SymetryML6.1
  • Introduction
  • Guides
    • Onboarding Guide
    • Technical Requirements
    • Admin User Guide
    • Installation Guide
      • Installation Guide - GPU
      • Installation Guide - Spark
  • SymetryML GUI
    • ML Toolkit
      • The SymetryML Difference
      • Data Mining Lifecycle
      • SymetryML Concepts
      • Data Sources
      • Streams
      • Encoders
      • Projects
      • Models
    • Sequence Models
    • SymetryML Federated Learning
      • Creating the Federation
      • Load data to local project
      • Requesting Federation Information from Admin Node
      • Joining a Federation with a peer node
      • Federated Data & Modelling
      • Appendix
    • DEM Generator
  • SymetryML Rest Client
    • REST API Reference Guide
      • SymetryML REST API Security
      • SymetryML JSON API Objects
      • Encoder Object REST API
      • SymetryML Projects REST API
      • About Federated Learning
      • Hipaa Compliance and Federated Learning
      • Federated Learning API
        • Federated Learning Topologies
        • Federated Learning with Nats
        • Federated Learning with AWS
        • Fusion Projects
      • Exploration API
      • Modeling API
      • Exporting and Importing Model
      • Third Party Model Rest API
      • SymetryML Job Information
      • Prediction API
      • Data Source API
      • Project Data Source Logs
      • Stream Data Source API
      • AutoML with SymetryML
      • Transform Dataframe
      • Select Model with SymetryML
      • Auto Select with SymetryML
      • Tasks API
      • Miscellaneous API
      • WebSocket API
      • Appendix A JSON Data Structure Schema
      • Appendix B Sample Code
  • SymetryML SaaS
    • SaaS Homepage
    • SaaS Dashboard
    • SaaS Account
    • SaaS Users
    • SaaS Licence
Powered by GitBook
On this page
  • Creating AWS Federation
  • Creating NATS Federation
  1. SymetryML GUI
  2. SymetryML Federated Learning

Creating the Federation

PreviousSymetryML Federated LearningNextLoad data to local project

Last updated 2 years ago

Creating a federation usually starts with a single node. This node will be the administrator of the federation and will control things such as who gets access to a federation and the overall sync schedule of the federation

After you log in to the SymetryML GUI, you arrive at the main SymetryML GUI interface. In the upper left corner you see the heading Projects, and directly below is the Create Project icon, which provides a drop-down menu with the following options:

  • Empty Project

  • Local Project

  • Create Federation

  • Join Federation

  • Fusion Project

In this tutorial, we are going to focus on the latter two options which constitute the federated learning capabilities of SymetryML:

  • Create Federation

  • Join Federation

When you click on the Create Federation option, you are presented with the following consecutive windows:

The following information needs to be supplied:

Parameters Needed
Description

Project Name

This is the name of the local project.

Project Type

This applies to the local project, declare whether the local project is CPU, GPU, Partition, etc.

Project Encoder

Optional parameter specifying the encoder to be used.

Click Next to choose the type of Federation backend.

Creating AWS Federation

Parameters Needed
Description

Federation Name

Name for your federation.

AWS Region

This is the region where S3 bucket will reside.

AWS S3 Bucket

This is the name of the S3 bucket where updates will be saved to.

Sync Schedule

The frequency at which your nodes will share their information.

AWS S3 Access Key

Enter AWS Access Key information here.

AWS S3 Secret Key

Enter AWS Secret Key information here.

Creating NATS Federation

Parameters Needed
Description

Federation Name

Name for your federation.

NATS Hosts

Location of the NATS server

Sync Schedule

The frequency at which your nodes will share their information.

NATS Request Timeout(s)

Timeout in seconds

NATS Max Memory

NATS maximum memory

Authentication Type

One of None, Password, Token

User Name

User name if Authentication Type is Password

Password

Password if Authentication Type is Password

Token

Token if Authentication Type is Token

Federation Restrictions

Parameters Needed
Description

HIPAA Compliant

SMPC

AWS is the default choice for Federation Type. If you wish to use NATS proceed

Once all the information above is satisfied, click Finish. Click to continue to the next section, Load data to local project.

Once all the information above is satisfied, click Finish. Click to continue to the next section, Load data to local project.

Adhere to strict .

Enable

here
here
here
HIPAA Compliance
SMPC
SymetryML GUI : Creating a Federation
Federated Learning: Creating a Federation (window1)
Federated Learning: Creating a Federation (window2)
Federated Learning: Creating a Federation - NATS