Skip to main content

Task 7: Model diagnostics

In this task, you will perform model diagnostics to evaluate the performance of the model on the test set. Running diagnostics helps assess various metrics and gain insights into how well the model performs and where improvements may be needed. The diagnostics include important performance scores such as AUC, precision, recall, and others, along with insights into the model's behavior.

To run the diagnostics on the experiment and test dataset, use the following command:

experiment_diagnostics = dai.model_diagnostics.create(experiment, test)

The experiment_diagnostics object contains detailed performance metrics and insights for the trained model.

Feedback