Cloudera Docs

Testing the deployed workflow using APIs

The deployed workbench model provides a dedicated API for interaction.

To test the deployed workflow or integrate it with a custom application, use the following endpoint schema to communicate with the deployed workflows.

  1. In the Cloudera console, click the Cloudera AI tile. The Cloudera AI Home page displays.
  2. Click on the name of the workbench. The Cloudera AI Workbench Home page displays.
  3. Click Model Deployments in the left navigation pane. The Models page displays.
  4. Click a Model Name from the list. The Overview page displays for the selected model. You can see the Input field for the Test Model section.
    The deployed workbench model has its own API. To test or integrate the deployed workflow, use the endpoint schema below to interact with it. Requests are sent directly to the model endpoint using supported APIs ("action types"). The following code snippet is an example of a payload used to initiate a workflow task:
    MODEL_ENDPOINT = "https://modelservice.<CDSW_DOMAIN>/model?accessKey=<ACCESS_KEY>"
CDSW_APIV2_KEY = "..." # An API key used for authorization to the model endpoint

# Defined during workflow task creation
workflow_inputs = {
  "input1": "value_1"
}
inputs_encoded = base64.b64encode(json.dumps(workflow_inputs).encode("utf-8")).decode("utf-8")

payload = {
  "action_type": "kickoff",
  "kickoff_inputs": inputs_encoded
}

resp = requests.post(
  MODEL_ENDPOINT, 
  json={
    "request": payload
  },
  headers={"authorization": f"Bearer {CDSW_APIV2_KEY}", "Content-Type": "application/json"},
)
out = resp.json()
trace_id = out["response"]["trace_id"]

# This trace ID can be used to extract events from the Operations & Metrics server
print("trace_id: ", trace_id)