Math Vector Library Machine Learning Extension
Programming Reference Manual
Version 1.0 DD-00004-110

2.1 Introduction

The Math Vector Library Machine Learning Extension is an extension to the Math Vector Library which provides high-performance function library with vectorized versions of standard mathematical functions. The Machine Learning extension builds upon the infrastructure of the Math Vector Library to provide perfomance tuned primitive functions for machine learning applications.

The functions can operate both on dense and strided vectors, the latter can be supplied individually for result and operand vectors. Stride only executes the function on every n -th element leaving the elements in between untouched.

This manual describes the Application Programming Interface (API) of the machine learning extension functions.