Tensors and Neural Networks with 'GPU' Acceleration

Provides functionality to define and train neural networks similar to 'PyTorch' by Paszke et al (2019) but written entirely in R using the 'libtorch' library. Also supports low-level tensor operations and 'GPU' acceleration.

D3 JavaScript Network Graphs from R

Creates 'D3' 'JavaScript' network, tree, dendrogram, and Sankey graphs from 'R'.

Fit, Simulate and Diagnose Exponential-Family Models for Networks

An integrated set of tools to analyze and simulate networks based on exponential-family random graph models (ERGMs). 'ergm' is a part of the Statnet suite of packages for network analysis. See Hunter, Handcock, Butts, Goodreau, and Morris (2008) and Krivitsky, Hunter, Morris, and Klumb (2023) .

Bayesian Network Structure Learning, Parameter Learning and Inference

Bayesian network structure learning, parameter learning and inference. This package implements constraint-based (PC, GS, IAMB, Inter-IAMB, Fast-IAMB, MMPC, Hiton-PC, HPC), pairwise (ARACNE and Chow-Liu), score-based (Hill-Climbing and Tabu Search) and hybrid (MMHC, RSMAX2, H2PC) structure learning algorithms for discrete, Gaussian and conditional Gaussian networks, along with many score functions and conditional independence tests. The Naive Bayes and the Tree-Augmented Naive Bayes (TAN) classifiers are also implemented. Some utility functions (model comparison and manipulation, random data generation, arc orientation testing, simple and advanced plots) are included, as well as support for parameter estimation (maximum likelihood and Bayesian) and inference, conditional probability queries, cross-validation, bootstrap and model averaging. Development snapshots with the latest bugfixes are available from < https://www.bnlearn.com/>.

Linear Networks Functionality of the 'spatstat' Family

Defines types of spatial data on a linear network and provides functionality for geometrical operations, data analysis and modelling of data on a linear network, in the 'spatstat' family of packages. Contains definitions and support for linear networks, including creation of networks, geometrical measurements, topological connectivity, geometrical operations such as inserting and deleting vertices, intersecting a network with another object, and interactive editing of networks. Data types defined on a network include point patterns, pixel images, functions, and tessellations. Exploratory methods include kernel estimation of intensity on a network, K-functions and pair correlation functions on a network, simulation envelopes, nearest neighbour distance and empty space distance, relative risk estimation with cross-validated bandwidth selection. Formal hypothesis tests of random pattern (chi-squared, Kolmogorov-Smirnov, Monte Carlo, Diggle-Cressie-Loosmore-Ford, Dao-Genton, two-stage Monte Carlo) and tests for covariate effects (Cox-Berman-Waller-Lawson, Kolmogorov-Smirnov, ANOVA) are also supported. Parametric models can be fitted to point pattern data using the function lppm() similar to glm(). Only Poisson models are implemented so far. Models may involve dependence on covariates and dependence on marks. Models are fitted by maximum likelihood. Fitted point process models can be simulated, automatically. Formal hypothesis tests of a fitted model are supported (likelihood ratio test, analysis of deviance, Monte Carlo tests) along with basic tools for model selection (stepwise(), AIC()) and variable selection (sdr). Tools for validating the fitted model include simulation envelopes, residuals, residual plots and Q-Q plots, leverage and influence diagnostics, partial residuals, and added variable plots. Random point patterns on a network can be generated using a variety of models.

Coercion Routines for Network Data Objects

Functions implemented in this package allow to coerce (i.e. convert) network data between classes provided by other R packages. Currently supported classes are those defined in packages: network and igraph.

Weighted Correlation Network Analysis

Functions necessary to perform Weighted Correlation Network Analysis on high-dimensional data as originally described in Horvath and Zhang (2005) and Langfelder and Horvath (2008) . Includes functions for rudimentary data cleaning, construction of correlation networks, module identification, summarization, and relating of variables and modules to sample traits. Also includes a number of utility functions for data manipulation and visualization.

Functions for Visualising Simple Graphs (Networks), Plotting Flow Diagrams

Visualises simple graphs (networks) based on a transition matrix, utilities to plot flow diagrams, visualising webs, electrical networks, etc. Support for the book "A practical guide to ecological modelling - using R as a simulation platform" by Karline Soetaert and Peter M.J. Herman (2009), Springer. and the book "Solving Differential Equations in R" by Karline Soetaert, Jeff Cash and Francesca Mazzia (2012), Springer. Includes demo(flowchart), demo(plotmat), demo(plotweb).

Neural Networks using the Stuttgart Neural Network Simulator (SNNS)

The Stuttgart Neural Network Simulator (SNNS) is a library containing many standard implementations of neural networks. This package wraps the SNNS functionality to make it available from within R. Using the 'RSNNS' low-level interface, all of the algorithmic functionality and flexibility of SNNS can be accessed. Furthermore, the package contains a convenient high-level interface, so that the most common neural network topologies and learning algorithms integrate seamlessly into R.

Network Meta-Analysis using Frequentist Methods

A comprehensive set of functions providing frequentist methods for network meta-analysis (Balduzzi et al., 2023) and supporting Schwarzer et al. (2015) , Chapter 8 "Network Meta-Analysis": - frequentist network meta-analysis following Rücker (2012) ; - additive network meta-analysis for combinations of treatments (Rücker et al., 2020) ; - network meta-analysis of binary data using the Mantel-Haenszel or non-central hypergeometric distribution method (Efthimiou et al., 2019) , or penalised logistic regression (Evrenoglou et al., 2022) ; - rankograms and ranking of treatments by the Surface under the cumulative ranking curve (SUCRA) (Salanti et al., 2013) ; - ranking of treatments using P-scores (frequentist analogue of SUCRAs without resampling) according to Rücker & Schwarzer (2015) ; - split direct and indirect evidence to check consistency (Dias et al., 2010) , (Efthimiou et al., 2019) ; - league table with network meta-analysis results; - 'comparison-adjusted' funnel plot (Chaimani & Salanti, 2012) ; - net heat plot and design-based decomposition of Cochran's Q according to Krahn et al. (2013) ; - measures characterizing the flow of evidence between two treatments by König et al. (2013) ; - automated drawing of network graphs described in Rücker & Schwarzer (2016) ; - partial order of treatment rankings ('poset') and Hasse diagram for 'poset' (Carlsen & Bruggemann, 2014) ; (Rücker & Schwarzer, 2017) ; - contribution matrix as described in Papakonstantinou et al. (2018) and Davies et al. (2022) ; - network meta-regression with a single continuous or binary covariate; - subgroup network meta-analysis.