{ "_id": "6abfaf6599f7aa8c6aa5941f", "Package": "timesift", "Title": "Learn Predictive Representations of Time-Varying Data", "Version": "0.3.1", "Authors@R": "person(\"Gilles\", \"Colling\", email = \"gilles.colling051@gmail.com\",\nrole = c(\"aut\", \"cre\", \"cph\"),\ncomment = c(ORCID = \"0000-0003-3070-6066\"))", "Description": "Fits and compares representations of time-varying data\nagainst a prediction target. Given a table of targets and a\ntable of time-stamped series belonging to them, it builds each\ncandidate representation, from the record unreduced through a\ncalendar grain such as a week or a month to a lookback anchored\non each target, fits the requested learners on each, scores\nevery candidate on one set of held-out folds, and stacks the\nout-of-fold predictions into an ensemble. Calendar-aware\nbinning keeps a bin a real week or month rather than a fixed\nblock of hours. Learners, response heads and metrics are\nregistered rather than hard-coded, so adding one is a\nregistration and not a fork of the fitting code. The penalised\nbaseline is an elastic net fitted by cyclic coordinate descent\nalong a warm-started path, following Friedman, Hastie and\nTibshirani (2010) . The shipped\ndefault is presence-absence with a joint multi-label head\nscored by the true skill statistic of Allouche, Tsoar and\nKadmon (2006) , the\nsetting used for species distribution modelling from\nmicroclimate loggers.", "License": "MIT + file LICENSE", "Encoding": "UTF-8", "Language": "en-GB", "URL": "https://gillescolling.com/timesift/,\nhttps://github.com/gcol33/timesift", "BugReports": "https://github.com/gcol33/timesift/issues", "VignetteBuilder": "knitr", "Config/testthat/edition": "3", "Config/roxygen2/version": "8.1.0", "NeedsCompilation": "yes", "Packaged": { "Date": "2026-10-02 13:11:49 UTC", "User": "root" }, "Author": "Gilles Colling [aut, cre, cph] (ORCID:\n)", "Maintainer": "Gilles Colling ", "Repository": "https://cran.r-universe.dev", "Date/Publication": "2026-10-02 12:53:17 UTC", "RemoteUrl": "https://github.com/cran/timesift", "RemoteRef": "HEAD", "RemoteSha": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "_user": "cran", "_type": "src", "_file": "timesift_0.3.1.tar.gz", "_fileid": "https://r2.ropensci.org/e6af79e2cfbdce7a2d07eef49ee723bc08265cf0577cdb3dd5be7e36b8e9f14e", "_filesize": 1207704, "_sha256": "e6af79e2cfbdce7a2d07eef49ee723bc08265cf0577cdb3dd5be7e36b8e9f14e", "_expires": "2027-01-10T13:19:31.000Z", "_created": "2026-10-02T13:11:49.000Z", "_published": "2026-10-02T13:19:33.840Z", "_jobs": [ { "job": 110851909025, "time": 237, "config": "linux-devel-arm64", "r": "4.7.0", "check": "OK", "artifact": "11228186809" }, { "job": 110851909128, "time": 276, "config": "linux-devel-x86_64", "r": "4.7.0", "check": "OK", "artifact": "11228272241" }, { "job": 110851909068, "time": 262, "config": "linux-release-arm64", "r": "4.6.1", "check": "OK", "artifact": "11228331527" }, { "job": 110851909099, "time": 287, "config": "linux-release-x86_64", "r": "4.6.1", "check": "OK", "artifact": "11228048448" }, { "job": 110850397067, "time": 254, "config": "source", "r": "4.6.1", "check": "OK", "artifact": "11228386330" }, { "job": 110851908966, "time": 142, "config": "wasm-release", "r": "4.6.0", "check": "OK", "artifact": "11228056782" }, { "job": 110851909165, "time": 399, "config": "windows-devel-arm64", "r": "4.7.0", "check": "OK", "artifact": "11228696244" }, { "job": 110851909339, "time": 290, "config": "windows-devel-x86_64", "r": "4.7.0", "check": "OK", "artifact": "11228531487" }, { "job": 110851909204, "time": 405, "config": "windows-release-arm64", "r": "4.6.1", "check": "OK", "artifact": "11227848342" }, { "job": 110851909042, "time": 