FV 1.0 Developers Guidebook
GitHub icon
About
Taiwan AI Labs - Federated Validation (FV) System : Developers' Guidebook
Search
1 - What is Federated Validation (FV)?
1.1 What is Federated Validation?
1.3 What's the difference between Federated Validation (FV) and Federated Learning (FL)?
1.4 What's the advantage of using Federated Validation?
1.5 How many sites (edges) are required for Federated Validation?
2 - System Overview
2.1 Taiwan AI Labs Federated Validation (FV) System overview
2.2 How does Taiwan AI Labs Federated Validation System work?
2.3 What is FV project and how to create it?
2.4 What is FV plan and how to create it?
3 - Ground-Truth Data
3.1 What datasets will be used to validate the AI model?
3.2 How do I prepare the Ground-Truth Data?
3.3 How to upload the Ground-Truth Data to the system for validation?
4 - How to implement AI model docker image for Valdation?
4.1 The AI Model hooking interface for validation plan.
4.2 The FV (federated validation) diagram
4.3 Input Interface
4.4 Output Interface
4.5 FV - Error Output
5 - How to "hook up" my AI model to the FV system?
5.1 Is there a right FV plan ready?
5.2 How to upload the AI model?
5.3 How to upload the AI model weight?
5.4 How to start the validating plan and get the result?
6 - Hands-on Examples:
6.1 Hello FV
6.2 Hello FV input datasets.
6.3 Hello FV output
7 - Others
7.1 How to know how much bandwidth is used during the training period?
8 - More
8.1 What is Taiwan AI Labs Federated Validation System?
8.2 What is Taiwan AI Labs Federated Learning Platform?
About
7.1 How to know how much bandwidth is used during the training period?
(Peiling)