The model task
NOT_SET can be used if Fiddler doesn't provide the model task needed or if XAI and scoring functionalities are not necessary. This model task doesn't have any restrictions for the outputs and targets field, meaning those can be omitted or specified for any columns (no restriction on the number or type of columns).
Suppose you would like to onboard a model for the following dataset.
Following is an example of how you could construct a
fdl.ModelInfo object and onboard such a model.
PROJECT_ID = 'example_project'
DATASET_ID = 'wine_data'
MODEL_ID = 'example_model'
dataset_info = client.get_dataset_info(
model_task = fdl.ModelTask.NOT_SET
model_info = fdl.ModelInfo.from_dataset_info(
Updated 4 months ago