{ "_id": "6a1ea8b3b25058d4daad9344", "Package": "cito", "Type": "Package", "Title": "Building and Training Neural Networks", "Version": "1.1.1", "Authors@R": "c(person(given = \"Christian\",\nfamily = \"Amesöder\",\nrole = c(\"aut\"),\nemail = \"Christian.Amesoeder@informatik.uni-regensburg.de\"),\nperson(given = \"Maximilian\",\nfamily = \"Pichler\",\nrole = c(\"aut\", \"cre\"),\nemail = \"maximilian.pichler@biologie.uni-regensburg.de\",\ncomment = c(ORCID = \"0000-0003-2252-8327\")),\nperson(given = \"Florian\",\nfamily = \"Hartig\",\nrole = \"ctb\",\nemail = \"florian.hartig@biologie.uni-regensburg.de\",\ncomment = c(ORCID = \"0000-0002-6255-9059\")),\nperson(given = \"Armin\",\nfamily = \"Schenk\",\nrole = \"ctb\",\nemail = \"armin.schenk99@gmail.com\")\n)", "Description": "The 'cito' package provides a user-friendly interface for\ntraining and interpreting deep neural networks (DNN). 'cito'\nsimplifies the fitting of DNNs by supporting the familiar\nformula syntax, hyperparameter tuning under cross-validation,\nand helps to detect and handle convergence problems. DNNs can\nbe trained on CPU, GPU and MacOS GPUs. In addition, 'cito' has\nmany downstream functionalities such as various explainable AI\n(xAI) metrics (e.g. variable importance, partial dependence\nplots, accumulated local effect plots, and effect estimates) to\ninterpret trained DNNs. 'cito' optionally provides confidence\nintervals (and p-values) for all xAI metrics and predictions.\nAt the same time, 'cito' is computationally efficient because\nit is based on the deep learning framework 'torch'. The 'torch'\npackage is native to R, so no Python installation or other API\nis required for this package.", "Encoding": "UTF-8", "LazyData": "true", "Roxygen": "list(markdown = TRUE)", "RoxygenNote": "7.3.3", "License": "GPL (>= 3)", "VignetteBuilder": "knitr", "BugReports": "https://github.com/citoverse/cito/issues", "URL": "https://citoverse.github.io/cito/", "Language": "en-US", "Config/pak/sysreqs": "cmake make libjpeg-dev libpng-dev libtiff-dev\nlibuv1-dev", "Repository": "https://citoverse.r-universe.dev", "Date/Publication": "2026-03-04 08:55:34 UTC", "RemoteUrl": "https://github.com/citoverse/cito", "RemoteRef": "HEAD", "RemoteSha": "54d21232aab657a797cb5de0cc0c5ef49ccf28b0", "NeedsCompilation": "no", "Packaged": { "Date": "2026-06-02 09:31:28 UTC", "User": "root" }, "Author": "Christian Amesöder [aut],\nMaximilian Pichler [aut, cre] (ORCID:\n),\nFlorian Hartig [ctb] (ORCID: ),\nArmin Schenk [ctb]", "Maintainer": "Maximilian Pichler ", "MD5sum": "f7f7d625664c98bb12b1e21075e14c10", "_user": "citoverse", "_type": "src", "_file": "cito_1.1.1.tar.gz", "_fileid": "65afff73c53e7ff19a6d897b01e55e7401e8cc15ff9ad05eaa700e3420fcd029", "_filesize": 5602958, "_sha256": "65afff73c53e7ff19a6d897b01e55e7401e8cc15ff9ad05eaa700e3420fcd029", "_created": "2026-06-02T09:31:28.000Z", "_published": "2026-06-02T09:56:02.957Z", "_distro": "noble", "_jobs": [ { "job": 79041175978, "time": 182, "config": "linux-devel-x86_64", "r": "4.7.0", "check": "WARNING", "artifact": "7354102910" }, { "job": 79041176047, "time": 170, "config": "linux-release-x86_64", "r": "4.6.0", "check": "WARNING", "artifact": "7354098759" }, { "job": 79041175976, "time": 147, "config": "macos-oldrel-arm64", "r": "4.5.3", "check": "WARNING", "artifact": "7354083435" }, { "job": 