LoslegenKostenlos starten

Query chatbot using Databricks Servings Endpoint API

In the previous exercise, you learned how to use the Databricks SDK to query an AI Foundation model. The goal of this exercise is for you to experiment with different user and system prompts when querying AI models using the Databricks SDK

Diese Übung ist Teil des Kurses

<Kurs>Databricks with the Python SDK</Kurs>
Kurs ansehen

Übungsanweisungen

  • Edit the System prompt and see how the AI model responds differently.
  • Edit the User prompt and see how the AI model responds differently.

Interaktive praktische Übung

Versuche dich an dieser Übung, indem du diesen Beispielcode vervollständigst.

from databricks.sdk import WorkspaceClient
from databricks.sdk.service.serving import ChatMessage, ChatMessageRole
w = WorkspaceClient()
response = w.serving_endpoints.query(
    name="databricks-meta-llama-3-3-70b-instruct",
    messages=[
        ChatMessage( 
          	# Edit the System prompt and see how the AI model responds differently
            role=ChatMessageRole.SYSTEM, content=""
        ),
        ChatMessage( 
          	# Edit the User prompt and see how the AI model responds differently
            role=ChatMessageRole.USER, content="" 
        ),
    ],
    max_tokens=128)
print(f"RESPONSE:\n{response.choices[0].message.content}")
Code bearbeiten und ausführen