Choose a starting point | A first regression fit | Prediction and reusable fits | Refit and extract smoother matrices | Function catalog | Shared model operations | Local regression and prediction synchronization | Model assessment and uncertainty | Lifting trend filters | Graph spectral smoothing and graph trend filtering | From a constructed graph to a fitted response | Hessian-energy regression | Harmonic interpolation | Probability mass and subject occupation | Experimental transport methods | Synthetic datasets: specifications, responses, and reproducibility | Geometry constructors and helpers supplied by dgraphs | Sampling constructors supplied by dgraphs | Finding detailed help | Advanced controls without a new fitter
Package: geosmooth 0.2.0
geosmooth: Geometric Methods for Nonparametric Regression and Density Estimation
Provides geometric methods for nonparametric regression and density estimation on data represented as coordinate matrices or weighted graphs. Methods include local polynomial smoothing, model-averaged local polynomial smoothing, local polynomial lifting trend filtering, synchronized local polynomial lifting trend filtering, graph low-pass filtering, and Hessian-energy regression. Methodological references include Gajer and Ravel (2025) "Adaptive Geometric Regression for High-Dimensional Structured Data" <doi:10.48550/arXiv.2511.03817>, Fan and Gijbels (1996, ISBN:9780412983214), Wang et al. (2016) "Trend Filtering on Graphs" <https://www.jmlr.org/papers/v17/15-147.html>, and Kim et al. (2009) "Semi-Supervised Regression Using Hessian Energy" <https://papers.nips.cc/paper/3741-semi-supervised-regression-using-hessian-energy-with-an-application-to-semi-supervised-dimensionality-reduction>.
Authors:
geosmooth_0.2.0.tar.gz
geosmooth_0.2.0.zip(r-4.7-x86_64)geosmooth_0.2.0.zip(r-4.7-arm64)geosmooth_0.2.0.zip(r-4.6-x86_64)geosmooth_0.2.0.zip(r-4.6-arm64)geosmooth_0.2.0.zip(r-4.5-x86_64)
geosmooth_0.2.0.tgz(r-4.6-x86_64)geosmooth_0.2.0.tgz(r-4.6-arm64)geosmooth_0.2.0.tgz(r-4.5-x86_64)geosmooth_0.2.0.tgz(r-4.5-arm64)
geosmooth_0.2.0.tar.gz(r-4.7-arm64)geosmooth_0.2.0.tar.gz(r-4.7-x86_64)geosmooth_0.2.0.tar.gz(r-4.6-arm64)geosmooth_0.2.0.tar.gz(r-4.6-x86_64)
geosmooth_0.2.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
geosmooth/json (API)
| # Install 'geosmooth' in R: |
| install.packages('geosmooth', repos = c('https://pgajer.r-universe.dev', 'https://cloud.r-project.org')) |
Bug tracker:https://github.com/pgajer/geosmooth/issues
Pkgdown/docs site:https://pgajer.github.io
Last updated from:5e8959743d. Checks:15 OK. Indexed: yes.
