File: sample_responses_structured_output.py

package info (click to toggle)
python-azure 20251118%2Bgit-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 783,356 kB
  • sloc: python: 6,474,533; ansic: 804; javascript: 287; sh: 205; makefile: 198; xml: 109
file content (67 lines) | stat: -rw-r--r-- 2,326 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
# ------------------------------------
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT License.
# ------------------------------------

"""
DESCRIPTION:
    This sample demonstrates how to run a basic responses operation
    using the synchronous AIProject and OpenAI clients, while defining
    a desired JSON schema for the response ("structured output").

    This sample is inspired by the OpenAI example here:
    https://platform.openai.com/docs/guides/structured-outputs/supported-schemas

USAGE:
    python sample_responses_structured_output.py

    Before running the sample:

    pip install "azure-ai-projects>=2.0.0b1" openai azure-identity python-dotenv

    Set these environment variables with your own values:
    1) AZURE_AI_PROJECT_ENDPOINT - The Azure AI Project endpoint, as found in the Overview
       page of your Microsoft Foundry portal.
    2) AZURE_AI_MODEL_DEPLOYMENT_NAME - The deployment name of the AI model, as found under the "Name" column in
       the "Models + endpoints" tab in your Microsoft Foundry project.
"""

import os
from dotenv import load_dotenv
from azure.identity import DefaultAzureCredential
from azure.ai.projects import AIProjectClient
from pydantic import BaseModel, Field

load_dotenv()


class CalendarEvent(BaseModel):
    model_config = {"extra": "forbid"}
    name: str
    date: str = Field(description="Date in YYYY-MM-DD format")
    participants: list[str]


endpoint = os.environ["AZURE_AI_PROJECT_ENDPOINT"]

with (
    DefaultAzureCredential() as credential,
    AIProjectClient(endpoint=endpoint, credential=credential) as project_client,
    project_client.get_openai_client() as openai_client,
):
    response = openai_client.responses.create(
        model=os.environ["AZURE_AI_MODEL_DEPLOYMENT_NAME"],
        instructions="""
            Extracts calendar event information from the input messages,
            and return it in the desired structured output format.
            """,
        text={
            "format": {
                "type": "json_schema",
                "name": "CalendarEvent",
                "schema": CalendarEvent.model_json_schema(),
            }
        },
        input="Alice and Bob are going to a science fair this Friday, November 7, 2025.",
    )
    print(f"Response output: {response.output_text}")