Reproduce “Fifty-Two Shades of Pig: What the UCR Archive Really Asks Classifiers to Do” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Omar Swelam, Jaris Küken, Shi Bin Hoo, Frank Hutter, Lennart Purucker

The authors audit 128 UCR datasets, tracing source studies, tasks, and train/test splits. They judge 52 fail the real-world task criterion; 41 have split concerns, while 65 are unclear because evidence is missing. In Crop, 2,753 of 16,800 test series are duplicates; MiniRocket accuracy (fraction classified correctly) is 0.992 on duplicates and 0.722 on the remainder.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.