matedor-talk.pdf (5.91 MB)
MAtrix, TEnsor, and Deep-learning Optimized Routines (MATEDOR)
Version 2 2018-04-28, 23:11
Version 1 2018-04-25, 03:04
presentation
posted on 2018-04-28, 23:11 authored by Azzam HaidarAzzam Haidar, Stan TomovThe MAtrix, TEnsor, and Deep-learning Optimized Routines (MATEDOR)
project seeks to develop software technologies and standard APIs, along
with a sustainable and portable library for large-scale computations,
but whose individual parts are very small matrix or tensor computations.
The main target is the acceleration of applications from important
fields that fit this profile, including deep learning, data mining,
astrophysics, image and signal processing, hydrodynamics, and more.