Added the Fixedpoint Saturating Add Template Specialization

google / gemmlowp

Low-precision matrix multiplication

Apache License 2.0

1.77k stars 451 forks source link

Closed lamarrr closed 4 years ago

lamarrr commented 4 years ago

I wanted to use this in Tensorflow Lite but found out it was only implemented for Int16. I added the Int8 equivalent.