| Target | Result | Time | Files | Syslog |
|---|---|---|---|---|
| linux-devel-arm64 | OK | 363 | ||
| linux-devel-x86_64 | OK | 442 | ||
| source / vignettes | OK | 415 | ||
| linux-release-arm64 | OK | 370 | ||
| linux-release-x86_64 | OK | 426 | ||
| macos-release-arm64 | OK | 360 | ||
| macos-release-x86_64 | OK | 1081 | ||
| macos-oldrel-arm64 | OK | 339 | ||
| macos-oldrel-x86_64 | OK | 1112 | ||
| windows-devel-arm64 | OK | 582 | ||
| windows-devel-x86_64 | OK | 526 | ||
| windows-release-arm64 | OK | 588 | ||
| windows-release-x86_64 | OK | 499 | ||
| windows-oldrel-x86_64 | OK | 534 | ||
| wasm-release | OK | 200 |
Exports:apply.metric.graph.lowpass.pathbootstrap.malpscompare.masked.graph.fieldscompare.synthetic.datasetcv.masked.graph.fieldfit.chart.kernelfit.densityfit.graph.trend.filteringfit.local.likelihoodfit.lpl.tffit.lpsfit.malpsfit.masked.graph.fieldfit.metric.graph.lowpassfit.ps.lpsfit.pttf.trend.filteringfit.slpl.tffit.ssrhe.hessian.l1.regressionfit.ssrhe.hessian.regressionfit.subject.odget.region.boundarygraph.trend.filtering.operatorharmonic.smootherlpl.tf.operatorlps.backend.diagnosticslps.grouped.foldidlps.nested.cvlps.pointwise.bandmalps.gcvmaterialize.syntheticmaterialize.synthetic.instancemetric.graph.heat.eta.gridmetric.graph.heat.extend.lowermetric.graph.lowpass.basismetric.graph.lowpass.operatornormalize.densityperform.harmonic.smoothingpttf.geometrypttf.operatorpttf.operator.filter.rowsrefitselect.masked.graph.fieldslpl.tf.operatorsmoother.matrixssrhe.hessian.operatorssrhe.support.gridsupport.masked.graph.fieldsynthetic.dataset.checksumsynthetic.registry.idssynthetic.registry.seedsynthetic.registry.specsynthetic.response.bernoullisynthetic.response.clustered.gaussiansynthetic.response.gaussiansynthetic.response.heteroskedastic.gaussiansynthetic.response.laplace.outliersynthetic.sampling.stratifiedsynthetic.specsynthetic.truth.gaussian.mixturesynthetic.truth.logitsynthetic.truth.namedsynthetic.truth.occupation.mixturesynthetic.truth.polynomialtransported.graph.hessian.operatorvalidate.synthetic.dataset
Dependencies:BHclicpp11dgraphsdigestFNNglueigraphjsonlitelatticelifecyclemagrittrMASSMatrixnloptrpkgconfigRcppRcppEigenrlangvctrs
Finding your way around geosmooth
Last update: 2026-09-15
Started: 2026-09-14
Synthetic datasets for geometric smoothing
Last update: 2026-09-15
Started: 2026-07-29
Readme and manuals
Help Manual
| Help page | Topics |
|---|---|
| geosmooth package | geosmooth-package geosmooth |
| Apply a Metric Graph Low-Pass Filter Path | apply.metric.graph.lowpass.path |
| Summarize a synthetic dataset as one metadata row | as.data.frame.synthetic_dataset |
| Bootstrap Uncertainty For A Fixed-Selection MALPS Fit | bootstrap.malps |
| Compare Masked Graph Fields with Independent Graph Workers | compare.masked.graph.fields |
| Compare two synthetic datasets scientifically | compare.synthetic.dataset |
| Participant-Grouped Nested Cross-Validation for a Masked Graph Field | cv.masked.graph.field |
| Fit a Chart-Kernel Smoother | fit.chart.kernel |
| Fit a Density Estimator | fit.density |
| Fit Graph Trend Filtering | fit.graph.trend.filtering |
| Fit a Local-Likelihood Density or Bernoulli Smoother | fit.local.likelihood |
| Fit A Local Polynomial Lifting Trend Filter | fit.lpl.tf |
| Fit a Local Polynomial Smoother | fit.lps |
| Fit Model-Averaged Local Polynomial Smoothing | fit.malps |
| Fit a Graph Field from Partially Observed Responses | fit.masked.graph.field |
| Fit Metric-Conductance Graph Low-Pass Regression | fit.metric.graph.lowpass |
| Fit Prediction-Synchronized Local Polynomial Smoothing | fit.ps.lps |
