Configure an FL project
After you create a Federated Learning (FL) project or are invited to join one, you must configure its nodes, data, and members. After the configuration is complete, you can proceed to the development phase. This topic describes how to configure an FL project in the management console.
Prerequisites
Before you configure an FL project, complete the following tasks:
Confirm that you have installed and deployed the nodes, authorized access between them, and that the nodes use the federated learning engine. For more information about installing, deploying, and authorizing nodes, see Privacy-Preserving Computation Nodes.
Confirm that you have added the data tables required for the federated learning project. For more information, see Data Tables.
Confirm that the members required for the federated learning project have been added to the management console. For more information, see Add a member.
Step 1: Configure nodes
Authorize the nodes created by your organization for the FL project. These nodes store the data that is authorized for the project.
Log on to the Ant Privacy-Preserving Computation Platform.
In the navigation pane on the left, click Privacy-Preserving Computation Projects.
On the Privacy-Preserving Computation Projects page, click My Projects or Invited Projects, depending on whether you created the project or were invited to it.
On the Privacy-Preserving Computation Projects page, click Resource Configuration in the row of the target FL project.
In the Resource Configuration panel, on the Node Configuration tab, click Add Federated Learning Node.
Select a federated learning node and click Save in the Actions column.
These are the nodes that you prepared in the Prerequisites section. For more information about the relationship between project types and node engine types, see Engines.
NoteTo authorize multiple nodes for the project, repeat Step 5 and Step 6 to add more federated learning nodes.
After you finish adding nodes, click Save.
The node configuration cannot be changed after it is submitted. Confirm that the configuration is correct, and then click OK in the dialog box that appears.
An Operation successful message indicates that the nodes are configured.
Step 2: Configure data
Authorize data for the nodes that you configured in Step 1. This allows the project to use the specified data.
On the Privacy-Preserving Computation Projects page, click Resource Configuration in the row of the target FL project.
In the Resource Configuration panel, click the Data Configuration tab, and then click Add Data Table Authorization.
In the Add Data Table Authorization panel, select the target data table in the Select Data Table area.
NoteEnsure that the target data table is in the Available state on the target node. If you cannot find the target data table, see Add a data table and confirm that the new data table is in the Available state.
In the Data Table Alias field, enter an alias for the data table. By default, the alias is the same as the data table name.
FL projects typically do not allow data tables with the same name. If multiple project participants have data tables with the same name, you can use aliases to distinguish them.
In the Add Data Table Authorization panel, click Save.
NoteIf you added multiple federated learning nodes on the node configuration tab, you must authorize data tables for each node by repeating Steps 1 to 5 in this section.
After you finish authorizing the data tables, click Save in the Resource Configuration panel.
An Operation successful message indicates that the data is configured.
Step 3: Configure members
Add members to the project. Authorized members can access the project console to perform development tasks.
On the Privacy-Preserving Computation Projects page, click Member Configuration in the row of the target FL project.
In the Member Configuration panel, click Add Member.
Select a member that is already associated with your organization and click Save.
An Operation successful message indicates that the member is configured.