{
  "_id": "6b00be07304827bb829a2e2c",
  "Package": "thangldKANr",
  "Type": "Package",
  "Title": "Kolmogorov-Arnold Networks with Interchangeable Edge Bases Using\n'torch'",
  "Version": "0.1.0",
  "Authors@R": "person(given = \"Thang\", family = \"Le-Dinh\",\nemail = \"thang87sg@gmail.com\",\nrole = c(\"aut\", \"cre\", \"cph\"))",
  "Description": "Implements Kolmogorov-Arnold Networks (KAN) in R on top of\nthe 'torch' package (no 'Python' needed) with one network\nengine and interchangeable edge bases: Chebyshev, Legendre and\nJacobi polynomials, Gaussian radial basis functions (FastKAN),\nMexican hat wavelets (Bozorgasl and Chen (2024)\n<doi:10.48550/arXiv.2405.12832>), safe Pade rational functions\n(Molina, Schramowski and Kersting (2020)\n<doi:10.48550/arXiv.1907.06732>), a hybrid of SiLU, Chebyshev\nand radial basis parts, and squared rectified linear unit bumps\n(Qiu et al. (2024) <doi:10.48550/arXiv.2406.02075>). Each layer\nmay use a different basis, and further bases can be registered\nby other packages. Besides tabular regression, the same layers\npower graph KANs for node regression (neighbourhood aggregation\nwith the normalized adjacency matrix, as in Kiamari, Kiamari\nand Krishnamachari (2024) <doi:10.48550/arXiv.2406.06470>) and\nlag-window KANs for one-step time series forecasting. Provides\nmulti-layer networks with Adam training, training-only\nstandardization, validation-based early stopping, reproducible\nseeds, edge-function evaluation and plots, and model\npersistence. A base R reference implementation with analytic\ngradients serves as a numerical oracle and as a training\nbackend that does not need 'libtorch'. The network follows Liu\net al. (2024) <doi:10.48550/arXiv.2404.19756>.",
  "License": "GPL-3",
  "Encoding": "UTF-8",
  "Language": "en-US",
  "Config/testthat/edition": "3",
  "VignetteBuilder": "knitr",
  "URL": "https://github.com/Thangpro9999/thangldKANr",
  "BugReports": "https://github.com/Thangpro9999/thangldKANr/issues",
  "Config/roxygen2/version": "8.1.0",
  "NeedsCompilation": "no",
  "Packaged": {
    "Date": "2026-11-20 21:36:05 UTC",
    "User": "root"
  },
  "Author": "Thang Le-Dinh [aut, cre, cph]",
  "Maintainer": "Thang Le-Dinh <thang87sg@gmail.com>",
  "Repository": "https://cran.r-universe.dev",
  "Date/Publication": "2026-11-20 13:10:02 UTC",
  "RemoteUrl": "https://github.com/cran/thangldKANr",
  "RemoteRef": "HEAD",
  "RemoteSha": "3d951cc797648d707e4598e689da37b770c01d45",
  "_user": "cran",
  "_type": "src",
  "_file": "thangldKANr_0.1.0.tar.gz",
  "_fileid": "https://r2.ropensci.org/10a2403d7d0c9f4ebb77635a100c6d68e88d57111f4f0d84cc518b3dabd19650",
  "_filesize": 921758,
  "_sha256": "10a2403d7d0c9f4ebb77635a100c6d68e88d57111f4f0d84cc518b3dabd19650",
  "_expires": "2027-02-28T21:39:17.000Z",
  "_created": "2026-11-20T21:36:05.000Z",
  "_published": "2026-11-20T21:39:19.779Z",
  "_upstream": "https://github.com/cran/thangldKANr",
  "_status": "success",
  "_maintainer": {
    "name": "Thang Le-Dinh",
    "email": "thang87sg@gmail.com",
    "login": "thangpro9999",
    "description": "PhD in Statistics (Business Applications)",
    "uuid": 266899457
  },
  "_jobs": [
    {
      "job": 136584359888,
      "time": 147,
      "config": "linux-devel-x86_64",
      "r": "4.7.0",
      "check": "OK",
      "artifact": "14671888452"
    },
    {
      "job": 136584359937,
      "time": 148,
      "config": "linux-release-x86_64",
      "r": "4.6.2",
      "check": "OK",
      "artifact": "14672033305"
    },
    {
      "job": 136582320197,
      "time": 236,
      "config": "source",
      "r": "4.6.2",
      "check": "OK",
