_llamar'). Built on top of ggmlR for efficient tensor operations.

This package provides two methods of estimating income inequality statistics from binned income data, such as the income data provided in the Census. These methods use different interpolation techniques to infer the distribution of incomes within income bins. One method is an implementation of Jargowsky and Wheeler's mean-constrained integration over brackets (MCIB). The other method is based on a new technique, Lorenz interpolation, which estimates income inequality by constructing an interpolated Lorenz curve based on the binned income data. These methods can be used to estimate three income inequality measures: the Gini (the default measure returned), the Theil, and the Atkinson's index. Jargowsky and Wheeler (2018) <doi:10.1177/0081175018782579>.

Clustering or classification of longitudinal data based on a mixture of multivariate t or Gaussian distributions with a Cholesky-decomposed covariance structure. Details in McNicholas and Murphy (2010) <doi:10.1002/cjs.10047> and McNicholas and Subedi (2012) <doi:10.1016/j.jspi.2011.11.026>.

The goal of LCMSQA is to make it easy to check the quality of liquid chromatograph/mass spectrometry (LC/MS) experiments using a shiny application. This package provides interactive data visualizations for quality control (QC) samples, including total ion current chromatogram (TIC), base peak chromatogram (BPC), mass spectrum, extracted ion chromatogram (XIC), and feature detection results from internal standards or known metabolites.

Detect feedback loops (cycles, circuits) between species (nodes) in ordinary differential equation (ODE) models. Feedback loops are paths from a node to itself without visiting any other node twice, and they have important regulatory functions. Loops are reported with their order of participating nodes and their length, and whether the loop is a positive or a negative feedback loop. An upper limit of the number of feedback loops limits runtime (which scales with feedback loop count). Model parametrizations and values of the modelled variables are accounted for. Computation uses the characteristics of the Jacobian matrix as described e.g. in Thomas and Kaufman (2002) <doi:10.1016/s1631-0691(02)01452-x>. Input can be the Jacobian matrix of the ODE model or the ODE function definition; in the latter case, the Jacobian matrix is determined using numDeriv'. Graph-based algorithms from igraph are employed for path detection.

Regression analysis of mixed sparse synchronous and asynchronous longitudinal covariates. Please cite the manuscripts corresponding to this package: Sun, Z. et al. (2023) <arXiv:2305.17715> and Liu, C. et al. (2023) <arXiv:2305.17662>.

Import, processing, validation, and visualization of personal light exposure measurement data from wearable devices. The package implements features such as the import of data and metadata files, conversion of common file formats, validation of light logging data, verification of crucial metadata, calculation of common parameters, and semi-automated analysis and visualization.

We provide a solution for performing permutation tests on linear and mixed linear regression models. It allows users to obtain accurate p-values without making distributional assumptions about the data. By generating a null distribution of the test statistics through repeated permutations of the response variable, permutation tests provide a powerful alternative to traditional parameter tests (Holt et al. (2023) <doi:10.1007/s10683-023-09799-6>). In this early version, we focus on the permutation tests over observed t values of beta coefficients, i.e.original t values generated by parameter tests. After generating a null distribution of the test statistic through repeated permutations of the response variable, each observed t values would be compared to the null distribution to generate a p-value. To improve the efficiency,a stop criterion (Anscombe (1953) <doi:10.1111/j.2517-6161.1953.tb00121.x>) is adopted to force permutation to stop if the estimated standard deviation of the value falls below a fraction of the estimated p-value. By doing so, we avoid the need for massive calculations in exact permutation methods while still generating stable and accurate p-values.

LimeSurvey is Free/Libre Open Source Software for the development and administrations of online studies, using sophisticated tailoring capabilities to support multiple study designs (see <https://www.limesurvey.org>). This package supports programmatic creation of surveys that can then be imported into LimeSurvey', as well as user friendly import of responses from LimeSurvey studies.

Time-dependent Receiver Operating Characteristic curves, Area Under the Curve, and Net Reclassification Indexes for repeated measures. It is based on methods in Barbati and Farcomeni (2017) <doi:10.1007/s10260-017-0410-2>.

This package creates plots of peptides from shotgun proteomics analysis of secretome and lysate samples. These plots contain associated protein features and scores for potential secretion and truncation.

Lake morphometry metrics are used by limnologists to understand, among other things, the ecological processes in a lake. Traditionally, these metrics are calculated by hand, with planimeters, and increasingly with commercial GIS products. All of these methods work; however, they are either outdated, difficult to reproduce, or require expensive licenses to use. The lakemorpho package provides the tools to calculate a typical suite of these metrics from an input elevation model and lake polygon. The metrics currently supported are: fetch, major axis, minor axis, major/minor axis ratio, maximum length, maximum width, mean width, maximum depth, mean depth, shoreline development, shoreline length, surface area, and volume.

