Electricity Load Curves Forecasting at Individual Level R package elcf4R version 0.4.0 Implements forecasting methods for individual electricity load curves, including Kernel Wavelet Functional (KWF), clustered KWF, Generalized Additive Models (GAM), Multivariate Adaptive Regression Splines (MARS), and Long Short-Term Memory (LSTM) models. Provides normalized dataset adapters for iFlex, StoreNet, Low Carbon London, and REFIT; download and read support for IDEAL and GX; explicit Python backend selection for TensorFlow-based LSTM fits; helpers for daily segmentation and rolling-origin benchmarking; and compact shipped example panels and benchmark-result datasets. Software Depends: R (>= 3.5.0) Imports: stats, utils, mgcv, earth, keras3, Rcpp, tensorflow, data.table, wavelets, jsonlite, xml2, DBI, RSQLite LinkingTo: Rcpp Suggests: knitr, reticulate, rmarkdown, testthat (>= 3.0.0) Frederic Bertrand <frederic.bertrand@lecnam.net> Comprehensive R Archive Network (CRAN) Frederic Bertrand [cre, aut] (ORCID: <https://orcid.org/0000-0002-0837-8281>), Fatima Fahs [aut], Myriam Maumy-Bertrand [aut] (ORCID: <https://orcid.org/0000-0002-4615-1512>) GPL-3 2026-04-21 application/tgz https://CRAN.R-project.org/package=elcf4R doi:10.32614/CRAN.package.elcf4R