| Fit A Parallel-Transport Trend Filter From A PTTF Operator | fit.pttf.trend.filtering |
| Fit A Fixed-Operator Synchronized LPL-TF Model | fit.slpl.tf |
| Fit SSRHE-Style Hessian L1 Regression | fit.ssrhe.hessian.l1.regression |
| Fit SSRHE-Style Hessian-Energy Regression | fit.ssrhe.hessian.regression fit.ssrhe.hessian.regression.cv fit.ssrhe.hessian.regression.gcv |
| Fit Subject-Occupation Density | fit.subject.od |
| Get Boundary Vertices of a Region in a Graph | get.region.boundary |
| Construct a Graph Trend-Filtering Operator | graph.trend.filtering.operator |
| Perform Harmonic Smoothing with Topology Tracking | harmonic.smoother |
| Construct a Local Polynomial Lifting Trend-Filtering Operator | lpl.tf.operator |
| Report LPS Backend and Chart-Dimension Diagnostics | lps.backend.diagnostics |
| Build a Cluster-Respecting Fold Assignment | lps.grouped.foldid |
| Nested Cross-Validation for LPS with Explicit, Recorded Folds | lps.nested.cv |
| Pointwise Variance and Confidence Band for a Fixed-Configuration LPS Fit | lps.pointwise.band |
| Compute MALPS Linear-Smoother Diagnostics | malps.gcv |
| Materialize a synthetic-data specification | materialize.synthetic |
| Materialize a frozen synthetic instance | materialize.synthetic.instance |
| Construct a Graph Heat-Time Grid | metric.graph.heat.eta.grid |
| Propose a Guarded Lower Graph Heat Time | metric.graph.heat.extend.lower |
| Construct a Reusable Metric Graph Low-Pass Spectral Basis | metric.graph.lowpass.basis |
| Construct a Metric-Conductance Graph Low-Pass Operator | metric.graph.lowpass.operator |
| Summarize a Geometric Smoother or Probability-Mass Fit | model-summary print.chart_kernel print.density_fit print.local_likelihood print.lpl_tf print.ps_lps print.slpl_tf print.summary.geosmooth_fit summary.chart_kernel summary.density_fit summary.graph.trend.filtering.fit summary.local_likelihood summary.lpl_tf summary.lps summary.malps summary.metric.graph.lowpass.fit summary.metric.graph.lowpass.refit summary.ps_lps summary.pttf.trend.filtering.fit summary.slpl_tf summary.ssrhe.hessian.fit summary.ssrhe.hessian.l1.fit summary.ssrhe.hessian.refit |
| Extract Stored Fitted Values and Response Residuals | fitted.chart_kernel fitted.graph.trend.filtering.fit fitted.harmonic_smoother fitted.local_likelihood fitted.lpl_tf fitted.lps fitted.malps fitted.metric.graph.lowpass.fit fitted.metric.graph.lowpass.refit fitted.ps_lps fitted.pttf.trend.filtering.fit fitted.slpl_tf fitted.ssrhe.hessian.fit fitted.ssrhe.hessian.l1.fit fitted.ssrhe.hessian.refit model-values residuals.chart_kernel residuals.graph.trend.filtering.fit residuals.local_likelihood residuals.lpl_tf residuals.lps residuals.malps residuals.metric.graph.lowpass.fit residuals.metric.graph.lowpass.refit residuals.ps_lps residuals.pttf.trend.filtering.fit residuals.slpl_tf residuals.ssrhe.hessian.fit residuals.ssrhe.hessian.l1.fit residuals.ssrhe.hessian.refit |
| Normalize Fitted Values Into a Density | normalize.density normalize.density.default normalize.density.lps normalize.density.metric.graph.lowpass.fit normalize.density.metric.graph.lowpass.refit normalize.density.numeric normalize.density.ps_lps |
| Perform Harmonic Smoothing on Graph Function Values | perform.harmonic.smoothing |
| Plot Method for Harmonic Smoother Results | plot.harmonic_smoother |
| Plot a canonical synthetic dataset | plot.synthetic_dataset |
| Predict From A Local Polynomial Lifting Trend Filter | predict.lpl_tf |
| Predict from an LPS Fit | predict.lps |
| Predict From A MALPS Fit | predict.malps |
| Predict From A Fixed-Operator Synchronized LPL-TF Model | predict.slpl_tf |
