Sebastian Raschka, PhD (@rasbt): "Took a while, but here it is, a mega write-up on language models for text classification and, of course, Jev. It's basically a visual guide to RNNs, CNNs, transformers, and calibration, with hands-on experiments on accuracy and efficiency. https://magazine.sebastianraschka.co…"

Make money doing the work you believe in

It’s incredible the things you don’t realize are sacred until you can no longer rely on them

Women artists you should know: Isabel Quintanilla (1938–2017)

A pioneer of Spanish realism - known for her dreamy still lifes and incredible use of light and shadow.

When I say I love Cheesecake Factory, I mean this cat who was found malnourished and wandering through a Bay Area mall parking lot. Cheesecake Factory has since been rescued and nursed back to health. He's now healthy, vaccinated, and sweet, but still needs a loving home.

You made it, you own it

You always own your intellectual property, mailing list, and subscriber payments. With full editorial control and no gatekeepers, you can do the work you most believe in.

Language Models for Text Classification: From Bag-of-Words to Jev

278 Likes
15 Replies
22 Restacks

Calibration is the part of this that deserves more attention. Accuracy says how often a model was right on the test set; calibration says whether to trust it on the next input, and it is usually the first thing to drift when the data moves.

Jev is a nice example of why a model history is worth teaching as a system, not a list of architectures. The jump from RNNs to transformers gets all the attention, but calibration is where the practical story gets less tidy.