File size: 1,058 Bytes
b70c4a4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import os
import requests
from langchain_core.messages import HumanMessage
from agent import build_graph
from huggingface_hub import hf_hub_download
import pyarrow.parquet as pq
from dotenv import load_dotenv

load_dotenv(override=True)

DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"

graph = build_graph()
resp = requests.get(f"{DEFAULT_API_URL}/questions")
questions = resp.json()

token = os.getenv("HF_TOKEN") or os.getenv("HUGGINGFACEHUB_API_TOKEN")
path = hf_hub_download(repo_id='gaia-benchmark/GAIA', filename='2023/validation/metadata.parquet', repo_type='dataset', token=token)
df = pq.read_table(path).to_pandas()
answer_map = dict(zip(df['task_id'], df['Final answer']))

# Show questions with files
for i in [3, 9, 11, 13, 18]:
    q = questions[i]
    task_id = q['task_id']
    question = q['question']
    ground_truth = answer_map.get(task_id, "NOT FOUND")
    file_name = q.get('file_name', '')
    
    print(f"\n=== Q{i+1} | File: {file_name} ===")
    print(f"Q: {question[:100]}...")
    print(f"GT: {ground_truth}")