Deep Learning for POWER
PPA description
IBM Deep Learning packages now include:
- caffe-bvlc (upstream Berkeley Vision and Learning Center)
- caffe-nv (NVIDIA fork)
- digits (NVIDIA DIGITS)
- theano
- torch
- power-mldl (meta-package to install all above)
Pre-requisites now include cuDNN v5.1 .deb packages:
1. Ubuntu 14.04
2. CUDA 7.5
3. cuDNN v5.1 for CUDA 7.5 .debs (https://developer.nvidia.com/cudnn)
- cuDNN v5.1 Runtime Library for Power8 (Deb)
- cuDNN v5.1 Developer Library for Power8 (Deb)
- cuDNN v5.1 Code Samples and User Guide Power8 (Deb)
4. "trusty-backports universe" and "trusty multiverse" repos
$ sudo apt-get install software-properties-common
$ sudo apt-add-repository "deb http://ports.ubuntu.com/ubuntu-ports/ trusty-backports universe"
$ sudo apt-add-repository "deb http://ports.ubuntu.com/ubuntu-ports/ trusty multiverse"
$ sudo apt-get update
Pre-requisites (for DIGITS) include a liblmdb0 backport. Installer may
suggest to omit Deep Learning packages rather than install the backport.
Rejecting that suggestion should cause the installer to offer to install
the liblmdb0 backport.
UPGRADING:
The easiest method to upgrade from the previous release is to uninstall
the old packages first then install the cuDNN .debs and new packages:
$ sudo apt-get purge libopenblas cudnn-req caffe torch theano
$ sudo dpkg -i libcudnn*.deb
$ sudo apt-get install power-mldl
NOTES:
Each package provides an environment setup script. We recommend updating
shell rc file (e.g. `.bashrc`) to source setup scripts:
source /opt/DL/<framework>/bin/<framework>-activate
Each package provides a test script to run some self-tests:
$ <framework>-test
digits-activate script will create $HOME/.digits for cfg and log files.
Caffe packages install to separate directories (/opt/DL/caffe-{bvlc|nv}).
By default caffe-nv is presented as /opt/DL/caffe; can be changed with:
$ sudo update-alternatives --config caffe
Caffe packages include scripts to copy samples to a local directory:
$ caffe-install-samples <somedir>
Adding this PPA to your system
You can update your system with unsupported packages from this untrusted PPA by adding ppa:hartb/ppa to your system's Software Sources. (Read about installing)
sudo add-apt-repository ppa:hartb/ppa sudo apt update
For questions and bugs with software in this PPA please contact Brian W Hart.
PPA statistics
- Activity
- 0 updates added during the past month.
Overview of published packages
1 → 19 of 19 results | First • Previous • Next • Last |
Package | Version | Uploaded by |
---|---|---|
bazel | 0.2.0-2ibm1ppa2 | Brian W Hart () |
caffe | 0.3-1ibm4ppa2 | Brian W Hart () |
caffe-bvlc | 0.3-2ibm3ppa1 | Brian W Hart () |
caffe-nv | 0.14.5-2ibm3ppa1 | Brian W Hart () |
cudnn-req | 1-1ibm2ppa1 | Brian W Hart () |
digits | 3.2-2ibm1ppa1 | Brian W Hart () |
flask-babel | 0.9-2.1~mldl.1 | Frédéric Bonnard () |
flask-wtf | 0.12+notheme-2~mldl.2 | Frédéric Bonnard () |
libopenblas | 0.2.18-2ibm1ppa1 | Brian W Hart () |
mldl-repo-local | 1.1-1ibm1ppa1 | Brian W Hart () |
power-mldl | 2-2ibm2ppa2 | Brian W Hart () |
protobuf | 3.0.0-2ibm1ppa2 | Brian W Hart () |
py-lmdb | 0.87-1~mldl.2 | Frédéric Bonnard () |
python-flask-socketio | 0.6.0+notheme-1~mldl.2 | Frédéric Bonnard () |
speaklater | 1.3-3~mldl.1 | Frédéric Bonnard () |
tensorflow | 0.8.0a-2ibm3ppa1 | Brian W Hart () |
theano | 0.8.2-2ibm1ppa1 | Brian W Hart () |
torch | 7a-2ibm4ppa1 | Brian W Hart () |
wtforms | 2.1-1~mldl.1 | Frédéric Bonnard () |
1 → 19 of 19 results | First • Previous • Next • Last |