Skip to main content
Version: Next

TensorFlow

TensorFlow is TensorFlow is an end-to-end open source machine learning platform., used for Training Framework AI Simulation Virtual Reality . This product integrates TensorFlow, which is an end-to-end open source platform for machine learning. It has a comprehensive, flexible ecosystem of tools, libraries and community resources that lets researchers push the state-of-the-art in ML and developers easily build and deploy ML powered applications.

gui

Prepare​

When referring to this document to use TensorFlow, please read and ensure the following points:

  • This application is installed by Websoft9 console.

  • The purpose of this application complies with the apache2 open source license agreement.

  • Configure the domain name or server security group opens external network ports for application access.

Getting started​

Initial setup​

  1. When completed installation of TensorFlow at Websoft9 Console, get the applicaiton's Overview and Access information from My Apps

  2. Access Jupyter URL locally prompts for a login token

  3. Login to the Jupyter backend with a Token or set a password

Run TensorBoard​

  1. Open the Jupter backend in sequence: New > Python 3 (ipykernel)

  2. Refer to Using TensorBoard in Notebooks, and run the example programs in sequence.Add a parameter--host 0.0.0.0 to the last command (to allow external access).

  3. The TensorBoard is now displayed in the Notebook.

Configuration options​

  • Container Ports

    • 8888: Jupter port
    • 6006: TensorBoard port

Administer​

Troubleshooting​

TensorBoard not visible in Notebook?​

  • Ensure that the TensorBoard command is started with the --host 0.0.0.0
  • Ensure the host port for container 6006 port mapping is enabled.