Package: leakaudit 0.1.0

leakaudit: Detect and Audit Train/Test Leakage from Near-Duplicate Images

Detects near-duplicate images across dataset splits using perceptual hashing, reports the resulting train/validation/test contamination, and produces a corrected, leak-free split assignment. Intended for machine learning researchers who need to verify that image classification splits do not share near-duplicate samples across partitions before reporting model metrics.

Authors:Kartik Patel [aut, cre], Samruddhi Amol Shah [aut], BitandByte [cph]

leakaudit_0.1.0.tar.gz
leakaudit_0.1.0.zip(r-4.7-any)leakaudit_0.1.0.zip(r-4.6-any)leakaudit_0.1.0.zip(r-4.5-any)
leakaudit_0.1.0.tgz(r-4.6-any)leakaudit_0.1.0.tgz(r-4.5-any)
leakaudit_0.1.0.tar.gz(r-4.7-any)leakaudit_0.1.0.tar.gz(r-4.6-any)
leakaudit_0.1.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
leakaudit/json (API)

# Install 'leakaudit' in R:

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

Bug tracker:https://github.com/anakincodex/leakaudit/issues

On CRAN:

computer-vision cran-package data-leakage dhash image-processing machine-learning perceptual-hashing

4.00 score 25 downloads 5 exports 4 dependencies

Last updated from:5b981afb3e. Checks:9 OK. Indexed: yes.

Target	Result	Time
linux-devel	OK	139
source / vignettes	OK	187
linux-release	OK	121
macos-release	OK	76
macos-oldrel	OK	74
windows-devel	OK	64
windows-release	OK	67
windows-oldrel	OK	62
wasm-release	OK	110

Exports:clean_splits compute_hashes dhash_audit find_duplicate_groups hamming_distance

Dependencies:curl magick magrittr Rcpp

Getting started with leakaudit

The problem | Workflow | A minimal, reproducible example | Choosing a threshold

Rendered fromleakaudit.Rmdusingknitr::rmarkdown

Last update: 2026-07-20
Started: 2026-07-20

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
Produce a leak-free split assignment	clean_splits
Compute perceptual hashes for a set of images	compute_hashes
Audit train/val/test leakage from duplicate groups	dhash_audit
Group near-duplicate images by hash distance	find_duplicate_groups
Hamming distance between two hex-encoded perceptual hashes	hamming_distance