caffe


Introduction:

CAFFE is a deep learning framework, a set of libraries and utitlies
that enable the construction of deep learning programs.

CAFFE’s original focus was on computer vision and image analysis,
but it has already branched out to processing speech and text,
sequences, and reinforcement learning. CAFFE can easily switch its
computational base between a CPU and a GPU by changing a single
parameter. CAFFE offers model definitions, optimization settings,
and pre-trained weights.

Web Site:

The CAFFE home page at berkeleyvision:

http://caffe.berkeleyvision.org/

Reference:

Usage:

On any ARC cluster, check the installation details
by typing “module spider caffe”.

CAFFE requires that the appropriate modules be loaded before it can
be used. One version of the appropriate commands for use on NewRiver is:

module purge
module load gcc/4.7.2 
module load yasm/1.3
module load x264/1.0
module load fdk-aac/1.0
module load lame/3.99.5
module load ffmpeg/2.5.4 
module load cuda/7.0.28
module load Anaconda/4.2.0
module load Anaconda-boost/1.58.0 
module load glog/0.3.3
module load atlas/3.11.36
module load opencv/2.4.11
module load python/2.7.10
module load caffe/1.0
    

Examples:

The following batch file demonstrates the use of CAFFE:

#! /bin/bash
#
#PBS -l walltime=00:15:00
#PBS -l nodes=1:ppn=1:gpus=1
#PBS -W group_list=newriver
#PBS -q open_q
#PBS -j oe
#
cd $PBS_O_WORKDIR
#
module purge
module load gcc/4.7.2 
module load yasm/1.3
module load x264/1.0
module load fdk-aac/1.0
module load lame/3.99.5
module load ffmpeg/2.5.4 
module load cuda/7.0.28
module load Anaconda/4.2.0
module load Anaconda-boost/1.58.0 
module load glog/0.3.3
module load atlas/3.11.36
module load opencv/2.4.11
module load python/2.7.10
module load caffe/1.0
#
caffe train -solver solverlmdb.prototxt
#
caffe test -model net_train_test.prototxt -weights mynets_iter_10000.caffemodel 
#

The data for this example is too extensive to be provided in our example
directory. A PARTIAL set of files to carry out a similar process are available in
caffe_example.tar