Welcome to the STADLE documentation!

STADLE documentation is divided into several sections on the left sidebar. If you’re unable to find what you’re looking for, please let us know! Further information on STADLE and its features can be found on our website. If this is your first time here, you may want to start with the Overview section.

Table of Contents:

• Overview
  • Introduction to STADLE
  • STADLE Architecture
  • Initial Base Model Upload Process
  • Federated Learning Cycle with STADLE
  • Client-Side Local Training Cycle
• Quickstart
  • STEP 1: Set up the STADLE Server
  • STEP 2: Install the STADLE Client
  • STEP 3: Run a Local STADLE Example
• Usage
  • STADLE Aggregator Functionalities
    • Creating a Project
    • Initiating an Aggregator
    • Downloading Models
    • Completing the Current Round
    • Aggregation Threshold
    • Agent Timeout
    • Aggregation Method Selection
    • Synthesize Semi-Global Models
    • Aggregation Management
    • Performance Tracking
    • Stopping & Restarting Aggregators
  • Client-side STADLE Integration
    • Local Training Code
    • Integration with BasicClient
  • Running Client-Side STADLE Components
    • Uploading the Base Model via CLI
    • Running Agents
    • Execution Order Summary
• Documentation
  • Client API Documentation
    • BasicClient
  • Config File Documentation
    • Configuration of Agent
  • Client CLI Documentation
    • Upload Model Command