Package: dsPUcopula 0.1.0

dsPUcopula: DataSHIELD Server Functions for Partition-of-Unity Copula Based Synthesis

Implements the server-side components required to generate privacy-protecting synthetic data in the DataSHIELD infrastructure. The package orchestrates the preprocessing, copula fitting, synthetic data generation and privacy scoring workflows by combining R and Python tooling.

Authors:Andreas Mändle [aut, cre]

dsPUcopula_0.1.0.tar.gz
dsPUcopula_0.1.0.zip(r-4.7)dsPUcopula_0.1.0.zip(r-4.6)dsPUcopula_0.1.0.zip(r-4.5)
dsPUcopula_0.1.0.tgz(r-4.6-any)dsPUcopula_0.1.0.tgz(r-4.5-any)
dsPUcopula_0.1.0.tar.gz(r-4.7-any)dsPUcopula_0.1.0.tar.gz(r-4.6-any)
dsPUcopula_0.1.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
card.svg |card.png
dsPUcopula/json (API)

# Install 'dsPUcopula' in R:

install.packages('dsPUcopula', repos = c('https://amaendle.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/amaendle/dspucopula/issues

On CRAN:

2.40 score 1 scripts 13 exports 144 dependencies

Last updated from:ce5cc23fb9. Checks:7 WARNING, 2 OK. Indexed: yes.

Target	Result	Time
linux-devel-x86_64	WARNING	150
source / vignettes	OK	208
linux-release-x86_64	WARNING	149
macos-release-arm64	WARNING	105
macos-oldrel-arm64	WARNING	67
windows-devel	WARNING	89
windows-release	WARNING	78
windows-oldrel	WARNING	79
wasm-release	OK	134

Exports:%>%estimateMarginalsDS fitPUcopulaDS generateSyntheticDS postprocessDataDS preprocessDataDS py_anonymeter_Inference py_anonymeter_SinglingOut py_syndat_scores save_original_classesDS save_original_varnamesDS setup_python simulateCopulaDS

Dependencies:admisc askpass backports bit bit64 boot broom caret cellranger childsds class cli clipr clock codetools colorspace cpp11 crayon curl data.table deldir DescTools diagram digest dplyr dsBase e1071 Exact expm farver forcats foreach fs future future.apply gamlss gamlss.data gamlss.dist generics ggplot2 gld glmnet globals glue gower gtable hardhat haven here hms httr interp ipred isoband iterators jomo jsonlite KernSmooth labeling lattice lava lifecycle listenv lme4 lmom logspline lubridate magrittr MASS Matrix matrixStats mice mime minqa mitml ModelMetrics msm mvtnorm nlme nloptr nnet numDeriv openssl ordinal pan parallelly pillar pkgconfig plyr png polycor pracma prettyunits pROC prodlim progress progressr proxy PUcopula purrr purrrlyr R6 RANN rappdirs rbibutils RColorBrewer Rcpp RcppEigen RcppTOML Rdpack readr readxl recipes reformulas rematch reshape2 reticulate rlang rootSolve rpart rprojroot rstudioapi S7 scales shape sparsevctrs SQUAREM stringi stringr survival sys tibble tidyr tidyselect timechange timeDate tzdb ucminf utf8 vctrs VGAM viridisLite vroom withr

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
Estimate marginal distributions for the fitted copula	estimateMarginalsDS
Fit the PU copula model on the server side	fitPUcopulaDS
Generate synthetic data and privacy scores	generateSyntheticDS
Restore the original factor structure after synthesis	postprocessDataDS
Helpers for preserving and restoring server-side metadata	preprocess_helpers
Prepare server-side data prior to copula fitting	preprocessDataDS
Run the Anonymeter inference attack evaluation	py_anonymeter_Inference
Run the Anonymeter singling out attack evaluation	py_anonymeter_SinglingOut
Compute utility scores using the 'syndat' Python module	py_syndat_scores
Capture the original server-side classes	save_original_classesDS
Capture the original server-side variable names	save_original_varnamesDS
Configure the Python environment used by dsPUcopula	setup_python
Draw samples from the fitted PUcopula model	simulateCopulaDS