yeongeon/tensorflow-playground repository overview

⁠tensorflow-playground

You can use tensorflow and pytorch on your own docker after to get clone this branch. Let's begin to learn a way that how to use this project with some of bash scripts.

⁠included things

• pytorch
• tensorflow
• tensorboard
• jupyter
• python3
• apache tinkerpop (with gremlin-server/ipython)

⁠hub repository

• yeongeon/tensorflow-playground⁠

  docker pull yeongeon/tensorflow-playground
  

  Copy

⁠runnable scripts

⁠common things

• env-dockerfile.sh

⁠for dockerizing

• build-dockerfile.sh

  • To make docker image on your own pc.

• tensorflow.sh

  • You can attach your terminal to the container.

⁠docker launchers

1. startup-container.sh

   • Start docker container without come into itself.
   • Also you can take a token to use in jupyter.
   • ! Don't forget it typing 'mkdir ~/data' on your pc for VOLUME_HOST environment value.
   • After that let's move to http://localhost:8888⁠ on your browser.
   • Input the password for jupyter that set in env-dockerfile.sh⁠.

     DEFAULT: jupyter
     

     Copy
   • Also you can use a image in docker hub directly with option value as --hub then without build on your own pc.

     usage : startup-container.sh [--hub] [-h|--help]
     --hub
         To use image in the docker hub.
     -h, --help
         The help
     

     Copy

2. shutdown-container.sh

   • Halt docker container

⁠http pages

⁠jupyter

• http://localhost:8888/⁠
  • Input the password for jupyter that set in env-dockerfile.sh⁠.

    DEFAULT: jupyter
    

    Copy

⁠tensorboard

• http://localhost:6006/⁠
• tip
  • You can see a message as "To store a graph, create a tf.summary.FileWriter" at landing page of the tensorboard, then you can avoid it after to use some of codes on below.
    • sample

    log_dir = "/data/tensorboard"
    tw = tf.summary.FileWriter(log_dir, graph=sess.graph)
    

    Copy

⁠tinkerpop

• ws://localhost:8182/gremlin
• sample code

  from __future__  import print_function  # Python 2/3 compatibility
  
  from gremlin_python import statics
  from gremlin_python.structure.graph import Graph
  from gremlin_python.process.graph_traversal import __
  from gremlin_python.process.strategies import *
  from gremlin_python.driver.driver_remote_connection import DriverRemoteConnection
  
  graph = Graph()
  g = graph.traversal().withRemote(DriverRemoteConnection('ws://localhost:8182/gremlin','g'))
  print(g.V().limit(2).toList())
  

  Copy