309, "config": "windows-release-x86_64", "r": "4.6.1", "check": "OK", "artifact": "11227583203" } ], "_host": "GitHub-Actions", "_buildurl": "https://github.com/r-universe/cran/actions/runs/37010943941", "_status": "success", "_upstream": "https://github.com/cran/timesift", "_commit": { "id": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "author": "Gilles Colling ", "committer": "cran-robot ", "message": "version 0.3.1\n", "time": 1790945597 }, "_maintainer": { "name": "Gilles Colling", "email": "gilles.colling051@gmail.com", "login": "gcol33", "linkedin": "in/gilles-colling-0b3747306", "orcid": "0000-0003-3070-6066", "twitter": "@Gilles__Colling", "description": "Building tools for ecology in R, C++, and Rust.\nWhy did Microsoft have to buy it. Sadge :(", "uuid": 25009600 }, "_distro": "resolute", "_registered": true, "_dependencies": [ { "package": "R", "version": ">= 4.1", "role": "Depends" }, { "package": "cpp11", "role": "LinkingTo" }, { "package": "grDevices", "role": "Imports" }, { "package": "graphics", "role": "Imports" }, { "package": "rlang", "role": "Imports" }, { "package": "stats", "role": "Imports" }, { "package": "tidyselect", "version": ">= 1.2.0", "role": "Imports" }, { "package": "tools", "role": "Imports" }, { "package": "utils", "role": "Imports" }, { "package": "testthat", "version": ">= 3.0.0", "role": "Suggests" }, { "package": "torch", "role": "Suggests" }, { "package": "glmnet", "role": "Suggests" }, { "package": "ranger", "role": "Suggests" }, { "package": "lme4", "role": "Suggests" }, { "package": "lmerTest", "role": "Suggests" }, { "package": "emmeans", "role": "Suggests" }, { "package": "knitr", "role": "Suggests" }, { "package": "rmarkdown", "role": "Suggests" }, { "package": "spelling", "role": "Suggests" }, { "package": "withr", "role": "Suggests" } ], "_owner": "cran", "_selfowned": false, "_usedby": 0, "_updates": [ { "week": "2026-40", "n": 1 } ], "_tags": [ { "name": "0.3.1", "date": "2026-10-02" } ], "_stars": 0, "_contributors": [ { "user": "gcol33", "count": 1, "uuid": 25009600 } ], "_userbio": { "uuid": 6899542, "type": "organization", "name": "cran", "followers": 630, "description": "Unofficial read-only mirror of all CRAN R packages" }, "_downloads": { "count": 0, "source": "https://cranlogs.r-pkg.org/downloads/total/last-month/timesift" }, "_devurl": "https://github.com/gcol33/timesift", "_pkgdown": "https://gillescolling.com/timesift/", "_topics": [ "cpp" ], "_rbuild": "4.6.1", "_assets": [ "extra/citation.cff", "extra/citation.html", "extra/citation.json", "extra/citation.txt", "extra/contents.json", "extra/manual.pdf", "extra/NEWS.html", "extra/NEWS.txt", "extra/readme.html", "extra/readme.md", "extra/timesift.html", "LICENSE" ], "_cranurl": false, "_exports": [ "all_of", "any_of", "as_sift", "average_precision", "bind_channels", "build_representation", "calendar_channels", "cnn", "contains", "coverage", "cv", "decision_threshold", "digest_array", "elasticnet", "ends_with", "ensemble", "ensemble_combine", "ensemble_fit", "ensemble_weights", "everything", "feature_matrix", "fit_learner", "fold_map", "forest", "grain", "grain_contrasts", "grain_ladder", "grain_matrix", "grains", "grouped_cv", "implied_skill", "kappa_score", "learner", "learners", "lookback", "lookback_matrix", "lookbacks", "matches", "metrics", "mlp", "model_agreement", "multigrain", "native", "occlusion", "paired_contrast", "positive_weights", "read_cells", "read_folds", "read_response", "register_learner", "register_metric", "register_response", "rescnn", "responses", "roc_auc", "scorable_cells", "score_predictions", "select_grain", "simulate_records", "starts_with", "stepwise", "timesift", "timesift_set", "train_control", "tss", "tss_inflation", "where", "write_cells", "write_folds", "write_response" ], "_help": [ { "page": "artifacts", "title": "Read and write the artifacts that cross the language boundary", "topics": [ "artifacts", "read_cells", "read_folds", "read_response", "write_cells", "write_folds", "write_response" ] }, { "page": "average_precision", "title": "Average precision", "topics": [ "average_precision" ] }, { "page": "bind_channels", "title": "Put channels side by side", "topics": [ "bind_channels" ] }, { "page": "build_representation", "title": "Build one representation for a set of targets", "topics": [ "build_representation" ] }, { "page": "calendar_channels", "title": "Where in the year, or the day, each bin sits", "topics": [ "calendar_channels" ] }, { "page": "combine", "title": "Combine learners, or representations, into a set", "topics": [ "c.timesift_learner", "c.timesift_models", "c.timesift_representation", "c.timesift_sift", "combine" ] }, { "page": "coverage", "title": "Which units reach which bins", "topics": [ "coverage" ] }, { "page": "cv", "title": "How the folds are drawn", "topics": [ "cv", "grouped_cv" ] }, { "page": "digest_array", "title": "The cross-language digest of a representation", "topics": [ "digest_array" ] }, { "page": "elasticnet", "title": "Penalised regression on the flattened representation", "topics": [ "elasticnet" ] }, { "page": "ensemble", "title": "How the candidates are combined", "topics": [ "ensemble" ] }, { "page": "ensemble_combine", "title": "Combine one prediction per member into one prediction", "topics": [ "ensemble_combine" ] }, { "page": "ensemble_fit", "title": "Fit the combiner on the out-of-fold predictions", "topics": [ "ensemble_fit" ] }, { "page": "ensemble_weights", "title": "The weights the combiner fitted", "topics": [ "ensemble_weights" ] }, { "page": "feature_matrix", "title": "Bring an already-reduced feature table into a ladder", "topics": [ "feature_matrix" ] }, { "page": "fit_learner", "title": "Fit one learner at one grain", "topics": [ "fit_learner", "predict.timesift_fit" ] }, { "page": "fold_map", "title": "Assign units to cross-validation folds", "topics": [ "fold_map" ] }, { "page": "forest", "title": "Random forest on the flattened representation", "topics": [ "forest" ] }, { "page": "grain_contrasts", "title": "Compare every grain against a learner's best one", "topics": [ "grain_contrasts" ] }, { "page": "grain_ladder", "title": "Fit at every grain and see where skill saturates", "topics": [ "grain_ladder", "summary.timesift_ladder" ] }, { "page": "grain_matrix", "title": "Reduce sensor series to a temporal grain", "topics": [ "grain_matrix" ] }, { "page": "grains", "title": "Several representations to run the same learners across", "topics": [ "as_sift", "grains", "lookbacks" ] }, { "page": "implied_skill", "title": "What population skill a reported level is consistent with", "topics": [ "implied_skill" ] }, { "page": "kappa_score", "title": "Cohen's kappa, and where two models disagree", "topics": [ "decision_threshold", "kappa_score", "model_agreement" ] }, { "page": "learner", "title": "Define a learner", "topics": [ "learner" ] }, { "page": "lookback_matrix", "title": "Reduce sensor series to a lookback anchored on each target", "topics": [ "lookback_matrix" ] }, { "page": "native", "title": "How a series becomes an array a learner reads", "topics": [ "grain", "lookback", "multigrain", "native" ] }, { "page": "occlusion", "title": "What part of the record a fitted model reads", "topics": [ "occlusion", "occlusion.default", "occlusion.timesift", "occlusion.timesift_ladder" ] }, { "page": "paired_contrast", "title": "Compare two arms cell by cell", "topics": [ "paired_contrast" ] }, { "page": "plot.timesift", "title": "Draw a run", "topics": [ "plot.timesift" ] }, { "page": "plot.timesift_ladder", "title": "Draw a ladder", "topics": [ "plot.timesift_ladder" ] }, { "page": "plot.timesift_selection", "title": "Draw how stable the choice of grain was", "topics": [ "plot.timesift_selection" ] }, { "page": "positive_weights", "title": "Case weights that balance a rare response", "topics": [ "positive_weights" ] }, { "page": "predict.timesift", "title": "Predict from a fitted timesift", "topics": [ "predict.timesift" ] }, { "page": "register_learner", "title": "Register a learner", "topics": [ "learners", "register_learner" ] }, { "page": "register_metric", "title": "Register a metric", "topics": [ "metrics", "register_metric" ] }, { "page": "register_response", "title": "Register a response head", "topics": [ "register_response", "responses" ] }, { "page": "roc_auc", "title": "The area under the ROC curve", "topics": [ "roc_auc" ] }, { "page": "scorable_cells", "title": "Which cells a score is defined on", "topics": [ "scorable_cells" ] }, { "page": "score_predictions", "title": "Score held-out predictions on the cells the mask allows", "topics": [ "score_predictions" ] }, { "page": "select_grain", "title": "Choose the grain inside the training data, and score the whole procedure", "topics": [ "select_grain", "summary.timesift_selection" ] }, { "page": "simulate_records", "title": "Simulate sensor records whose response acts at a known temporal grain", "topics": [ "simulate_records" ] }, { "page": "stepwise", "title": "Forward selection by AIC on the flattened representation", "topics": [ "stepwise" ] }, { "page": "timesift", "title": "Fit and compare representations of time-varying data", "topics": [ "timesift" ] }, { "page": "timesift_report", "title": "What a run found", "topics": [ "print.timesift", "print.timesift_summary", "summary.timesift", "timesift_report" ] }, { "page": "timesift_set", "title": "Several built representations of the same targets", "topics": [ "timesift_set" ] }, { "page": "torch_learners", "title": "Sequence encoders with a joint multi-label head", "topics": [ "cnn", "mlp", "rescnn", "torch_learners" ] }, { "page": "train_control", "title": "Training settings every neural learner reads", "topics": [ "train_control" ] }, { "page": "tss", "title": "The true skill statistic", "topics": [ "tss" ] }, { "page": "tss_inflation", "title": "How much a self-selected threshold inflates the reported level", "topics": [ "tss_inflation" ] } ], "_readme": "https://github.com/cran/timesift/raw/HEAD/README.md", "_rundeps": [ "cli", "cpp11", "glue", "lifecycle", "rlang", "tidyselect", "vctrs", "withr" ], "_sysdeps": [ { "shlib": "libstdc++", "package": "libstdc++6", "source": "gcc", "version": "16-20260322-1ubuntu1", "name": "c++", "homepage": "http://gcc.gnu.org/", "description": "GNU Standard C++ Library v3" } ], "_vignettes": [ { "source": "timesift.Rmd", "filename": "timesift.html", "title": "Choosing how a record is read", "engine": "knitr::rmarkdown", "headings": [ "Two tables", "One call", "Representations", "What a learner may be handed", "The split, and the cells a score is defined on", "Learners, and how they are trained", "The combination", "Reading a level honestly", "The arrays on their own", "An extreme day is not an extreme reading", "One grain at a time", "What was read" ], "created": "2026-10-02 12:53:17", "modified": "2026-10-02 12:53:17", "commits": 1 }, { "source": "reproducing-schrankogel.Rmd", "filename": "reproducing-schrankogel.html", "title": "Reproducing the Schrankogel grid", "engine": "knitr::rmarkdown", "headings": [ "What the reproduction needs", "The contract", "The