This package provides an extension to factors called lfactor that are similar to factors but allows users to refer to lfactor levels by either the level or the label.

This package provides tools for fast and accurate evaluation of skew stable distributions (CDF, PDF and quantile functions), random number generation, and parameter estimation. This is libstableR as per Royuela del Val, Simmross-Wattenberg, and Alberola López (2017) <doi:10.18637/jss.v078.i01> under a new maintainer.

To decompose symmetric matrices such as brain connectivity matrices so that one can extract sparse latent component matrices and also estimate mixing coefficients, a blind source separation (BSS) method named LOCUS was proposed in Wang and Guo (2023) <arXiv:2008.08915>. For brain connectivity matrices, the outputs correspond to sparse latent connectivity traits and individual-level trait loadings.

These functions take a gene expression value matrix, a primary covariate vector, an additional known covariates matrix. A two stage analysis is applied to counter the effects of latent variables on the rankings of hypotheses. The estimation and adjustment of latent effects are proposed by Sun, Zhang and Owen (2011). "leapp" is developed in the context of microarray experiments, but may be used as a general tool for high throughput data sets where dependence may be involved.

R lists, especially nested lists, can be very difficult to visualize or represent. Sometimes str() is not enough, so this suite of htmlwidgets is designed to help see, understand, and maybe even modify your R lists. The function reactjson() requires a package reactR that can be installed from CRAN or <https://github.com/timelyportfolio/reactR>.

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. Broman et al. (2015) <doi:10.1534/g3.115.019778>.

In addition to modeling the expectation (location) of an outcome, mixed effects location scale models (MELSMs) include submodels on the variance components (scales) directly. This allows models on the within-group variance with mixed effects, and between-group variances with fixed effects. The MELSM can be used to model volatility, intraindividual variance, uncertainty, measurement error variance, and more. Multivariate MELSMs (MMELSMs) extend the model to include multiple correlated outcomes, and therefore multiple locations and scales. The latent multivariate MELSM (LMMELSM) further includes multiple correlated latent variables as outcomes. This package implements two-level mixed effects location scale models on multiple observed or latent outcomes, and between-group variance modeling. Williams, Martin, Liu, and Rast (2020) <doi:10.1027/1015-5759/a000624>. Hedeker, Mermelstein, and Demirtas (2008) <doi:10.1111/j.1541-0420.2007.00924.x>.

Effectively simulates the discretization process inherent to Likert scales while minimizing distortion. It converts continuous latent variables into ordinal categories to generate Likert scale item responses. Particularly useful for accurately modeling and analyzing survey data that use Likert scales, especially when applying statistical techniques that require metric data.

This package provides tools to decompose differences in cohort health expectancy (HE) by age and cause using longitudinal data. The package implements a novel longitudinal attribution method based on a semiparametric additive hazards model with time-dependent covariates, specifically designed to address interval censoring and semi-competing risks via a copula framework. The resulting age-cause-specific contributions to disability prevalence and death probability can be used to quantify and decompose differences in cohort HE between groups. The package supports stepwise replacement decomposition algorithms and is applicable to cohort-based health disparity research across diverse populations. Related methods include Sun et al. (2023) <doi:10.1177/09622802221133552>.

Calculation of rectifying LTPD and AOQL plans for sampling inspection by variables which minimize mean inspection cost per lot of process average quality.

Alternate font rendering is useful when rendering text to novel graphics outputs where modern font rendering is not available or where bespoke text positioning is required. Bitmap and vector fonts allow for custom layout and rendering using pixel coordinates and line drawing. Formatted text is created as a data.frame of pixel coordinates (for bitmap fonts) or stroke coordinates (for vector fonts). All text can be easily previewed as a matrix or raster image. A selection of fonts is included with this package.

This package provides functions for estimating the gliding box lacunarity (GBL), covariance, and pair-correlation of a random closed set (RACS) in 2D from a binary coverage map (e.g. presence-absence land cover maps). Contains a number of newly-developed covariance-based estimators of GBL (Hingee et al., 2019) <doi:10.1007/s13253-019-00351-9> and balanced estimators, proposed by Picka (2000) <http://www.jstor.org/stable/1428408>, for covariance, centred covariance, and pair-correlation. Also contains methods for estimating contagion-like properties of RACS and simulating 2D Boolean models. Binary coverage maps are usually represented as raster images with pixel values of TRUE, FALSE or NA, with NA representing unobserved pixels. A demo for extracting such a binary map from a geospatial data format is provided. Binary maps may also be represented using polygonal sets as the foreground, however for most computations such maps are converted into raster images. The package is based on research conducted during the author's PhD studies.