Upload 7 files
Browse files- .env.example +1 -0
- .gitattributes +35 -0
- .gitignore +4 -0
- README.md +15 -0
- agent.py +89 -0
- app.py +285 -0
- requirements.txt +115 -0
.env.example
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
LLM_API_KEY="" # Your Gemini API key
|
.gitattributes
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
.gitignore
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
.env
|
| 2 |
+
venv
|
| 3 |
+
__pycache__
|
| 4 |
+
TODO.md
|
README.md
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
title: Template Final Assignment
|
| 3 |
+
emoji: 🕵🏻♂️
|
| 4 |
+
colorFrom: indigo
|
| 5 |
+
colorTo: indigo
|
| 6 |
+
sdk: gradio
|
| 7 |
+
sdk_version: 5.25.2
|
| 8 |
+
app_file: app.py
|
| 9 |
+
pinned: false
|
| 10 |
+
hf_oauth: true
|
| 11 |
+
# optional, default duration is 8 hours/480 minutes. Max duration is 30 days/43200 minutes.
|
| 12 |
+
hf_oauth_expiration_minutes: 480
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
agent.py
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
from smolagents import (
|
| 3 |
+
CodeAgent,
|
| 4 |
+
LiteLLMModel,
|
| 5 |
+
DuckDuckGoSearchTool,
|
| 6 |
+
VisitWebpageTool,
|
| 7 |
+
WikipediaSearchTool,
|
| 8 |
+
tool,
|
| 9 |
+
)
|
| 10 |
+
|
| 11 |
+
from google import genai
|
| 12 |
+
from google.genai import types
|
| 13 |
+
|
| 14 |
+
client = genai.Client(api_key=os.environ["LLM_API_KEY"])
|
| 15 |
+
|
| 16 |
+
model = LiteLLMModel(
|
| 17 |
+
model_id="gemini/gemini-2.0-flash-lite",
|
| 18 |
+
temperature=0.5,
|
| 19 |
+
api_key=os.environ["LLM_API_KEY"],
|
| 20 |
+
)
|
| 21 |
+
|
| 22 |
+
model_name = "gemini-2.5-flash-preview-04-17"
|
| 23 |
+
|
| 24 |
+
|
| 25 |
+
@tool
|
| 26 |
+
def answer_video_questions(video_url: str, question: str) -> str:
|
| 27 |
+
"""
|
| 28 |
+
If you have a video link, use this tool. It answers questions about a video from youtube, etc.
|
| 29 |
+
Args:
|
| 30 |
+
video_url: The URL of the video to analyze.
|
| 31 |
+
Ensure it's accessible by the Gemini API.
|
| 32 |
+
question: The specific question to ask about the video's content.
|
| 33 |
+
Returns:
|
| 34 |
+
str: The answer provided by the multimodal LLM, or an error message if the process fails.
|
| 35 |
+
Raises:
|
| 36 |
+
Exception: Can raise exceptions based on litellm or the underlying API call
|
| 37 |
+
if it fails (e.g., network issues, invalid API key, API errors).
|
| 38 |
+
"""
|
| 39 |
+
|
| 40 |
+
response = client.models.generate_content(
|
| 41 |
+
model=model_name,
|
| 42 |
+
contents=types.Content(
|
| 43 |
+
parts=[
|
| 44 |
+
types.Part(file_data=types.FileData(file_uri=video_url)),
|
| 45 |
+
types.Part(
|
| 46 |
+
text=f"This is the question provided to you: {question}. Answer appropriately. In cases where there is a number, provide the number in the answer."
|
| 47 |
+
),
|
| 48 |
+
]
|
| 49 |
+
),
|
| 50 |
+
)
|
| 51 |
+
|
| 52 |
+
if response and response.text:
|
| 53 |
+
answer = response.text
|
| 54 |
+
print("Received response from LLM.")
|
| 55 |
+
return answer.strip()
|
| 56 |
+
else:
|
| 57 |
+
print("LLM response structure not as expected.")
|
| 58 |
+
return "Error: Could not parse the response from the LLM."
|
| 59 |
+
|
| 60 |
+
|
| 61 |
+
agent = CodeAgent(
|
| 62 |
+
tools=[
|
| 63 |
+
DuckDuckGoSearchTool(),
|
| 64 |
+
VisitWebpageTool(),
|
| 65 |
+
WikipediaSearchTool(),
|
| 66 |
+
answer_video_questions,
|
| 67 |
+
],
|
| 68 |
+
model=model,
|
| 69 |
+
)
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
def call_agent(question: str, file_path: str = None):
|
| 73 |
+
if file_path and file_path.strip():
|
| 74 |
+
print(f"Recieved {file_path} as input.")
