Running and deploying the Federated Learning experiment

After you have completed the configuration for the Federated Learning experiment and have each involved party call out to the aggregator from the previous step, the next step is to run the Federated Learning experiment to train your model.

Cloud open beta

This is a Cloud open preview and is not supported for use in production environments.

Run the Federated Learning experiment

After you have all the parties call out to the aggregator in the previous step by running the yml file, the Federated Learning experiment runs automatically.

Check the progress of the experiment

On the top right, you can see the progress of steps for training. It goes from Create experiment to Setup at the start. For each round of training, the closed hoop that consists of four circles loops:

  • Sending model: Federated Learning sends the model data each party.
  • Training: The process of training the collected data locally.
  • Receiving models: Once the data is trained, they are sent back to the Federated Learning experiment.
  • Aggregating: The aggregator combines the results of all the trained data to produce a final model.

View your results

Once the training is complete, a chart demonstrating model accuracy over each round of training is drawn. Hover over the points on the chart for more information on a single point’s exact metrics.

A Training rounds table is shown at the bottom for more details on each training round.

When you are done with the viewing, click on Save model to project to save the Federated Learning model to your project.

Deploy your model

After you save your Federated Learning model, you can deploy and score the model like other machine learning models. See Deploying models.