79041175969, "time": 204, "config": "macos-release-arm64", "r": "4.6.0", "check": "WARNING", "artifact": "7354099755" }, { "job": 79040461114, "time": 236, "config": "source", "r": "4.6.0", "check": "OK", "artifact": "7354037031" }, { "job": 79041176038, "time": 1433, "config": "wasm-release", "r": "4.6.0", "check": "FAIL", "artifact": "" }, { "job": 79041175943, "time": 139, "config": "windows-devel", "r": "4.7.0", "check": "WARNING", "artifact": "7354087775" }, { "job": 79041175979, "time": 119, "config": "windows-oldrel", "r": "4.5.3", "check": "WARNING", "artifact": "7354082127" }, { "job": 79041176004, "time": 108, "config": "windows-release", "r": "4.6.0", "check": "WARNING", "artifact": "7354077262" } ], "_buildurl": "https://github.com/r-universe/citoverse/actions/runs/26810928244", "_status": "success", "_host": "GitHub-Actions", "_upstream": "https://github.com/citoverse/cito", "_commit": { "id": "54d21232aab657a797cb5de0cc0c5ef49ccf28b0", "author": "Maximilian Pichler ", "committer": "Maximilian Pichler ", "message": "Update\n\n- baseline loss, allow TRUE or FALSE\n- Calculate ce at the end of the training\n- Reduce number of permutation samples in importance\n", "time": 1772614534 }, "_maintainer": { "name": "Maximilian Pichler", "email": "maximilian.pichler@biologie.uni-regensburg.de", "login": "maximilianpi", "bluesky": "@maximilianpichler.bsky.social", "orcid": "0000-0003-2252-8327", "twitter": "@_Max_Pichler", "description": "Postdoc at the University of Regensburg", "uuid": 29951520 }, "_registered": true, "_dependencies": [ { "package": "R", "version": ">= 4.1.0", "role": "Depends" }, { "package": "coro", "role": "Imports" }, { "package": "checkmate", "role": "Imports" }, { "package": "torch", "role": "Imports" }, { "package": "gridExtra", "role": "Imports" }, { "package": "parabar", "role": "Imports" }, { "package": "abind", "role": "Imports" }, { "package": "progress", "role": "Imports" }, { "package": "cli", "role": "Imports" }, { "package": "torchvision", "role": "Imports" }, { "package": "tibble", "role": "Imports" }, { "package": "lme4", "role": "Imports" }, { "package": "rlang", "role": "Imports" }, { "package": "spelling", "role": "Suggests" }, { "package": "rmarkdown", "role": "Suggests" }, { "package": "testthat", "role": "Suggests" }, { "package": "plotly", "role": "Suggests" }, { "package": "ggraph", "role": "Suggests" }, { "package": "igraph", "role": "Suggests" }, { "package": "stats", "role": "Suggests" }, { "package": "ggplot2", "role": "Suggests" }, { "package": "knitr", "role": "Suggests" } ], "_owner": "citoverse", "_selfowned": true, "_usedby": 0, "_updates": [ { "week": "2025-29", "n": 4 }, { "week": "2025-30", "n": 3 }, { "week": "2025-34", "n": 1 }, { "week": "2025-35", "n": 1 }, { "week": "2025-37", "n": 1 }, { "week": "2025-42", "n": 1 }, { "week": "2025-43", "n": 1 }, { "week": "2025-44", "n": 1 }, { "week": "2026-10", "n": 1 } ], "_tags": [], "_topics": [ "machine-learning", "neural-network" ], "_stars": 54, "_contributors": [ { "user": "christianames", "count": 172, "uuid": 51171608 }, { "user": "arminschenk", "count": 160, "uuid": 65034337 }, { "user": "maximilianpi", "count": 120, "uuid": 29951520 }, { "user": "florianhartig", "count": 4, "uuid": 5457753 } ], "_userbio": { "uuid": 100435685, "type": "organization", "name": "citoverse" }, "_downloads": { "count": 