-
Notifications
You must be signed in to change notification settings - Fork 148
Home
Welcome to the OpenCL-caffe wiki!
#OpenCL caffe Wiki
- a. System: Ubuntu >=12.04 with AMD's GPU
- b. OpenCL runtime environment
- c. [clBLAS] (https://github.com/clMathLibraries/clBLAS)
##Caffe dependency The OpenCL caffe project is based on Berkeley's deep learning framework Caffe, so first you need to install the dependencies by following Caffe's installation instructions at : http://caffe.berkeleyvision.org/installation.html,
NOTE: The CUDA SDK is required because some *.cu files are still in the project (to be cleaned up)
More information about CAFFE can be found at the CAFFE homepage
Go to the root directory of OpenCL-Caffe (CAFFEROOT in the following)
mkdir build
cd build
cmake ..
make
make runtest
The log files generated by caffe are redirected to a log subdir. Under the caffe directory, issue mkdir log.
This will create the log dir for all the generated log files.
For exampleexample, to check training loss, issue grep -ni loss log/caffe.INFO
- CIFAR data take cifar10 (http://www.cs.toronto.edu/~kriz/cifar.html) for example. cd data/cifar10, then run ./get_cifar10.sh. After the download is finished, cd examples/cifar10, run ./create_cifar10.sh;
- ImageNet data
Under the directory CAFFEROOT/ choose the network you want to train, eg ./examples/imagenet/train_alexnet.sh
./build/tools/caffe time -model=models/bvlc_alexnet/deploy.prototxt -gpu 0
NOTe: Timing info it is not yet accurate in this test case.
See log/caffe.INFO for timing.