Databricks Integration

Fiddler allows your team to monitor, explain and analyze your models developed and deployed in Databricks Workspace by integrating with MLFlow for model asset management and utilizing Databricks Spark environment for data management.

To validate and monitor models built on Databricks using Fiddler, you can follow these steps:

1. Creating a Fiddler Project

2. Uploading a Baseline Dataset

3. Adding Model Information

4. Uploading Model Files (for Explainability)

5. Publishing Events

   1. Batch Models

   2. Live Models

Creating a Fiddler Project

Launch a Databricks notebook from your workspace and run the following code:

!pip install -q fiddler-client
import fiddler as fdl

Now that you have the Fiddler library installed, you can connect to your Fiddler environment. Please use the UI administration guide to help you find your Fiddler credentials.

URL = ""
ORG_ID = ""
AUTH_TOKEN = ""
client = fdl.FiddlerApi(url=URL, org_id=ORG_ID, auth_token=AUTH_TOKEN)

Finally, you can set up a new project using:

client.create_project("YOUR_PROJECT_NAME")

Uploading a Baseline Dataset

You can grab your baseline dataset from a delta table and share it with Fiddler as a baseline dataset:

baseline_dataset = spark.read.table("YOUR_DATASET").select("*").toPandas()

dataset_info = fdl.DatasetInfo.from_dataframe(baseline_upload, max_inferred_cardinality=100)
  
client.upload_dataset(
  project_id=PROJECT_ID,
  dataset_id=DATASET_ID,
  dataset={'baseline': baseline_upload},
  info=dataset_info)

Adding Model Information

Using the **MLFlow API ** you can query the model registry and get the model signature which describes the inputs and outputs as a dictionary. You can use this dictionary to build out the ModelInfo object required to the model to Fiddler:

import mlflow 
from mlflow.tracking import MlflowClient

client = MlflowClient() #initiate MLFlow Client 

#Get the model URI
model_version_info = client.get_model_version(model_name, model_version)
model_uri = client.get_model_version_download_uri(model_name, model_version_info) 

#Get the Model Signature
mlflow_model_info = mlflow.models.get_model_info(model_uri)
model_inputs_schema = model_info.signature.inputs.to_dict()
model_inputs = [ sub['name'] for sub in model_inputs_schema ]

Now you can use the model signature to build the Fiddler ModelInfo object :

features = model_inputs

model_task = fdl.ModelTask.BINARY_CLASSIFICATION

model_info = fdl.ModelInfo.from_dataset_info(
	dataset_info = client.get_dataset_info(YOUR_PROJECT,YOUR_DATASET),
	target =  "TARGET COLUMN", 
  dataset_id=DATASET_ID,
  model_task=model_task, 
  features=features,
  outputs=['output_column'])

Uploading Model Files

Sharing your model artifacts helps Fiddler explain your models. By leveraging the MLFlow API you can download these model files:

import os  
import mlflow  
from mlflow.store.artifact.models_artifact_repo import ModelsArtifactRepository

model_name = "example-model-name"  
model_stage = "Staging"  # Should be either 'Staging' or 'Production'

mlflow.set_tracking_uri("databricks")  
os.makedirs("model", exist_ok=True)  
local_path = ModelsArtifactRepository(
  f'models:/{model_name}/{model_stage}').download_artifacts("", dst_path="model")  

print(f'{model_stage} Model {model_name} is downloaded at {local_path}')  

Once you have the model file, you can create a package.py file in this model directory that describes how to access this model.

Finally, you can upload all the model artifacts to Fiddler:

client.add_model_artifact(  
    project_id=PROJECT_ID,
    model_id=MODEL_ID,
    model_dir='model/',
)

Alternatively, you can skip uploading your model and use Fiddler to generate a surrogate model to get low-fidelity explanations for your model.

Publishing Events

Now you can publish all the events from your models. You can do this in two ways:

Batch Models

If your models run batch processes with your models or your aggregate model outputs over a timeframe, then you can use the table change feed from Databricks to select only the new events and send them to Fiddler:

changes_df = spark.read.format("delta") \
.option("readChangeFeed", "true") \
.option("startingVersion",last_version) \
.option("endingVersion", new_version) \
.table("inferences").toPandas()


client.publish_events_batch(
   project_id=PROJECT_ID,
   model_id=MODEL_ID,
   batch_source=changes_df,
   timestamp_field='timestamp')

Live Models

For models with live predictions or real-time applications, you can add the following code snippet to your prediction pipeline and send every event to Fiddler in real-time:

example_event = model_output.toPandas() #turn your model's ouput in a pandas datafram 

client.publish_event(
    project_id=PROJECT_ID,
    model_id=MODEL_ID,
    event=example_event,
    event_id='event_001',
    event_timestamp=1637344470000)

Support for Inference tables and hosted endpoints is coming soon!