PowerAI release 3.1 includes packages for both NV Caffe 0.14.5 and 0.15.13. The versions may behave differently (e.g. in performance or convergence) with different models.
NV Caffe 0.15.13 is NCCL enabled and will be installed by default. Version 0.14.5 can be installed as follows:
Uninstall 0.15.13 (if needed)
$ sudo apt-get purge caffe-nv
...
The following packages were automatically installed and are no longer required:
bazel caffe-bvlc caffe-ibm digits libnccl1 libopenblas tensorflow theano torch
Use 'sudo apt autoremove' to remove them.
The following packages will be REMOVED:
caffe-nv* power-mldl*
...
Do you want to continue? [Y/n]
Install 0.14.15 version specifically
$ sudo apt-get install caffe-nv=0.14.5-3ibm1
PowerAI release 3.1 was tested with NVIDIA GPU driver version 361.93.03. That driver package can be downloaded from the NVIDIA site as follows:
Using Torch with DIGITS requires additional packages that are not part of the PowerAI release 3.1 distribution.
Torch can be made to work with DIGITS as follows:
Install PowerAI Torch and DIGITS packages
$ sudo apt-get install digits torch
Install pre-requisite packages from Ubuntu
$ sudo apt-get install libhdf5-serial-dev liblmdb-dev
Install additional luarocks needed for DIGITS' Torch support
$ source /opt/DL/torch/bin/torch-activate
$ luarocks install --local --dep-mode=order tds
$ luarocks install --local --dep-mode=order totem
$ luarocks install --local --dep-mode=order"https://raw.github.com/deepmind/torch-hdf5/master/hdf5-0-0.rockspec"
$ luarocks install --local --dep-mode=order"https://raw.github.com/Neopallium/lua-pb/master/lua-pb-scm-0.rockspec"
$ luarocks install --local --dep-mode=order lightningmdb 0.9.18.1-1 LMDB_INCDIR=/usr/include LMDB_LIBDIR=/usr/lib/powerpc64le-linux-gnu
$ luarocks install --local --dep-mode=order "https://raw.githubusercontent.com/ngimel/nccl.torch/master/nccl-scm-1.rockspec"