Multiple ways to bin numeric columns with a tidy output. Wraps a variety of existing binning methods into one function, and includes a new method for binning by equal value, which is useful for sales data. Provides a function to automatically summarize the properties of the binned columns.

This package provides a framework for the creation and use of Neural ordinary differential equations with the tensorflow and keras packages. The idea of Neural ordinary differential equations comes from Chen et al. (2018) <doi:10.48550/arXiv.1806.07366>, and presents a novel way of learning and solving differential systems.

Allows forecasting time series using nearest neighbors regression Francisco Martinez, Maria P. Frias, Maria D. Perez-Godoy and Antonio J. Rivera (2019) <doi:10.1007/s10462-017-9593-z>. When the forecasting horizon is higher than 1, two multi-step ahead forecasting strategies can be used. The model built is autoregressive, that is, it is only based on the observations of the time series. The nearest neighbors used in a prediction can be consulted and plotted.

Enhances koRpus text object classes and methods to also support large corpora. Hierarchical ordering of corpus texts into arbitrary categories will be preserved. Provided classes and methods also improve the ability of using the koRpus package together with the tm package. To ask for help, report bugs, suggest feature improvements, or discuss the global development of the package, please subscribe to the koRpus-dev mailing list (<https://korpusml.reaktanz.de>).

The goal of this package will be to provide a simple interface for automatic machine learning that fits the tidymodels framework. The intention is to work for regression and classification problems with a simple verb framework.

Statistical exploration of textual corpora using several methods from French Textometrie (new name of Lexicometrie') and French Data Analysis schools. It includes methods for exploring irregularity of distribution of lexicon features across text sets or parts of texts (Specificity analysis); multi-dimensional exploration (Factorial analysis), etc. Those methods are used in the TXM software.

The goal of tidyheatmaps is to simplify the generation of publication-ready heatmaps from tidy data. By offering an interface to the powerful pheatmap package, it allows for the effortless creation of intricate heatmaps with minimal code.

This package implements inverse and augmented inverse probability weighted estimators for common treatment effect parameters at an interim analysis with time-lagged outcome that may not be available for all enrolled subjects. Produces estimators, standard errors, and information that can be used to compute stopping boundaries using software that assumes that the estimators/test statistics have independent increments. Tsiatis, A. A. and Davidian, M., (2022) <doi:10.1002/sim.9580> .

Analyse data from longitudinal studies to characterise changes in values of semi-quantitative outcome variables within individual subjects, using high performance C++ code to enable rapid processing of large datasets. A flexible methodology is available for codifying these state transitions.

This package provides bindings to a C grammar for Tree-sitter, to be used alongside the treesitter package. Tree-sitter builds concrete syntax trees for source files and can efficiently update them or generate code like producing R C API wrappers from C functions, structs and global definitions from header files.

Most estimators implemented by the video game industry cannot obtain reliable initial estimates nor guarantee comparability between distant estimates. TrueSkill Through Time solves all these problems by modeling the entire history of activities using a single Bayesian network allowing the information to propagate correctly throughout the system. This algorithm requires only a few iterations to converge, allowing millions of observations to be analyzed using any low-end computer. Landfried G, Mocskos E (2025). "TrueSkill Through Time: Reliable Initial Skill Estimates and Historical Comparability with Julia, Python, and R." <doi:10.18637/jss.v112.i06>. The core ideas implemented in this project were developed by Dangauthier P, Herbrich R, Minka T, Graepel T (2007). "Trueskill through time: Revisiting the history of chess.".

Access Open Trade Statistics API from R to download international trade data.

This package contains performance analysis metrics of track records including entropy-based correlation and dynamic beta based on a state/space algorithm. The normalized sample entropy method has been implemented which produces accurate entropy estimation even on smaller datasets. On a separate stream, trades from the five major assets classes and also functionality to use pricing curves, rating tables, Credit Support Annex and add-on tables. The implementation follows an object oriented logic whereby each trade inherits from more abstract classes while also the curves/tables are objects. Furthermore, odds calculators and P&L back-testing functionality has been implemented for the most widely used betting/trading strategies including martingale, DAlembert', Labouchere and Fibonacci. Back testing has also been included for the EuroMillions', the EuroJackpot', the UK Lotto, the Set For Life and the UK ThunderBall lotteries. Furthermore, some basic functionality about climate risk has been included.

