Model Evaluation
Dr. Smith is over-the-moon excited that her federated training and validation have been completed. She now has a Code Run where she can download her trained model weights and view the output Datasets created by the validation stage. Dr. Smith really would like to understand how her model has performed with both validation Datasets using the ROC evaluation metric. Armed with her evaluation metric, Dr. Smith decides to create a report with custom visualizations within the reports section of her Code Run.
Help Dr. Smith build her report with custom visualizations within the FCP by completing Notebook #5: Federated Evaluations
Supplemental Resources
🎉 Congratulations! 🎉
You helped Dr. Smith produce groundbreaking work in the field of pneumonia detection. Because of your help, Dr. Smith and Dr. Jones have jointly decided to publish their work in a medical journal and present it at an upcoming conference. Dr. Smith hopes to expand the work by adding additional collaborators, and Dr. Jones thinks he may be able to further improve model performance by doing some configuration tweaking of NVFlare and model hyperparameter tuning.
Thank you for choosing to invest your time in learning how to use the Rhino Health FCP. We can't wait to see what you will do with it!
Looking for additional learning materials?