This is an open-source R package for the statistical analysis of building's data within the framework of BIGG project. It contains a set of AI toolbox that allows the clustering, classification and modelling of building time series and its metadata. This package contains the needed functionalities to take advantage of the BIGG Ontology, thus the authors recommends its usage when elaborating your own data pipelines. This would generate a bigger chance of re-usage among different companies/users.
Once R is installed (>=4.1 is recommended), execute the following sentences to install all dependent libraries:
install.packages(c("devtools","pastecs","purrr","arules","glmnet","ranger","rdflib",
"magrittr","parsedate","lubridate","readr","dplyr","tidyr","tibble","zoo",
"roll","padr","quantreg","testthat","kernlab","fastDummies","caret","penalized",
"GA","parallel","mgcv","qgam","FinCal","onlineforecast","matrixStats","gratia")
# Additionally, is recommended to install libraries that are often used in pipelines based on 'biggr' functions
install.packages(c("plotly","ggplot2","carrier","mlflow","mongolite","gridExtra"))
Then, install the package through the GitHub installer if you want to use the last version of the code, or install an stable release located in the releases repository directly using the install.packages command.
# Install development version from GitHub
devtools::install_github("biggproject/biggr")
# Install stable version from the source package (*.tar.gz)
install.packages(<.tar.gz file>, repos = NULL, type ="source")
This R package is licensed under the MIT License. It also depends on other popular open-source R libraries, from which it will retain their licenses.