Specifying Custom Features

"Patented Fiddler Technology"

Vector Monitoring for Unstructured Data

CF1 = fdl.CustomFeature.from_columns(['f1','f2','f3'], custom_name = 'vector1')
CF2 = fdl.CustomFeature.from_columns(['f1','f2','f3'], n_clusters=5, custom_name = 'vector2')
CF3 = fdl.TextEmbedding(name='text_embedding',column='embedding',source_column='text')
CF4 = fdl.ImageEmbedding(name='image_embedding',column='embedding',source_column='image_url')

Passing Custom Features List to Model Spec

model_spec = fdl.ModelSpec(
    inputs=['CreditScore', 'Geography', 'Gender', 'Age', 'Tenure', 'Balance'],
    outputs=['probability_churned'],
    targets=['Churned'],
    decisions=[],
    metadata=[],
    custom_features=[CF1,CF2,CF3,CF4],
)

📘
Quick Start for NLP Monitoring
Check out our Quick Start guide for NLP monitoring for a fully functional notebook example.

Updated 9 days ago