FL 2.0 Manuals
About
Taiwan AI Labs - Federated Learning framework: User Manuals
Search
1 - System deployment
1.1 System specification
1.2 Preparation for on-premises machine installation
1.3 How to deploy the FLP system?
1.4 Maintenance for system
1.5 Roles and Terminology
1.6 Task Flows
1.7 Roles, Tasks and Operations
1.8 How to start?
2 - How to create a Federated Learning project?
2.1 How to create a Federated Learning project? (Fed-Admin)
2.2 How to invite partners to join my FL project? (PI)
2.3 How to join a Federated Learning project as alliance partner? (Co-PIs and PI)
3 - How to Setup a Federated Learning Project? (PI)
3.1 How to setup a Federated Learning project? (PI)
3.2 How to upload the initial AI Model (PI)?
- 3.2.1 Build AI Model Image Online
- 3.2.2 Build AI Model Image Locally
- 3.2.3 View the pushed AI Model images
3.3 How to upload the AI model pretrained weights? (PI)
4 - How to initial and setup a Federated Learning plan? (PI)
4.1 What's the difference between Federated Learning project and Federated Learning plan?
4.2 How to create and setup a Federated Learning plan? (PI)
4.3 How to fulfill FL aggregator algorithm settings? (PI)
5 - What aggregation algorithm should I choose? (PI)
5.1 How do I select the right aggregation algorithm?
5.2 Algorithm - FedAvg
5.3 Algorithm - FedAvgM
5.4 Algorithm - FedAdaGrad
5.5 Algorithm - FedAdam
5.6 How do I view current aggregator settings?
6 - How to start a Federated Learning plan?
6.1 What to prepare before kicking off a training plan?
6.2 How to upload training datasets? (PI and Co-PIs)
6.3 How to adjust the training schedule? (PI and Co-PIs, optional)
6.4 How to start a training plan?
6.5 How to know a training plan is complete?
7 - How to review the result of Federated Learning plan?
7.1 Where could I see the training result? (PI)
7.2 How to evaluate the model performance?
7.3 How to obtain the (global) trained models? (PI)
7.4 How to obtain the best local models? (Co-PIs and/or PI)
8 - Others
8.1 How to check the bandwidth usage during the training period?
8.2 How to push an AI model image that is saved as a tar file ?
9 - More
9.1 What is Taiwan AI Labs Federated Learning Framework?
- 9.1.1 How does Federated Learning work?
9.2 What is Taiwan AI Labs Federated Learning Platform?
- 9.2.1 Federated Learning Platform Specification
9.3 Dataflow of Taiwan AI Labs FL Framework
- 9.3.1 Federated Learning Platform (FLP) - Whole Picture
9.4 Taiwan AI Labs Federated Learning Edge Machine Security Compliance
- 9.4.1 Data Security Compliance
9.5 AutoML and Data Governance
9.6 AutoFL - Model Zoo and CDM
9.7 Federated Analysis
9.8 Code of AI Ethics
About
7.4 How to obtain the best local models? (Co-PIs and/or PI)
Please contact Taiwan AI Labs for detailed information of this advanced feature.