Learn Predictive Representations of Time-Varying Data R package timesift version 0.3.1 Fits and compares representations of time-varying data against a prediction target. Given a table of targets and a table of time-stamped series belonging to them, it builds each candidate representation, from the record unreduced through a calendar grain such as a week or a month to a lookback anchored on each target, fits the requested learners on each, scores every candidate on one set of held-out folds, and stacks the out-of-fold predictions into an ensemble. Calendar-aware binning keeps a bin a real week or month rather than a fixed block of hours. Learners, response heads and metrics are registered rather than hard-coded, so adding one is a registration and not a fork of the fitting code. The penalised baseline is an elastic net fitted by cyclic coordinate descent along a warm-started path, following Friedman, Hastie and Tibshirani (2010) <doi:10.18637/jss.v033.i01>. The shipped default is presence-absence with a joint multi-label head scored by the true skill statistic of Allouche, Tsoar and Kadmon (2006) <doi:10.1111/j.1365-2664.2006.01214.x>, the setting used for species distribution modelling from microclimate loggers. Software Depends: R (>= 4.1) Imports: grDevices, graphics, rlang, stats, tidyselect (>= 1.2.0), tools, utils LinkingTo: cpp11 Suggests: testthat (>= 3.0.0), torch, glmnet, ranger, lme4, lmerTest, emmeans, knitr, rmarkdown, spelling, withr Gilles Colling <gilles.colling051@gmail.com> Comprehensive R Archive Network (CRAN) Gilles Colling [aut, cre, cph] (ORCID: <https://orcid.org/0000-0003-3070-6066>) MIT + file LICENSE (https://CRAN.R-project.org/package=timesift/LICENSE) 2026-10-02 application/tgz https://CRAN.R-project.org/package=timesift doi:10.32614/CRAN.package.timesift en-GB