Estimates the time-varying (tv) parameters of the GARCH(1,1) model, enabling the modeling of non-stationary volatilities by allowing the model parameters to change gradually over time. The estimation and prediction processes are facilitated through the application of the Kalman filter and state-space equations. This package supports the estimation of tv parameters for various deterministic functions, which can be identified through exploratory analysis of different time periods or segments of return data. The methodology is grounded in the framework presented by Ferreira et al. (2017) <doi:10.1080/00949655.2017.1334778>.

Base R sometimes requires verbose statements for simple, often recurring tasks, such as printing text without trailing space, ending with newline. This package aims at providing shorthands for such tasks.

This package provides a calculator for the two-dimensional clinical Disease Activity index for Psoriatic Arthritis (TwoDcDAPSA), a principal component-derived measure that complements the conventional clinical DAPSA score. TwoDcDAPSA captures residual variation in patient-reported outcomes (pain and patient global assessment) and joint counts (swollen and tender) after adjusting for standardized cDAPSA using natural spline coefficients derived from published models. Residuals are standardized and combined with fixed principal component loadings to yield two continuous component scores: the PROs-Joint Contrast (PJC) and the Swollenâ Tender joints Contrast (STC), along with quartile-based groupings (including optional combined quartile groupings). The package applies pre-specified coefficients, residual standardization, and loadings to new datasets but does not estimate spline models or principal components itself.

Collection of ancillary functions and utilities to be used in conjunction with the TraMineR package for sequence data exploration. Includes, among others, specific functions such as state survival plots, position-wise group-typical states, dynamic sequence indicators, and dissimilarities between event sequences. Also includes contributions by non-members of the TraMineR team such as methods for polyadic data and for the comparison of groups of sequences.

Fast calculation of the Subtree Prune and Regraft (SPR), Tree Bisection and Reconnection (TBR) and Replug distances between unrooted trees, using the algorithms of Whidden and Matsen (2017) <doi:10.48550/arXiv.1511.07529>.

The goal of TailID is to detect sensitive points in the tail of a dataset using techniques from Extreme Value Theory (EVT). It utilizes the Generalized Pareto Distribution (GPD) for assessing tail behavior and detecting inconsistent points with the Identical Distribution hypothesis of the tail. For more details see Manau (2025)<doi:10.4230/LIPIcs.ECRTS.2025.20>.

Data filtering module for teal applications. Allows for interactive filtering of data stored in data.frame and MultiAssayExperiment objects. Also displays filtered and unfiltered observation counts.

This package provides functions for tabulating and summarising categorical variables. Most functions are designed to work with dataframes, and use the tidyverse idiom of taking the dataframe as the first argument so they work within pipelines. Equivalent functions that operate directly on vectors are also provided where it makes sense. This package aims to make exploratory data analysis involving categorical variables quicker, simpler and more robust.

Identifies clusters of individual longitudinal trajectories. In the spirit of Leffondre et al. (2004), the procedure involves identifying each trajectory to a point in the space of measures. In this context, a measure is a quantity meant to capture a certain characteristic feature of the trajectory. The points in the space of measures are then clustered using a version of spectral clustering.

Turn complex JSON data into tidy data frames.

This package provides tools for reading, parsing, indexing, and exporting LAS (Log ASCII Standard) well log files into tidy, analysis-ready tabular formats. The package separates LAS header information and log data into structured components, builds a searchable index across collections of LAS files, and enables reproducible subsetting of wells based on metadata or curve availability. Output tables can be written to CSV or Parquet formats to support large-scale statistical, machine learning, and earth science workflows. The tidy data structure follows Wickham (2014) <doi:10.18637/jss.v059.i10>. The LAS file structure follows the Canadian Well Logging Society LAS standard <https://www.cwls.org/wp-content/uploads/2017/02/Las2_Update_Jan2017.pdf>.