Follow

A new 16-bit floating point format for machine learning! hub.packtpub.com/why-intel-is-

It increases range at the expense of precision, and often allows 16-bit computation (smaller, faster hardware) to replace 32-bit ML logic.

Sign in to participate in the conversation
Mastodon

Server run by the main developers of the project 🐘 It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!