      "artifact": "14672396459"
    },
    {
      "job": 136584359805,
      "time": 98,
      "config": "wasm-release",
      "r": "4.6.0",
      "check": "OK",
      "artifact": "14672830092"
    },
    {
      "job": 136584359879,
      "time": 113,
      "config": "windows-devel-x86_64",
      "r": "4.7.0",
      "check": "OK",
      "artifact": "14672511593"
    },
    {
      "job": 136584359885,
      "time": 121,
      "config": "windows-release-x86_64",
      "r": "4.6.2",
      "check": "OK",
      "artifact": "14671429598"
    }
  ],
  "_buildurl": "https://github.com/r-universe/cran/actions/runs/44820989457",
  "_commit": {
    "id": "3d951cc797648d707e4598e689da37b770c01d45",
    "author": "Thang Le-Dinh <thang87sg@gmail.com>",
    "committer": "cran-robot <csardi.gabor+cran@gmail.com>",
    "message": "version 0.1.0\n",
    "time": 1795180202
  },
  "_distro": "resolute",
  "_host": "GitHub-Actions",
  "_registered": true,
  "_dependencies": [
    {
      "package": "R",
      "version": ">= 4.1.0",
      "role": "Depends"
    },
    {
      "package": "graphics",
      "role": "Imports"
    },
    {
      "package": "stats",
      "role": "Imports"
    },
    {
      "package": "torch",
      "version": ">= 0.13.0",
      "role": "Imports"
    },
    {
      "package": "utils",
      "role": "Imports"
    },
    {
      "package": "knitr",
      "role": "Suggests"
    },
    {
      "package": "rmarkdown",
      "role": "Suggests"
    },
    {
      "package": "testthat",
      "version": ">= 3.0.0",
      "role": "Suggests"
    }
  ],
  "_owner": "cran",
  "_selfowned": false,
  "_usedby": 0,
  "_updates": [
    {
      "week": "2026-47",
      "n": 1
    }
  ],
  "_tags": [
    {
      "name": "0.1.0",
      "date": "2026-11-20"
    }
  ],
  "_stars": 0,
  "_contributors": [
    {
      "user": "thangpro9999",
      "count": 1,
      "uuid": 266899457
    }
  ],
  "_userbio": {
    "uuid": 6899542,
    "type": "organization",
    "name": "cran",
    "followers": 636,
    "description": "Unofficial read-only mirror of all CRAN R packages"
  },
  "_downloads": {
    "count": 0,
    "source": "https://cranlogs.r-pkg.org/downloads/total/last-month/thangldKANr"
  },
  "_devurl": "https://github.com/thangpro9999/thangldkanr",
  "_rbuild": "4.6.2",
  "_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/thangldKANr.html"
  ],
  "_cranurl": false,
  "_releases": [
    {
      "version": "0.1.0",
      "date": "2026-11-20"
    }
  ],
  "_authors": [
    {
      "name": "Thang Le-Dinh",
      "email": "thang87sg@gmail.com",
      "role": [
        "aut",
        "cre",
        "cph"
      ]
    }
  ],
  "_exports": [
    "chebyshev_basis",
    "chebyshev_basis_deriv",
    "hybrid_basis",
    "hybrid_basis_deriv",
    "hybrid_kan_components",
    "jacobi_basis",
    "jacobi_basis_deriv",
    "kan_bases",
    "kan_layer_bases",
    "legendre_basis",
    "legendre_basis_deriv",
    "mexican_hat",
    "mexican_hat_deriv",
    "rational",
    "rational_deriv",
    "rbf_basis",
    "rbf_basis_deriv",
    "rbf_grid",
    "register_kan_basis",
    "relu_basis",
    "relu_basis_deriv",
    "thangld_graph_kan",
    "thangld_graph_kan_adjacency",
    "thangld_graph_kan_fd_gradient_r",
    "thangld_graph_kan_fit",
    "thangld_graph_kan_forward_r",
    "thangld_graph_kan_gradient_r",
    "thangld_graph_kan_load",
    "thangld_graph_kan_metrics",
    "thangld_graph_kan_save",
    "thangld_graph_kan_simulate",
    "thangld_graph_kan_validate",
    "thangld_kan",
    "thangld_kan_edge",
    "thangld_kan_fd_gradient_r",
    "thangld_kan_fit",
    "thangld_kan_forward_r",
    "thangld_kan_gradient_r",
