Tag archive
A CC BY 4.0 dataset of the rates central banks actually publish, back to 1914, refreshed daily. One CSV per institution, loads in one line with datasets, pandas, Polars or DuckDB.
Field-level normalization for a cross-platform video corpus: identity, time, engagement, aspect ratio, and the null-vs-zero rule that keeps per-platform comparisons honest.
A dataset of 19th-century Russian texts in pre-reform orthography with images and ground truth, perfect for training OCR models and exploring historical linguistics.
We open-sourced 449 real equipment financing quotes so nobody has to trust our...
Robot learning is getting better very quickly. We now have better policy architectures, more capable...
Announcing QTE Technologies Open Industrial Dataset (1M+ SKUs) QTE Technologies Co., Ltd....
What clean, normalized SEC EDGAR data looks like vs raw XBRL — with a free 100-row sample and runnable pandas. S&P 500 fundamentals, Form 4 insider trades, 13F holdings.
Compute net insider buying from SEC Form 4 in a few lines of pandas, using a free 100-row sample of parsed insider-trade data. The transaction codes that matter, and the judgment calls that trip people up.
2026 年,美國國家科學基金會(NSF)發布了一個徵案計畫:Unlocking Dataset Value for AI-Enabled Scientific Discovery(暱稱 AI...
Last month I wrote about building a 1,800-page calculator site solo. Since then the recipe hub on...
Real-money benchmark of Claude prompt caching across a 30-day production deployment — $487 baseline to $52 with caching, exact API call traces.
I run a collector that snapshots every active Polymarket market's price every 15 minutes. After 92...