File size: 3,609 Bytes
0412d41 0af627b 0412d41 0af627b 0412d41 0af627b 0412d41 0af627b 0412d41 0af627b 63d6133 0af627b 63d6133 0412d41 0af627b 0412d41 2ae2c93 0af627b 2ae2c93 63d6133 a058371 0af627b 0412d41 2ae2c93 0412d41 0af627b a058371 0412d41 0af627b 63d6133 a058371 0412d41 63d6133 0af627b a058371 0af627b 0412d41 0af627b 2ae2c93 a058371 0af627b 0412d41 a058371 0412d41 a058371 0412d41 63d6133 0412d41 63d6133 0af627b 63d6133 0412d41 63d6133 0af627b 63d6133 0412d41 |
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 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 |
import os
import openai # 🍕 https://www.google.com/search?q=openai.ChatCompletion.create
import gradio as gr # 🍣 https://www.google.com/search?q=Gradio+python
import requests
import pandas as pd
# --- Configuration ---
openai.api_key = os.getenv("OPENAI_API_KEY") # 🍱 https://www.google.com/search?q=python+environment+variables
API_URL = "https://agents-course-unit4-scoring.hf.space"
MODEL_NAME = "gpt-4.1" # 🍜 https://www.google.com/search?q=gpt-4.1+openai+model+id
# --- ChatGPT-4.1 Caller ---
def ask_chatgpt_4_1(question: str) -> str:
response = openai.ChatCompletion.create(
model=MODEL_NAME,
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": question}
],
temperature=0.7,
max_tokens=1500
)
return response.choices[0].message.content
# --- Agent Class ---
class BasicAgent:
def __init__(self):
print("BasicAgent using OpenAI GPT-4.1 ready.")
def __call__(self, question: str) -> str:
print(f"Q>> {question}")
try:
return ask_chatgpt_4_1(question)
except Exception as e:
return f"AGENT ERROR: {e}"
# --- Evaluation & Submission ---
def run_and_submit_all(profile: gr.OAuthProfile | None):
if not profile:
return "Please log in to Hugging Face.", None
username = profile.username
space_id = os.getenv("SPACE_ID", "")
agent = BasicAgent()
code_link = f"https://huggingface.co/spaces/{space_id}/tree/main"
# 1. Fetch questions
try:
resp = requests.get(f"{API_URL}/questions", timeout=15)
resp.raise_for_status()
questions = resp.json()
if not questions:
return "No questions fetched.", None
except Exception as e:
return f"Error fetching questions: {e}", None
# 2. Answer each
logs = []
payload = []
for item in questions:
tid = item.get("task_id")
q = item.get("question")
if not tid or q is None:
continue
ans = agent(q)
logs.append({"Task ID": tid, "Question": q, "Answer": ans})
payload.append({"task_id": tid, "submitted_answer": ans})
if not payload:
return "No answers generated.", pd.DataFrame(logs)
# 3. Submit
submission = {
"username": username,
"agent_code": code_link,
"answers": payload
}
try:
post = requests.post(f"{API_URL}/submit", json=submission, timeout=60)
post.raise_for_status()
res = post.json()
status = (
f"Success! {res.get('username')} scored "
f"{res.get('score', 'N/A')}% "
f"({res.get('correct_count')}/{res.get('total_attempted')})"
)
return status, pd.DataFrame(logs)
except Exception as e:
return f"Submission Failed: {e}", pd.DataFrame(logs)
# --- Gradio Interface ---
with gr.Blocks() as demo:
gr.Markdown("# SmolAgent GAIA Evaluation Runner 🚀")
gr.Markdown(
"""
1. Clone space and modify if needed
2. Log in to Hugging Face
3. Click **Run Evaluation & Submit All Answers**
(May take several minutes)
"""
)
gr.LoginButton()
run_btn = gr.Button("Run Evaluation & Submit All Answers")
status_out = gr.Textbox(label="Status", lines=5, interactive=False)
table_out = gr.DataFrame(label="Q&A Log", wrap=True)
run_btn.click(fn=run_and_submit_all, outputs=[status_out, table_out])
if __name__ == "__main__":
demo.launch(debug=True, share=False)
|