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.
Versions across snapshots
| Version | Repository | File | Size |
|---|---|---|---|
0.1.0 |
rolling linux/jammy R-4.5 | leakaudit_0.1.0.tar.gz |
39.2 KiB |
0.1.0 |
rolling linux/noble R-4.5 | leakaudit_0.1.0.tar.gz |
39.0 KiB |
0.1.0 |
rolling source/ R- | leakaudit_0.1.0.tar.gz |
15.6 KiB |
0.1.0 |
latest linux/jammy R-4.5 | leakaudit_0.1.0.tar.gz |
39.2 KiB |
0.1.0 |
latest linux/noble R-4.5 | leakaudit_0.1.0.tar.gz |
39.0 KiB |
0.1.0 |
latest source/ R- | leakaudit_0.1.0.tar.gz |
15.6 KiB |
0.1.0 |
2026-04-23 source/ R- | leakaudit_0.1.0.tar.gz |
0 B |