    "thangld_kan_load",
    "thangld_kan_metrics",
    "thangld_kan_save",
    "thangld_kan_validate",
    "thangld_temporal_kan",
    "thangld_temporal_kan_embed",
    "thangld_temporal_kan_fit",
    "thangld_temporal_kan_load",
    "thangld_temporal_kan_rolling",
    "thangld_temporal_kan_save",
    "thangld_temporal_kan_validate",
    "unregister_kan_basis"
  ],
  "_help": [
    {
      "page": "chebyshev_basis",
      "title": "Chebyshev polynomials and their derivatives",
      "topics": [
        "chebyshev_basis",
        "chebyshev_basis_deriv"
      ]
    },
    {
      "page": "hybrid_basis",
      "title": "Basis functions of a hybrid KAN edge",
      "topics": [
        "hybrid_basis",
        "hybrid_basis_deriv"
      ]
    },
    {
      "page": "hybrid_kan_components",
      "title": "Size of each basis part on data",
      "topics": [
        "hybrid_kan_components"
      ]
    },
    {
      "page": "jacobi_basis",
      "title": "Jacobi polynomials and their derivatives",
      "topics": [
        "jacobi_basis",
        "jacobi_basis_deriv"
      ]
    },
    {
      "page": "kan_bases",
      "title": "List the available edge bases",
      "topics": [
        "kan_bases"
      ]
    },
    {
      "page": "kan_layer_bases",
      "title": "Bases used by the layers of a model",
      "topics": [
        "kan_layer_bases"
      ]
    },
    {
      "page": "legendre_basis",
      "title": "Legendre polynomials and their derivatives",
      "topics": [
        "legendre_basis",
        "legendre_basis_deriv"
      ]
    },
    {
      "page": "mexican_hat",
      "title": "Mexican hat (Ricker) wavelet",
      "topics": [
        "mexican_hat",
        "mexican_hat_deriv"
      ]
    },
    {
      "page": "plot.thangld_kan",
      "title": "Plot a Kolmogorov-Arnold network",
      "topics": [
        "plot.thangld_kan"
      ]
    },
    {
      "page": "predict.thangld_graph_kan",
      "title": "Predict node targets",
      "topics": [
        "predict.thangld_graph_kan"
      ]
    },
    {
      "page": "predict.thangld_kan",
      "title": "Predict with a fitted Kolmogorov-Arnold network",
      "topics": [
        "predict.thangld_kan"
      ]
    },
    {
      "page": "predict.thangld_temporal_kan",
      "title": "One-step-ahead forecast",
      "topics": [
        "predict.thangld_temporal_kan"
      ]
    },
    {
      "page": "rational",
      "title": "Safe Pade rational function",
      "topics": [
        "rational",
        "rational_deriv"
      ]
    },
    {
      "page": "rbf_basis",
      "title": "Gaussian radial basis functions",
      "topics": [
        "rbf_basis",
        "rbf_basis_deriv",
        "rbf_grid"
      ]
    },
    {
      "page": "register_kan_basis",
      "title": "Register an additional edge basis",
      "topics": [
        "register_kan_basis",
        "unregister_kan_basis"
      ]
    },
    {
      "page": "relu_basis",
      "title": "ReLU-KAN basis functions and their derivatives",
      "topics": [
        "relu_basis",
        "relu_basis_deriv"
      ]
    },
    {
      "page": "thangld_graph_kan",
      "title": "Create a graph KAN architecture",
      "topics": [
        "thangld_graph_kan"
      ]
    },
    {
      "page": "thangld_graph_kan_adjacency",
      "title": "Normalized adjacency matrix of an undirected graph",
      "topics": [
        "thangld_graph_kan_adjacency"
      ]
    },
    {
      "page": "thangld_graph_kan_fit",
      "title": "Fit a graph KAN for node regression",
      "topics": [
        "thangld_graph_kan_fit"
      ]
    },
    {
      "page": "thangld_graph_kan_forward_r",
