Why do you call TabPFN a revolution for data science?

Prof. Dr. Frank Hutter is one of the fathers of the AutoML approach. Now he has followed up and promised a revolution. TabPFN is its name. Prof. Dr. Frank Hutter and his team presented a "radically new approach" to tabular classification. It is a new tabular data classification method that takes < 1 second & yields SOTA performance (competitive with the best AutoML pipelines in an hour). So far, it is limited in scale, though: it can only tackle problems up to 1000 training examples, 100 features and 10 classes. It works best when all features are numerical and there are no missing values (but we believe when we focus on those cases we’ll also improve performance for them).

News: 

• Facebooks Language Model
• Arduino and ML
• The TabPFN Paper
• The Summary