Suppose you would like to onboard a multiclass classification model for the following dataset.
Following is an example of how you would construct a
fdl.ModelInfo object and onboard such a model.
For multiclass models, the
categorical_target_class_detailsargument is required.
This argument should be a list of your target classes in the order that your model outputs predictions for them.
PROJECT_ID = 'example_project' DATASET_ID = 'iris_data' MODEL_ID = 'multiclass_model' dataset_info = client.get_dataset_info( project_id=PROJECT_ID, dataset_id=DATASET_ID ) model_task = fdl.ModelTask.MULTICLASS_CLASSIFICATION model_target = 'species' model_outputs = [ 'probability_0', 'probability_1', 'probability_2' ] model_features = [ 'sepal_length', 'sepal_width', 'petal_length', 'petal_width' ] model_info = fdl.ModelInfo.from_dataset_info( dataset_info=dataset_info, dataset_id=DATASET_ID, target=model_target, outputs=model_outputs, model_task=model_task, categorical_target_class_details=[0, 1, 2] ) client.add_model( project_id=PROJECT_ID, dataset_id=DATASET_ID, model_id=MODEL_ID, model_info=model_info )
Using client.add_model() does not provide Fiddler with a model artifact. Onboarding a model in this fashion is a good start for model monitoring, but Fiddler will not be able to offer model explainability features without a model artifact. You can subsequently call client.add_model_surrogate() or client.add_model_artifact() to provide Fiddler with a model artifact. Please see Uploading a Model Artifact for more information.
Updated about 2 months ago