Running TensorFlow App on AWS

From dftwiki3
Revision as of 17:21, 8 October 2017 by Thiebaut (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

--D. Thiebaut (talk) 18:18, 8 October 2017 (EDT)


  • open aws console
  • EC2
  • Pick 64-bit Deep Learning AMI Ubuntu Version 2.3_Sep2017 - ami-06e80e7c
  • Pick instance (e.g. c3.4xlarge)
  • Launch
  • Connect
  • Open Terminal
ssh -i "$HOME/.ssh/mykeyDFT.pem" ubuntu@ec2-nnn-nnn-nnn-nnn.compute-1.amazonaws.com
  • mkdir a directory to hold the data + Tensorflow code
mkdir ML

  • In other Terminal, cd to directory with Tensorflow program & data


rsync -azv -e "ssh -i$HOME/.ssh/mykeyDFT.pem" * ubuntu@ec2-54-159-224-197.compute-1.amazonaws.com:ML

  • back in other Terminal window, launch app:
python theApp.py  parameter1=value parameter2=value