Deploying Trains Overview

You can use Trains and enjoy its automagical capturing of experiment inputs and outputs by simply installing the Trains Python Client Package and using our demo Trains Server, which works with Trains by default and is open to anyone. However, the demo Trains Server resets periodically deleting data.

To take full advantage of Trains features, we offer several deployment options for Trains Server, our backend infrastructure, and Trains Agent, our DevOps solution for experiment execution and autoML.

This page provides an overview of the Trains deployment options which are detailed in this "Deploying Trains" section.

Trains deployment options

Option 1. No Deployment: Our Trains demo server, only.

Description:

  • Automagically track, analyze, and compare experiments using the demo Trains Web-App.
  • Add two lines of code to your Python experiment script and use the demo Trains Server, see our Quick Start page.
  • Optionally, configure Trains options (for example cache, metrics, and log settings), see the Trains Configuration Reference.

Our demo Trains Server is periodically refreshed and data is deleted.

Option 2. Our Trains demo server and Trains Agent

Description:

  • Automagically track, analyze, and compare experiments using the demo Trains Web-App.
  • Manage your experimentation using Trains Agent (see workers and queues), but not your own backend infrastructure.

Deployment:

  1. Install Trains Agent.
  2. Optionally, configure Trains Agent options (for example, Git, package manager, worker, or Docker settings), see the Trains Agent Reference.
  3. Optionally, configure Trains options (for example, cache, metrics, and log settings), see the Trains Configuration Reference.

Option 3. Your own Trains Server

Description:

  • Automagically track, analyze, and compare experiments using the demo Trains Web-App.
  • Manage your own Trains backend infrastructure.

Deployment:

  1. Deploy your own Trains Server, using one of the available formats which include:
  2. Optionally, configure Trains Server for additional features, including sub-domains and load balancers, web login authentication, and the non-responsive task watchdog.
  3. Configure Trains for Trains Server.

Option 4. Your own Trains Server and Trains Agent.

Description:

  • Automagically track, analyze, and compare experiments using the demo Trains Web-App.
  • Manage your own Trains backend infrastructure.
  • Manage your experimentation using Trains Agent (workers and queues).

Deployment:

  1. Deploy your own Trains Server, using one of the available formats which include:
  2. Optionally, configure Trains Server for additional features, including sub-domains and load balancers, web login authentication, and the non-responsive task watchdog.
  3. Configure Trains for Trains Server.
  4. Install and Configure Trains Agent.