Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://github.com/sshpa/bayesvl below:

sshpa/bayesvl: BayesVL package for Bayesian statistical analyses in R

The R package for visually learning the graphical structures of Bayesian networks, and performing Hamiltonian Markov chain Monte Carlo (MCMC) with 'Stan'.

Features:

* Creating the (starting) graphical structure of Bayesian networks
* Creating one or more random Bayesian networks learned from dataset with customized constraints
* Generating Stan code for structures of Bayesian networks for sampling and parameter learning
* Plotting the Bayesian network graphs 
* Performing Markov chain Monte Carlo (MCMC) simulations and plotting various graphs for posteriors check
* Compatibility with R 3.4 or newer versions

Here is the CHANGELOG

• User guide (user manual, case studies, and more)
• Open an issue (GitHub issues for bug reports, feature requests)
• Example code (example code snippets referenced in user manual)

You'll need to install rstan first. Stan’s website http://mc-stan.org provides up-to-date information, follow the instructions for your platform. For instructions on installing a C++ compiler for use with RStan see https://github.com/stan-dev/rstan/wiki/RStan-Getting-Started are quite thorough.

Getting started and installing the latest snapshot, type in the R console:

install.packages(c("coda","devtools","loo","ggplot2"))
devtools::install_github("sshpa/bayesvl")

Creating a node for each variable in the proposed network

dag <- bayesvl()
dag <- bvl_addNode(dag, "Burden", "norm")
dag <- bvl_addNode(dag, "Res", "norm")
dag <- bvl_addNode(dag, "Insured", "norm")

Starting to add arcs between variables (nodes) using the survey data

dag <- bvl_addArc(dag, "Res", "Burden", "slope")
dag <- bvl_addArc(dag, "Insured", "Burden", "slope")

Generating the 'Stan' code required for building structures of the Bayesian networks required for sampling and parameter learning

stan_code <- bvl_model2Stan(dag)
cat(stan_code)

Getting the model's parameters

params <- bvl_stanParams(dag)
params

In this example, we used the data1042 data set that is built-in to the package.

# load data
data( data1042 )

Sampling the predefined 'Stan' model

stan_fit <- bvl_modelFit(dag, data1042, iter=5000 , warmup=2000 , chains=4 , cores=4)
summary(stan_fit)

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4