|
|
|
|
|
|
|
|
from smolagents import tool |
|
|
from openai import OpenAI |
|
|
import dotenv |
|
|
dotenv.load_dotenv() |
|
|
|
|
|
@tool |
|
|
def analyze_image(question: str, image_url: str) -> str: |
|
|
""" |
|
|
Analyze an image using OpenAI's API. |
|
|
Args: |
|
|
question (str): The question to ask about the image. eg. "What is in this image?" |
|
|
image_url (str): The URL of the image to analyze. |
|
|
""" |
|
|
client = OpenAI() |
|
|
|
|
|
response = client.responses.create( |
|
|
model="gpt-4o-mini", |
|
|
input=[ |
|
|
{ |
|
|
"role": "user", |
|
|
"content": [ |
|
|
{ "type": "input_text", "text": f"{question}" }, |
|
|
{ |
|
|
"type": "input_image", |
|
|
"image_url": f"{image_url}", |
|
|
} |
|
|
] |
|
|
} |
|
|
] |
|
|
) |
|
|
|
|
|
return response |
|
|
|
|
|
if __name__ == "__main__": |
|
|
question = "What is the main subject of this image?" |
|
|
image_url = "https://agents-course-unit4-scoring.hf.space/files/cca530fc-4052-43b2-b130-b30968d8aa44" |
|
|
answer = analyze_image(question, image_url) |
|
|
print(f"Answer: {answer}") |