| Print Geosmooth S3 Objects | print.graph.trend.filtering.fit print.lpl_tf_operator print.lps print.malps print.malps_bootstrap print.metric.graph.lowpass.fit print.metric.graph.lowpass.refit print.pttf.trend.filtering.fit print.pttf_operator print.slpl_tf_operator print.ssrhe.hessian.cv.fit print.ssrhe.hessian.fit print.ssrhe.hessian.gcv.fit print.ssrhe.hessian.l1.fit print.ssrhe.hessian.operator print.ssrhe.hessian.refit print.transported.graph.hessian.operator |
| Print Method for Harmonic Smoother Results | print.harmonic_smoother |
| Print Method for Summary of Harmonic Smoother Results | print.summary.harmonic_smoother |
| Print a canonical synthetic dataset | print.synthetic_dataset |
| Construct PTTF Local Geometry | pttf.geometry |
| Construct PTTF Transported-Difference Operators | pttf.operator |
| Filter Rows Of A PTTF Operator | pttf.operator.filter.rows |
| Refit A Model Using Its Stored Geometry | refit refit.default |
| Refit A Local Polynomial Lifting Trend Filter | refit.lpl.tf refit.lpl_tf |
| Refit A MALPS Smoother With A New Response | refit.malps |
| Refit Metric-Conductance Graph Low-Pass Regression | refit.metric.graph.lowpass refit.metric.graph.lowpass.fit |
| Refit A Fixed-Operator Synchronized LPL-TF Model | refit.slpl.tf refit.slpl_tf |
| Refit SSRHE-Style Hessian-Energy Regression | refit.ssrhe.hessian.fit refit.ssrhe.hessian.regression |
| Refit SSRHE-Style Hessian L1 Regression | refit.ssrhe.hessian.l1.fit refit.ssrhe.hessian.l1.regression |
| Evaluate Nested Selection Among Candidate Graphs | select.masked.graph.field |
| Construct A Synchronized LPL-TF Fixed Operator | slpl.tf.operator |
| Extract A Conditional Smoother Matrix | smoother.matrix smoother.matrix.default |
| Extract the Linear-Smoother Matrix of a Fixed-Configuration LPS Fit | lps.smoother.matrix smoother.matrix.lps |
| Construct The Conditional MALPS Smoother Matrix | malps.smoother.matrix smoother.matrix.malps |
| Construct an SSRHE-Style Local Hessian Energy Operator | ssrhe.hessian.operator |
| Build Candidate Support Profiles for SSRHE Adaptive-Radius Tuning | ssrhe.support.grid |
| Summary Method for Harmonic Smoother Results | summary.harmonic_smoother |
| Summarize Measurement Support for a Graph Field | support.masked.graph.field |
| Compute the canonical content checksum of a synthetic dataset | synthetic.dataset.checksum |
| List maintained synthetic recipe IDs | synthetic.registry.ids |
| Resolve the legacy seed for an SSRHE registry recipe | synthetic.registry.seed |
| Resolve a maintained synthetic recipe | synthetic.registry.spec |
| Bernoulli response specification | synthetic.response.bernoulli |
| Clustered Gaussian response specification | synthetic.response.clustered.gaussian |
| Gaussian response specification | synthetic.response.gaussian |
| Heteroskedastic Gaussian response specification | synthetic.response.heteroskedastic.gaussian |
| Laplace response with Gaussian contamination | synthetic.response.laplace.outlier |
| Legacy G4 stratified sampling specification | synthetic.sampling.stratified |
| Assemble a synthetic-data specification | synthetic.spec |
| Specify a Gaussian-mixture truth | synthetic.truth.gaussian.mixture |
| Specify a sinusoidal finite-design logit truth | synthetic.truth.logit |
| Specify a versioned named truth evaluator | synthetic.truth.named |
| Specify a finite-design occupation-mixture truth | synthetic.truth.occupation.mixture |
| Specify a polynomial truth | synthetic.truth.polynomial |
| Construct a Transported Graph Hessian Operator | transported.graph.hessian.operator |
| Validate a canonical synthetic dataset | validate.synthetic.dataset |