split and the cells", "The seven grains", "The two readings of a grain", "The channels the encoders are given", "The arms", "The grid", "The procedure the demonstration reports", "The environment a fitted encoder carries", "The grain contrast", "Running it", "What the reproduction is checked against" ], "created": "2026-10-02 12:53:17", "modified": "2026-10-02 12:53:17", "commits": 1 } ], "_score": 3, "_indexed": false, "_nocasepkg": "timesift", "_universes": [ "cran", "gcol33" ], "_indexurl": "https://gcol33.r-universe.dev/timesift", "_binaries": [ { "r": "4.7.0", "os": "linux", "version": "0.3.1", "date": "2026-10-02T13:14:13.000Z", "distro": "resolute", "arch": "aarch64", "commit": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "fileid": "https://r2.ropensci.org/b39c79aa66e8de83f83fe85060125b6093d441ea6b9f2f5756e725dec3951d90", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/37010943941" }, { "r": "4.7.0", "os": "linux", "version": "0.3.1", "date": "2026-10-02T13:14:29.000Z", "distro": "resolute", "arch": "x86_64", "commit": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "fileid": "https://r2.ropensci.org/1c4cb7c127fb3aca01ab2c1b04cf3155d7d042a8cc25756f4a87b7b0026d6dde", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/37010943941" }, { "r": "4.6.1", "os": "linux", "version": "0.3.1", "date": "2026-10-02T13:14:27.000Z", "distro": "resolute", "arch": "aarch64", "commit": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "fileid": "https://r2.ropensci.org/b573828a27713fe390d8077d41c4119659f169a9227275381f3b8e7ec3bb427d", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/37010943941" }, { "r": "4.6.1", "os": "linux", "version": "0.3.1", "date": "2026-10-02T13:14:25.000Z", "distro": "resolute", "arch": "x86_64", "commit": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "fileid": "https://r2.ropensci.org/a8ded8a9f9582a97b0b8e8d69862e1957274a229505914a460cd43973f321680", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/37010943941" }, { "r": "4.6.0", "os": "wasm", "version": "0.3.1", "date": "2026-10-02T13:14:32.000Z", "arch": "emscripten", "commit": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "fileid": "https://r2.ropensci.org/693bd89570385c3b7fbf57f1c0eea8641b2fa02b26392cc9d6fa7b42f9216bea", "status": "success", "buildurl": "https://github.com/r-universe/cran/actions/runs/37010943941" }, { "r": "4.7.0", "os": "win", "version": "0.3.1", "date": "2026-10-02T13:15:18.000Z", "arch": "aarch64", "commit": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "fileid": "https://r2.ropensci.org/c6c8aa7a7cd3482c425c3db2d333f7d93069c13b24d4ee221c01c6c00b68b3f9", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/37010943941" }, { "r": "4.7.0", "os": "win", "version": "0.3.1", "date": "2026-10-02T13:13:24.000Z", "arch": "x86_64", "commit": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "fileid": "https://r2.ropensci.org/bb0860f40fec897971a002dd2cd1b2784554d83b691e94f9cf50829d4f0255e4", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/37010943941" }, { "r": "4.6.1", "os": "win", "version": "0.3.1", "date": "2026-10-02T13:15:34.000Z", "arch": "aarch64", "commit": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "fileid": "https://r2.ropensci.org/e8e3d1e6c68eb9749230f38dbeec33ad955ab31b4e28c5cb09a5533b3a0bf597", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/37010943941" }, { "r": "4.6.1", "os": "win", "version": "0.3.1", "date": "2026-10-02T13:13:14.000Z", "arch": "x86_64", "commit": "0918701a1ffe26cc8cc43fb4560971eaffe83e62", "fileid": "https://r2.ropensci.org/9d146a09f4d26f72bf44cf30c184de6072ed8e07f4cb814f51aefecda9f9e0f1", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/37010943941" } ] }