4504, "source": "https://cranlogs.r-pkg.org/downloads/total/last-month/cito" }, "_devurl": "https://github.com/citoverse/cito", "_pkgdown": "https://citoverse.github.io/cito/", "_searchresults": 157, "_rbuild": "4.6.0", "_assets": [ "extra/citation.cff", "extra/citation.html", "extra/citation.json", "extra/citation.txt", "extra/cito.html", "extra/contents.json", "extra/NEWS.html", "extra/NEWS.txt", "extra/readme.html", "extra/readme.md", "manual.pdf" ], "_homeurl": "https://github.com/citoverse/cito", "_realowner": "citoverse", "_cranurl": true, "_releases": [ { "version": "1.0.0", "date": "2022-07-20" }, { "version": "1.0.1", "date": "2023-03-13" }, { "version": "1.0.2", "date": "2023-10-06" }, { "version": "1.1", "date": "2024-03-19" } ], "_exports": [ "ALE", "analyze_training", "avgPool", "cnn", "conditionalEffects", "config_lr_scheduler", "config_optimizer", "config_tuning", "continue_training", "conv", "create_architecture", "dnn", "e", "linear", "maxPool", "mmn", "PDP", "transfer", "tune" ], "_datasets": [ { "name": "tree_data", "title": "Tree Data", "object": "tree_data", "class": [ "list" ], "fields": [], "table": false, "tojson": true } ], "_help": [ { "page": "ALE", "title": "Accumulated Local Effect Plot (ALE)", "topics": [ "ALE", "ALE.citodnn", "ALE.citodnnBootstrap" ] }, { "page": "analyze_training", "title": "Visualize training of Neural Network", "topics": [ "analyze_training" ] }, { "page": "avgPool", "title": "This function creates an 'avgPool' layer object of class 'citolayer' for use in constructing a Convolutional Neural Network (CNN) architecture. The resulting layer object can be passed to the 'create_architecture' function to define the structure of the network.", "topics": [ "avgPool" ] }, { "page": "cito", "title": "'cito': Building and training neural networks", "topics": [ "cito-package", "cito" ] }, { "page": "cnn", "title": "Train a Convolutional Neural Network (CNN)", "topics": [ "cnn" ] }, { "page": "coef.citocnn", "title": "Retrieve parameters of a fitted CNN model", "topics": [ "coef.citocnn" ] }, { "page": "coef.citodnn", "title": "Returns list of parameters the neural network model currently has in use", "topics": [ "coef.citodnn", "coef.citodnnBootstrap" ] }, { "page": "coef.citommn", "title": "Retrieve parameters of a fitted MMN model", "topics": [ "coef.citommn" ] }, { "page": "conditionalEffects", "title": "Calculate average conditional effects", "topics": [ "conditionalEffects", "conditionalEffects.citodnn", "conditionalEffects.citodnnBootstrap" ] }, { "page": "config_lr_scheduler", "title": "Creation of customized learning rate scheduler objects", "topics": [ "config_lr_scheduler" ] }, { "page": "config_optimizer", "title": "Creation of customized optimizer objects", "topics": [ "config_optimizer" ] }, { "page": "config_tuning", "title": "Config hyperparameter tuning", "topics": [ "config_tuning" ] }, { "page": "continue_training", "title": "Continues training of a model generated with 'dnn', 'cnn' or 'mmn' for additional epochs.", "topics": [ "continue_training", "continue_training.citocnn", "continue_training.citodnn", "continue_training.citodnnBootstrap", "continue_training.citommn" ] }, { "page": "conv", "title": "Create a Convolutional Layer for a CNN Architecture", "topics": [ "conv" ] }, { "page": "create_architecture", "title": "Create a CNN