You can download the latest tensorflow via conda. For that you have to generate your own environment by

$ module load python
$ conda create -n my_python
$ source activate my_python
(my_python) $ conda install -c anaconda tensorflow

then you can start tensorflow via python

(my_python) $ python
Python 2.7.13 |Anaconda 4.3.1 (64-bit)| (default, Dec 20 2016, 23:09:15)
[GCC 4.4.7 20120313 (Red Hat 4.4.7-1)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
Anaconda is brought to you by Continuum Analytics.
Please check out: http://continuum.io/thanks and https://anaconda.org
>>> import tensorflow

You can use the following settings for better performance on KNL:

(my_python) $ export OMP_NUM_THREADS=128
(my_python) $ export KMP_BLOCKTIME=30
(my_python) $ export KMP_SETTINGS=1
(my_python) $ export KMP_AFFINITY=granularity=fine,verbose,compact,1,0
(my_python) $ numactl -m 0 python convolutional_network.py
  • No labels