      "title": "Forward pass and gradients in base R",
      "topics": [
        "thangld_graph_kan_fd_gradient_r",
        "thangld_graph_kan_forward_r",
        "thangld_graph_kan_gradient_r"
      ]
    },
    {
      "page": "thangld_graph_kan_metrics",
      "title": "Regression metrics",
      "topics": [
        "thangld_graph_kan_metrics"
      ]
    },
    {
      "page": "thangld_graph_kan_save",
      "title": "Save and load a graph KAN",
      "topics": [
        "thangld_graph_kan_load",
        "thangld_graph_kan_save"
      ]
    },
    {
      "page": "thangld_graph_kan_simulate",
      "title": "Simulate a stochastic block model graph with node features",
      "topics": [
        "thangld_graph_kan_simulate"
      ]
    },
    {
      "page": "thangld_graph_kan_validate",
      "title": "Run the numerical validation suite",
      "topics": [
        "thangld_graph_kan_validate"
      ]
    },
    {
      "page": "thangld_kan",
      "title": "Create a Kolmogorov-Arnold network with a chosen edge basis",
      "topics": [
        "thangld_kan"
      ]
    },
    {
      "page": "thangld_kan_edge",
      "title": "Evaluate a learned edge function",
      "topics": [
        "thangld_kan_edge"
      ]
    },
    {
      "page": "thangld_kan_fit",
      "title": "Fit a Kolmogorov-Arnold network for regression",
      "topics": [
        "thangld_kan_fit"
      ]
    },
    {
      "page": "thangld_kan_forward_r",
      "title": "Forward pass and gradients in base R (numerical oracle)",
      "topics": [
        "thangld_kan_fd_gradient_r",
        "thangld_kan_forward_r",
        "thangld_kan_gradient_r"
      ]
    },
    {
      "page": "thangld_kan_metrics",
      "title": "Regression metrics",
      "topics": [
        "thangld_kan_metrics"
      ]
    },
    {
      "page": "thangld_kan_save",
      "title": "Save and load a Kolmogorov-Arnold network",
      "topics": [
        "thangld_kan_load",
        "thangld_kan_save"
      ]
    },
    {
      "page": "thangld_kan_validate",
      "title": "Run the numerical validation suite",
      "topics": [
        "thangld_kan_validate"
      ]
    },
    {
      "page": "thangld_temporal_kan",
      "title": "Create a lag-window Temporal KAN",
      "topics": [
        "thangld_temporal_kan"
      ]
    },
    {
      "page": "thangld_temporal_kan_embed",
      "title": "Lag-window design matrix",
      "topics": [
        "thangld_temporal_kan_embed"
      ]
    },
    {
      "page": "thangld_temporal_kan_fit",
      "title": "Fit a Temporal KAN on a series",
      "topics": [
        "thangld_temporal_kan_fit"
      ]
    },
    {
      "page": "thangld_temporal_kan_rolling",
      "title": "Rolling-origin one-step evaluation",
      "topics": [
        "thangld_temporal_kan_rolling"
      ]
    },
    {
      "page": "thangld_temporal_kan_save",
      "title": "Save and load a Temporal KAN",
      "topics": [
        "thangld_temporal_kan_load",
        "thangld_temporal_kan_save"
      ]
    },
    {
      "page": "thangld_temporal_kan_validate",
      "title": "Run the numerical validation suite",
      "topics": [
        "thangld_temporal_kan_validate"
      ]
    }
  ],
  "_readme": "https://github.com/cran/thangldKANr/raw/HEAD/README.md",
  "_rundeps": [
    "bit",
    "bit64",
    "callr",
    "cli",
    "coro",
    "desc",
    "farver",
    "glue",
    "jsonlite",
    "labeling",
    "lifecycle",
    "magrittr",
    "otel",
    "processx",
    "ps",
    "R6",
    "RColorBrewer",
    "Rcpp",
    "rlang",
    "safetensors",
    "scales",
    "torch",
    "viridisLite",
    "withr"
  ],
  "_vignettes": [
    {
      "source": "thangldKANr.Rmd",
