Table of Contents

• Introduction to Federated Learning
  • FEDn: the SDK for scalable federated learning
• Quickstart Tutorial PyTorch (MNIST)
  • Prerequisites
  • Launch a pseudo-distributed FEDn Network
  • Install the FEDn SDK
  • Prepare the compute package and seed the FEDn network
  • Configure and attach clients
  • Start a training session
  • Automate experimentation with several clients
  • Access logs and validation data from MongoDB
  • Access model updates
  • Clean up
  • Where to go from here?
• APIClient
• Compute Package
  • The compute package
  • Designing the compute package
  • Packaging for distribution
  • More on local data access
  • Running the client
• Architecture overview
  • The clients: tier 1
  • The combiners: tier 2
  • The controller: tier 3
  • Notes on aggregating algorithms
• Aggregators
  • on_model_update
  • combine_models
• Model Serialization/Deserialization - Helpers
• Frequently asked questions
  • Q: How do I remove/replace the compute package?
  • Q: Can I skip fetching the remote package and instead use a local folder when developing the compute package
  • Q: How can other aggregation algorithms can be defined?
  • Q: What is needed to include other ML frameworks in FEDn like sklearn, xgboost, etc.?
  • Q: Can I start a client listening only to training requests or only on validation requests?:
  • Q: How do you approach the question of output privacy?
• API reference
  • fedn package

Indices and tables

• Index

• Module Index

• Search Page