8 comments

[ 4.4 ms ] story [ 28.9 ms ] thread
As a machine learning hobbyist who is too lazy to learn Tensorflow or -insert trending ML framework-, CoreML is a breeze to use. I love how you can set up an image classification model without writing any code.
It's a breeze ... until you need something slightly different.
Again and again I click these articles to see something about ML the programming language only to be disappointed by a machine learning topic. Overloading acronyms sucks.
People not understanding this is a joke lol......wait it is right?
If there's Core Haskell, surely there's some Core ML as well?
ugh, yeah, i've gotten better at telling, but this time i was really expecting some pared down ML dialect.
(comment deleted)
It's a shame the author didn't go further into CreateML , the app.

https://developer.apple.com/machine-learning/create-ml/

I've taught a couple Intro to Machine Learning classes to a few iOS developers. While I progress to showing them how various bits of machine learning work, I start with CreateML so they can be off and running with basic classifiers in a few minutes.

The App is really quite amazing even though it's limited to a few types of networks.

Of course, as is obligatory, we start with Not a Hotdog.