      "filename": "thangldKANr.html",
      "title": "Kolmogorov-Arnold networks with interchangeable edge bases",
      "engine": "knitr::rmarkdown",
      "headings": [
        "One engine, many edge types",
        "Fitting and comparing bases",
        "A different basis in each layer",
        "Adding a basis",
        "Persistence and numerical checks"
      ],
      "created": "2026-11-20 13:10:02",
      "modified": "2026-11-20 13:10:02",
      "commits": 1
    },
    {
      "source": "temporal.Rmd",
      "filename": "temporal.html",
      "title": "Lag-window KAN forecasting with thangldKANr",
      "engine": "knitr::rmarkdown",
      "headings": [
        "Model",
        "Leakage rules",
        "Example: Canadian lynx (log scale)",
        "Rolling-origin evaluation against baselines",
        "Validation"
      ],
      "created": "2026-11-20 13:10:02",
      "modified": "2026-11-20 13:10:02",
      "commits": 1
    },
    {
      "source": "graph.Rmd",
      "filename": "graph.html",
      "title": "Node regression with graph KANs",
      "engine": "knitr::rmarkdown",
      "headings": [
        "Model",
        "A simulated graph",
        "Deeper networks and the self branch",
        "Cautions",
        "Validation"
      ],
      "created": "2026-11-20 13:10:02",
      "modified": "2026-11-20 13:10:02",
      "commits": 1
    }
  ],
  "_score": 3.1760912590556813,
  "_indexed": true,
  "_nocasepkg": "thangldkanr",
  "_universes": [
    "cran",
    "thangpro9999"
  ],
  "_binaries": [
    {
      "r": "4.7.0",
      "os": "linux",
      "version": "0.1.0",
      "date": "2026-11-20T21:38:21.000Z",
      "distro": "resolute",
      "commit": "3d951cc797648d707e4598e689da37b770c01d45",
      "fileid": "https://r2.ropensci.org/9aa04e99b6a291b526884c18880d35a45811344c4fc00471e76ae82ef861173a",
      "status": "success",
      "check": "OK",
      "buildurl": "https://github.com/r-universe/cran/actions/runs/44820989457"
    },
    {
      "r": "4.6.2",
      "os": "linux",
      "version": "0.1.0",
      "date": "2026-11-20T21:38:11.000Z",
      "distro": "resolute",
      "commit": "3d951cc797648d707e4598e689da37b770c01d45",
      "fileid": "https://r2.ropensci.org/c51f94134c40bbaa6048febdfa046b4e624fa0c4a64946b156aef9087b3b30a6",
      "status": "success",
      "check": "OK",
      "buildurl": "https://github.com/r-universe/cran/actions/runs/44820989457"
    },
    {
      "r": "4.6.0",
      "os": "wasm",
      "version": "0.1.0",
      "date": "2026-11-20T21:38:08.000Z",
      "commit": "3d951cc797648d707e4598e689da37b770c01d45",
      "fileid": "https://r2.ropensci.org/204548aacba1f9a1df72cffedc838bf79d80f85cdfb44fefdb6a0d30dddfbc81",
      "status": "success",
      "buildurl": "https://github.com/r-universe/cran/actions/runs/44820989457"
    },
    {
      "r": "4.7.0",
      "os": "win",
      "version": "0.1.0",
      "date": "2026-11-20T21:37:19.000Z",
      "commit": "3d951cc797648d707e4598e689da37b770c01d45",
      "fileid": "https://r2.ropensci.org/51212fefd20dbbc77385071dbd26c913e1fc8b4aa093232bf6b7463857e190e4",
      "status": "success",
      "check": "OK",
      "buildurl": "https://github.com/r-universe/cran/actions/runs/44820989457"
    },
    {
      "r": "4.6.2",
      "os": "win",
      "version": "0.1.0",
      "date": "2026-11-20T21:37:26.000Z",
      "commit": "3d951cc797648d707e4598e689da37b770c01d45",
      "fileid": "https://r2.ropensci.org/d2880272b5f0ba0759c55417b787db5184b7c59d4823a3d96d1c4ec2ac2e6544",
      "status": "success",
      "check": "OK",
      "buildurl": "https://github.com/r-universe/cran/actions/runs/44820989457"
    }
  ]
}