Architecture", "topics": [ "create_architecture" ] }, { "page": "dnn", "title": "DNN", "topics": [ "dnn" ] }, { "page": "e", "title": "Embeddings", "topics": [ "e" ] }, { "page": "findReTrmClasses", "title": "list of specials - taken from enum.R", "topics": [ "findReTrmClasses" ] }, { "page": "linear", "title": "Create a Linear Layer for a CNN Architecture", "topics": [ "linear" ] }, { "page": "maxPool", "title": "Create a Maximum Pooling Layer for a CNN Architecture", "topics": [ "maxPool" ] }, { "page": "mmn", "title": "Train a Multi-Modal Neural Network (MMN)", "topics": [ "mmn" ] }, { "page": "PDP", "title": "Partial Dependence Plot (PDP)", "topics": [ "PDP", "PDP.citodnn", "PDP.citodnnBootstrap" ] }, { "page": "plot.citoarchitecture", "title": "Plot method for citoarchitecture objects", "topics": [ "plot.citoarchitecture" ] }, { "page": "plot.citocnn", "title": "Plot a fitted CNN model", "topics": [ "plot.citocnn" ] }, { "page": "plot.citodnn", "title": "Creates graph plot which gives an overview of the network architecture.", "topics": [ "plot.citodnn", "plot.citodnnBootstrap" ] }, { "page": "predict.citocnn", "title": "Predict with a fitted CNN model", "topics": [ "predict.citocnn" ] }, { "page": "predict.citodnn", "title": "Predict from a fitted dnn model", "topics": [ "predict.citodnn", "predict.citodnnBootstrap" ] }, { "page": "predict.citommn", "title": "Predict with a fitted MMN model", "topics": [ "predict.citommn" ] }, { "page": "print.citoarchitecture", "title": "Print method for citoarchitecture objects", "topics": [ "print.citoarchitecture" ] }, { "page": "print.citocnn", "title": "Print a fitted CNN model", "topics": [ "print.citocnn" ] }, { "page": "print.citodnn", "title": "Print class citodnn", "topics": [ "print.citodnn", "print.citodnnBootstrap" ] }, { "page": "print.citommn", "title": "Print a fitted MMN model", "topics": [ "print.citommn" ] }, { "page": "print.conditionalEffects", "title": "Print average conditional effects", "topics": [ "print.conditionalEffects", "print.conditionalEffectsBootstrap" ] }, { "page": "print.summary.citodnn", "title": "Print method for class summary.citodnn", "topics": [ "print.summary.citodnn", "print.summary.citodnnBootstrap" ] }, { "page": "residuals.citodnn", "title": "Extract Model Residuals", "topics": [ "residuals.citodnn" ] }, { "page": "simulate_shapes", "title": "Data Simulation for CNN", "topics": [ "simulate_shapes" ] }, { "page": "summary.citocnn", "title": "Summarize a fitted CNN model", "topics": [ "summary.citocnn" ] }, { "page": "summary.citodnn", "title": "Summarize Neural Network of class citodnn", "topics": [ "summary.citodnn", "summary.citodnnBootstrap" ] }, { "page": "summary.citommn", "title": "Summarize a fitted MMN model", "topics": [ "summary.citommn" ] }, { "page": "sumTerms", "title": "combine a list of formula terms as a sum", "topics": [ "sumTerms" ] }, { "page": "transfer", "title": "Include a Pretrained Model in a CNN Architecture", "topics": [ "transfer" ] }, { "page": "tree_data", "title": "Tree Data", "topics": [ "tree_data" ] }, { "page": "tune", "title": "Tune hyperparameter", "topics": [ "tune" ] } ], "_readme": "https://github.com/citoverse/cito/raw/HEAD/README.md", "_rundeps": [ "abind", "backports", "bit", "bit64", "boot", "callr", "checkmate", "cli", "coro", "crayon", "desc", "farver", "filelock", "fs", "glue", "gridExtra", "gtable", "hms", "jpeg", "jsonlite", "labeling", "lattice", "lifecycle", "lme4", "magrittr", "MASS", "Matrix", "minqa", "nlme", "nloptr", "parabar", "pillar", "pkgconfig", "png", "prettyunits", "processx", "progress", "ps", "R6", "rappdirs", "rbibutils", "RColorBrewer", "Rcpp", "RcppEigen", "Rdpack", "reformulas", "rlang", "safetensors", "scales", "tibble", "tiff", "torch", "torchvision", "utf8", "vctrs", "viridisLite", "withr", "zeallot" ], "_vignettes": [ { "source": "D-Advanced_custom_loss_functions.Rmd", "filename": "D-Advanced_custom_loss_functions.html", "title": "Advanced: Custom loss functions and prediction intervals", "author": "Maximilian Pichler", "engine": "knitr::rmarkdown", "headings": [ "Custom loss functions", "Example 1: Custom (likelihood/loss) functions", "Example 2: Quantile regression", "Example 3: Using cito for optimization / active learning" ], "created": "2023-09-29 08:45:03", "modified": "2024-03-05 13:02:25", "commits": 4 }, { "source": "E-CNN_and_MMN.Rmd", "filename": "E-CNN_and_MMN.html", "title": "Convultions neural networks and Multi modal neural networks", "author": "Maximilian Pichler", "engine": "knitr::rmarkdown", "headings": [ "Data Preparation of complex data", "The format of the inputs we expect in cito", "1. Preparing your data on disk", "2. Load images into R", "3. Normalize and check channel dimension", "4. Prepare tabular data (response and other tabular data such as altitude and spatial coordinates)", "Convolutional neural networks", "Multi-modal neural networks", "Computational Considerations and Constraints", "1. GPU Requirements", "2. System Memory (RAM)", "Example Calculation (500 observations)" ], "created": "2025-05-27 14:29:54", "modified": "2025-05-27 14:46:48", "commits": 2 }, { "source": "C-Example_Species_distribution_modeling.Rmd", "filename": "C-Example_Species_distribution_modeling.html", "title": "Example: (Multi-) Species distribution models with cito", "author": "Maximilian Pichler", "engine": "knitr::rmarkdown", "headings": [ "Species distribution model - African elephant", "Adjusting optimization parameters - Convergence", "Train final model with bootstrapping to obtain uncertainties", "Predictions", "Inference", "Multi-species distribution model", "Train model with bootstrapping", "Advanced: Joint species distribution model" ], "created": "2023-09-29 08:45:03", "modified": "2024-03-18 17:04:38", "commits": 8 }, { "source": "A-Introduction_to_cito.Rmd", "filename": "A-Introduction_to_cito.html", "title": "Introduction to cito", "author": "Christian Amesoeder & Maximilian Pichler", "engine": "knitr::rmarkdown", "headings": [ "Setup - Installing torch", "Details about the internal implementation of ‘cito’", "Internal Data Representation", "Construction of the neural networks", "Training and Evaluation", "Transferability", "Introduction to models and model structures", "Loss functions / Likelihoods", "Data", "Fitting a simple model", "Baseline loss", "Adding a validation set to the training process", "Methods", "Explainable AI - Understanding your model", "Uncertainties/p-Values", "Architecture", "Activation functions", "Tuning hyperparameters", "Regularization", "Elastic net regularization", "Dropout Regularization", "Learning rate", "Learning rate scheduler", "Optimizer", "Early Stopping", "Automatic hyperparameter