Package: scanCP 0.1.0

scanCP: Deep Learning–Based Changepoint Detection with Local Neural Models

Implementation of deep learning–based changepoint detection algorithm designed for time series with smooth local fluctuations. The method fits localized feed‑forward neural networks to approximate the underlying smooth component and constructs a residual‑based detector that isolates abrupt structural changes. A fully data‑adaptive empirical cumulative distribution function (ECDF) based thresholding rule and refinement procedures yield accurate changepoint localization without parametric assumptions on noise or trend structure.

Authors:Arman Azizyan [aut, cre], Abolfazl Safikhani [aut]

scanCP_0.1.0.tar.gz
scanCP_0.1.0.zip(r-4.7)scanCP_0.1.0.zip(r-4.6)scanCP_0.1.0.zip(r-4.5)
scanCP_0.1.0.tgz(r-4.6-any)scanCP_0.1.0.tgz(r-4.5-any)
scanCP_0.1.0.tar.gz(r-4.7-any)scanCP_0.1.0.tar.gz(r-4.6-any)
scanCP_0.1.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
card.svg |card.png
scanCP/json (API)

# Install 'scanCP' in R:

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

Bug tracker:https://github.com/armanazizyan/scancp/issues

On CRAN:

3.18 score 14 exports 71 dependencies

Last updated from:6646c6a61c. Checks:9 OK. Indexed: yes.

Target	Result	Time
linux-devel-x86_64	OK	147
source / vignettes	OK	178
linux-release-x86_64	OK	137
macos-release-arm64	OK	159
macos-oldrel-arm64	OK	357
windows-devel	OK	111
windows-release	OK	79
windows-oldrel	OK	101
wasm-release	OK	124

Exports:best_split_free calc_detector combine_detectors decompose_signal_core detect_cp_ecdf fit_global_mlp fit_mlp ma plot_mlp_fits_interactive scan_cp scan_cp_multi_async scan_cp_multi_sync select_best_spike simulate_piecewise_signal_idx

Dependencies:askpass base64enc bslib cachem cli codetools cpp11 crosstalk curl data.table digest doSNOW dplyr evaluate farver fastmap fontawesome foreach fs generics ggplot2 glue gtable highr htmltools htmlwidgets httr isoband iterators jquerylib jsonlite knitr labeling later lazyeval lifecycle magrittr memoise mime openssl otel pillar pkgconfig plotly pracma promises purrr R6 rappdirs RColorBrewer Rcpp rlang rmarkdown RSNNS S7 sass scales snow stringi stringr sys tibble tidyr tidyselect tinytex utf8 vctrs viridisLite withr xfun yaml

Help page	Topics
Best Split for Binary Labels (Free Assignment)	best_split_free
Compute the MLP-Based Changepoint Detector Statistic	calc_detector
Combine Multiple Detector Statistics into a Joint Detector	combine_detectors
Core Signal Decomposition (Changepoints + Piecewise Correction)	decompose_signal_core
ECDF-Based Changepoint Detection for a Detector Statistic	detect_cp_ecdf
Fit Global MLP and Compute Residual Diagnostics	fit_global_mlp
Fit Rolling MLP Models for Change Point Detection	fit_mlp
Two-Sided Moving Average Smoothing	ma
Interactive Plot of Rolling MLP Fits	plot_mlp_fits_interactive
Full Changepoint Detection Pipeline Using Rolling MLPs	scan_cp
Asynchronous Multivariate Changepoint Detection	scan_cp_multi_async
Multivariate Synchronized Changepoint Detection	scan_cp_multi_sync
Select the Most Significant Spike in an ECDF Spacing Curve	select_best_spike
Simulate a Piecewise-Constant Signal with Smooth Trend and Noise	simulate_piecewise_signal_idx

Package: scanCP 0.1.0

scanCP: Deep Learning–Based Changepoint Detection with Local Neural Models

Citation

Development and contributors

Readme and manuals

Help Manual

Usage by other packages (reverse dependencies)