Zeitpunkt Nutzer Delta Tröts TNR Titel Version maxTL Di 06.08.2024 00:00:37 61.975 +4 3.608.594 58,2 Fosstodon 4.2.10 500 Mo 05.08.2024 00:00:05 61.971 +5 3.606.116 58,2 Fosstodon 4.2.10 500 So 04.08.2024 00:00:02 61.966 +2 3.604.482 58,2 Fosstodon 4.2.10 500 Sa 03.08.2024 00:00:05 61.964 -2 3.602.552 58,1 Fosstodon 4.2.10 500 Fr 02.08.2024 00:00:23 61.966 -4 3.599.813 58,1 Fosstodon 4.2.10 500 Do 01.08.2024 00:01:11 61.970 +3 3.596.601 58,0 Fosstodon 4.2.10 500 Mi 31.07.2024 00:00:00 61.967 0 3.593.593 58,0 Fosstodon 4.2.10 500 Di 30.07.2024 00:00:20 61.967 +2 3.590.362 57,9 Fosstodon 4.2.10 500 Mo 29.07.2024 00:00:12 61.965 -1 3.587.302 57,9 Fosstodon 4.2.10 500 So 28.07.2024 00:00:15 61.966 0 3.586.611 57,9 Fosstodon 4.2.10 500
Heidi Seibold (@HeidiSeibold) · 11/2022 · Tröts: 1.110 · Folger: 2.071
Di 06.08.2024 14:00
I love the little stories in "Supervised Machine Learning for Science". Of course my favorite is the one on reproducibility.
The book (currently work-in-progress version) is available online: https://ml-science-book.com/reproducibility.html
Print and ebook expected to come out this fall 🤩
The hurricane forecasting system worked like a charm. Until it didn’t. One hot summer day, the server had a meltdown. And with it vanished the model. Fortunately, the researchers still had the code on an old laptop. All they had to do was retrain and deploy the model. It would take an afternoon, Product Raven estimated. But easier said than done. They had trained the model years ago, and the lead developer had retired, leaving them with a mess. A strange folder structure, mysterious file names, and unclear instructions. In what order should they run the scripts? Why does the training produce different models, sometimes even getting stuck in a local optimum? It was a stressful time for the researchers.
[Öffentlich] Antw.: 0 Wtrl.: 0 Fav.: 0 · via Buffer