tuning (experimental)", "Continue training process", "The best of both worlds - combining statistical models and deep learning" ], "created": "2023-09-29 08:45:03", "modified": "2024-03-16 13:50:47", "commits": 8 }, { "source": "B-Training_neural_networks.Rmd", "filename": "B-Training_neural_networks.html", "title": "Training neural networks", "author": "Maximilian Pichler", "engine": "knitr::rmarkdown", "headings": [ "Possible issues", "Convergence issues", "Epochs", "Learning rate", "Solution: learning rate scheduler", "Overfitting", "Early stopping and regularization" ], "created": "2023-09-29 08:45:03", "modified": "2024-03-06 13:11:47", "commits": 5 } ], "_score": 9.411195566906517, "_indexed": true, "_nocasepkg": "cito", "_universes": [ "citoverse", "maximilianpi" ], "_binaries": [ { "r": "4.6.0", "os": "wasm", "version": "1.1.1", "date": "2026-05-22T18:37:43.000Z", "commit": "54d21232aab657a797cb5de0cc0c5ef49ccf28b0", "fileid": "dcb700bc8e5d8239abe5a0b1884cc044e0e377e84b1ca9a5cb1c163e20f8dd29", "status": "success", "buildurl": "https://github.com/r-universe/citoverse/actions/runs/25283648276" }, { "r": "4.7.0", "os": "linux", "version": "1.1.1", "date": "2026-06-02T09:34:20.000Z", "distro": "noble", "commit": "54d21232aab657a797cb5de0cc0c5ef49ccf28b0", "fileid": "cde1781ccd01965254f6f3b4f88b25ebc6f115d4b5c0a0f78c0f2cd8e5ad3b7f", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/citoverse/actions/runs/26810928244" }, { "r": "4.6.0", "os": "linux", "version": "1.1.1", "date": "2026-06-02T09:34:07.000Z", "distro": "noble", "commit": "54d21232aab657a797cb5de0cc0c5ef49ccf28b0", "fileid": "1c63ab0211cb55822e4d26ded7f93da337e769892dd7249e77700bd8de1c8a11", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/citoverse/actions/runs/26810928244" }, { "r": "4.5.3", "os": "mac", "version": "1.1.1", "date": "2026-06-02T09:33:38.000Z", "commit": "54d21232aab657a797cb5de0cc0c5ef49ccf28b0", "fileid": "5b2b38d936b69aecc053a955c46895fa55ba078bdeadcda972cb768b622dbb41", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/citoverse/actions/runs/26810928244" }, { "r": "4.6.0", "os": "mac", "version": "1.1.1", "date": "2026-06-02T09:34:17.000Z", "commit": "54d21232aab657a797cb5de0cc0c5ef49ccf28b0", "fileid": "5e8351e1f67ec829b991dd75f91453cf705d22fce303801d4bed4b1c7638507c", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/citoverse/actions/runs/26810928244" }, { "r": "4.7.0", "os": "win", "version": "1.1.1", "date": "2026-06-02T09:33:27.000Z", "commit": "54d21232aab657a797cb5de0cc0c5ef49ccf28b0", "fileid": "d4decbda68be0b890b76ec9d356d9ada5ef8ccbc7d2c138cdf2587901e215915", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/citoverse/actions/runs/26810928244" }, { "r": "4.5.3", "os": "win", "version": "1.1.1", "date": "2026-06-02T09:33:06.000Z", "commit": "54d21232aab657a797cb5de0cc0c5ef49ccf28b0", "fileid": "98d4ebdedfafc2e0c7b07d7597522ad390bb0d09c72ff53ffe0c5e0712324a0c", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/citoverse/actions/runs/26810928244" }, { "r": "4.6.0", "os": "win", "version": "1.1.1", "date": "2026-06-02T09:32:54.000Z", "commit": "54d21232aab657a797cb5de0cc0c5ef49ccf28b0", "fileid": "aa2ac1d7180e4ba2f465752707832a49180b5f7854c4ed657f4eb2e57de307b8", "status": "success", "check": "WARNING", "buildurl": "https://github.com/r-universe/citoverse/actions/runs/26810928244" } ] }