News for hackers. Stuff that matters.

A comment on In Defense of Pure 16-Bit Floating-Point Neural Networks. Parent comment

Re:In Defense of Pure 16-Bit Floating-Point Neural Networks by _nalply on Tuesday May 23, 2023 @02:14AM (#36040550)
Posits seems to be better for 8-bit or even 6-bit. There is only one not-a-number place, the NaR (not a real). This means for 6-bit posits you have 63 points in the number space.

https://en.wikipedia.org/wiki/Unum_(number_format)#Unum_III

[ Parent | on HN ]
Re:In Defense of Pure 16-Bit Floating-Point Neural Networks by Dylan16807 on Tuesday May 23, 2023 @04:01PM (#36050080)
The way posits focus on numbers near 1.0 is probably going to have a bigger effect. A 6 bit float with 4 exponent bits is the best competitor to a 6 bit posit, and it would only have four non-finite numbers.
[ Parent | on HN ]