This fork of BVLC/Caffe is dedicated to improving performance of this deep learning framework when running on CPU, in particular Intel Xeon processors (HSW+) and Intel  Xeon Phi processors.This module contains Intel/Caffe with MLSL support for multi-node training/inference.

Version: intel-multinode-1.0.0

Availability: MARCONI

Target: all

Official web site:

Help and Documentation:

Module Specific Help for /cineca/prod/opt/modulefiles/deeplrn/applications/caffe/intel-multinode-1.0.0