Package: clustra 0.2.1

George Ostrouchov

clustra: Clustering Longitudinal Trajectories

Clusters longitudinal trajectories over time (can be unequally spaced, unequal length time series and/or partially overlapping series) on a common time axis. Performs k-means clustering on a single continuous variable measured over time, where each mean is defined by a thin plate spline fit to all points in a cluster. Distance is MSE across trajectory points to cluster spline. Provides graphs of derived cluster splines, silhouette plots, and Adjusted Rand Index evaluations of the number of clusters. Scales well to large data with multicore parallelism available to speed computation.

Authors:George Ostrouchov [aut, cre], David Gagnon [aut], Hanna Gerlovin [aut], Chen Wei-Chen [ctb], Schmidt Drew [ctb], Oak Ridge National Laboratory [cph], U.S. Department of Veteran's Affairs [fnd]

clustra_0.2.1.tar.gz
clustra_0.2.1.zip(r-4.7-any)clustra_0.2.1.zip(r-4.6-any)clustra_0.2.1.zip(r-4.5-any)
clustra_0.2.1.tgz(r-4.6-any)clustra_0.2.1.tgz(r-4.5-any)
clustra_0.2.1.tar.gz(r-4.7-any)clustra_0.2.1.tar.gz(r-4.6-any)
clustra_0.2.1.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
clustra/json (API)

# Install 'clustra' in R:
install.packages('clustra', repos = c('https://go-ski.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/mvp-champion/clustra/issues

Datasets:
  • bp10k - Simulated blood pressure data

On CRAN:

Conda:

4.48 score 8 scripts 404 downloads 11 exports 7 dependencies

Last updated from:6515ed4050. Checks:7 WARNING, 2 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-develWARNING113
source / vignettesOK360
linux-releaseWARNING103
macos-releaseWARNING192
macos-oldrelWARNING202
windows-develWARNING120
windows-releaseWARNING108
windows-oldrelWARNING61
wasm-releaseOK85

Exports:clustraclustra_randclustra_sildeltimedeltimeTgen_traj_dataplot_sampleplot_silhouetteplot_smoothsrand_plottrajectories

Dependencies:data.tablelatticeMASSMatrixmgcvMixSimnlme

clustra: Graphics with BP Data
Comparison between SAS and R clustering | Some notes about the SAS execution and results
Rendered fromclustra_bp_vignette.Rmdusingknitr::rmarkdown

Last update: 2024-01-11
Started: 2023-10-04

clustra: clustering trajectories
Rendered fromclustra_vignette.Rmdusingknitr::rmarkdown

Last update: 2023-10-16
Started: 2020-11-11

Readme and manuals

Help Manual

Help pageTopics
clustra-packageclustra-package
allpair_RandIndex: helper for replicated cluster comparisonallpair_RandIndex
Simulated blood pressure databp10k
Checks if non-empty groups have enough data for spline fit degrees of freedom.check_df
Cluster longitudinal trajectories over timeclustra
clustra_rand: Rand Index cluster evaluationclustra_rand
clustra_sil: Prepare silhouette plot data for several k or for a previous clustra runclustra_sil
Timing functiondeltime deltimeT
Data Generatorsgen_traj_data
gendatagendata
Function to test information criteria. Not exported and used by internal function 'kchoose'.ic_fun
A test function to evaluate information criteria for several k values. Not exported and only for debugging internal use.kchoose
Various Loss functions used internally by clustramae_g mme_g mse_g mxe_g
Generates data for one 'id'oneid
Plots a sample of ids in a small mutiples layoutplot_sample
Plots a list item, a silhouette, from the result of 'clustra_sil' along with the average silhouette value. Typically used via 'lapply(list, plot_silhouette)'plot_silhouette
plot_smoothsplot_smooths
Function to predict for new data based on fitted gam object.pred_g
Matrix plot of Rand Index comparison of replicated clustersrand_plot
Function to assign starting groups.start_groups
Fits a thin plate spline to a single group with 'bam'.tps_g
Function to run trajectories inside mclapply with one core.traj_rep
Cluster longitudinal trajectories over time.trajectories
xit_reportxit_report