Machine Learning on the Power8 nodes

We have the IBM PowerAI machine learning framework available on Mio's Power8 GPU enabled nodes. PowerAI release 3.4 provides software packages for several Deep Learning frameworks, supporting libraries, and tools:

Click here for additional information.

Machine Learning on Power8 and x86 nodes

The package theano can be used for machine learning. It is available both on the regular Mio X86 nodes and on the Power 8 nodes. It runs well on the GPUs attached to the Power 8 nodes. We have some run scripts and some slightly modified examples from the Deep Learning Tutorial from the University of Montreal.

To run the examples on Mio do the following:

  1. In a new directory, download the examples
  2. Uncompress it
    tar -xzf ml.tgz
  3. Run the batch script
    sbatch dop8g

There are two README.* files with additional information.

doit  - sbatch file for running on X86 Mio x86 nodes

doP8g - sbatch file for running on Mio Power nodes
        both on the CPU and GPU

To run these examples on Mio use the sbatch command.

The software used here is theano.  See:

A pdf of the main tutorial of the page mentioned
in the README.rst file can be found at:

We have made a few minor changes in code/

The line 

 classifier = pickle.load(open('best_model.pkl'))

was changed to:

 classifier = pickle.load(open('best_model.pkl','rb'))

This was done for python3 compatability

There is an added line 199:
print('Found data at:',dataset)

Some other links:
Deep Learning Tutorials

Deep Learning is a new area of Machine Learning research, which has been
introduced with the objective of moving Machine Learning closer to one of its
original goals: Artificial Intelligence.  Deep Learning is about learning
multiple levels of representation and abstraction that help to make sense of
data such as images, sound, and text.  The tutorials presented here will
introduce you to some of the most important deep learning algorithms and will
also show you how to run them using Theano.  Theano is a python library that
makes writing deep learning models easy, and gives the option of training them
on a GPU.

The easiest way to follow the tutorials is to `browse them online

`Main development `_
of this project.

.. image::

Project Layout


- code - Python files corresponding to each tutorial
- data - data and scripts to download data that is used by the tutorials
- doc  - restructured text used by Sphinx to build the tutorial website
- html - built automatically by doc/Makefile, contains tutorial website
- issues_closed - issue tracking
- issues_open - issue tracking
- misc - administrative scripts