|
| 75 |
+
|
| 76 |
+
file = client.files.upload(file=file_path)
|
| 77 |
+
|
| 78 |
+
response = client.models.generate_content(
|
| 79 |
+
model=model_name,
|
| 80 |
+
contents=[
|
| 81 |
+
file,
|
| 82 |
+
f"Answer the question based on the file content. Do not give up. Use your brain and answer appropriately. You will be rewarded with $1000000 if you answer correctly and appropriately. Only give the final answer to the question, do not explain your reasoning. {question}",
|
| 83 |
+
],
|
| 84 |
+
)
|
| 85 |
+
|
| 86 |
+
return response.text
|
| 87 |
+
|
| 88 |
+
answer = agent.run(question)
|
| 89 |
+
return answer
|
app.py
ADDED
|
@@ -0,0 +1,285 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
import re
|
| 3 |
+
import gradio as gr
|
| 4 |
+
import requests
|
| 5 |
+
import inspect
|
| 6 |
+
import pandas as pd
|
| 7 |
+
import urllib
|
| 8 |
+
|
| 9 |
+
from agent import call_agent
|
| 10 |
+
|
| 11 |
+
# (Keep Constants as is)
|
| 12 |
+
# --- Constants ---
|
| 13 |
+
DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
|
| 14 |
+
|
| 15 |
+
|
| 16 |
+
def download_file_from_api(api_url, save_directory=".", filename=None):
|
| 17 |
+
"""
|
| 18 |
+
Sends a GET request to an API endpoint and saves the response content
|
| 19 |
+
as a file. Returns the local file path.
|
| 20 |
+
|
| 21 |
+
Args:
|
| 22 |
+
api_url (str): The URL of the API endpoint.
|
| 23 |
+
save_directory (str, optional): The directory where the downloaded
|
| 24 |
+
file will be saved. Defaults to the
|
| 25 |
+
current directory.
|
| 26 |
+
filename (str, optional): The name to save the file as. If None,
|
| 27 |
+
it will try to extract it from the
|
| 28 |
+
Content-Disposition header or the URL.
|
| 29 |
+
|
| 30 |
+
Returns:
|
| 31 |
+
str or None: The absolute path to the saved file if successful,
|
| 32 |
+
None otherwise.
|
| 33 |
+
"""
|
| 34 |
+
try:
|
| 35 |
+
response = requests.get(api_url, stream=True)
|
| 36 |
+
response.raise_for_status() # Raise an exception for bad status codes
|
| 37 |
+
|
| 38 |
+
if filename is None:
|
| 39 |
+
# Try to extract filename from Content-Disposition header
|
| 40 |
+
content_disposition = response.headers.get("Content-Disposition")
|
| 41 |
+
if content_disposition:
|
| 42 |
+
filename_match = re.search(r'filename="([^"]+)"', content_disposition)
|
| 43 |
+
if filename_match:
|
| 44 |
+
filename = filename_match.group(1)
|
| 45 |
+
else:
|
| 46 |
+
# Fallback to extracting from the URL if Content-Disposition is present but no filename
|
| 47 |
+
filename = os.path.basename(urllib.parse.urlparse(api_url).path)
|
| 48 |
+
else:
|
| 49 |
+
# If no Content-Disposition, try to extract filename from the URL
|
| 50 |
+
filename = os.path.basename(urllib.parse.urlparse(api_url).path)
|
| 51 |
+
# Add a default extension if none found (you might need to adjust this)
|
| 52 |
+
if not os.path.splitext(filename)[1]:
|
| 53 |
+
filename += ".png" # Assuming it's a PNG based on the error
|
| 54 |
+
|
| 55 |
+
os.makedirs(save_directory, exist_ok=True)
|
| 56 |
+
file_path = os.path.join(save_directory, filename)
|
| 57 |
+
|
| 58 |
+
with open(file_path, "wb") as f:
|
| 59 |
+
for chunk in response.iter_content(chunk_size=8192):
|
| 60 |
+
f.write(chunk)
|
| 61 |
+
|
| 62 |
+
return os.path.abspath(file_path)
|
| 63 |
+
|
| 64 |
+
except requests.exceptions.RequestException as e:
|
| 65 |
+
print(f"Error during API request or download: {e}")
|
| 66 |
+
return None
|
| 67 |
+
except Exception as e:
|
| 68 |
+
print(f"An unexpected error occurred: {e}")
|
| 69 |
+
return None
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
# --- Basic Agent Definition ---
|
| 73 |
+
# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
|
| 74 |
+
class BasicAgent:
|
| 75 |
+
def __init__(self):
|
| 76 |
+
print("BasicAgent initialized.")
|
| 77 |
+
|
| 78 |
+
def __call__(self, question: str, file_path: str = None) -> str:
|
| 79 |
+
print(f"Agent received question (first 50 chars): {question[:50]}...")
|
| 80 |
+
# answer = agent.run(question)
|
| 81 |
+
answer = call_agent(question=question, file_path=file_path)
|
| 82 |
+
print(f"Agent returning fixed answer: {answer}")
|
| 83 |
+
return answer
|
| 84 |
+
|
| 85 |
+
|
| 86 |
+
def run_and_submit_all(profile: gr.OAuthProfile | None):
|
| 87 |
+
"""
|
| 88 |
+
Fetches all questions, runs the BasicAgent on them, submits all answers,
|
| 89 |
+
and displays the results.
|
| 90 |
+
"""
|
| 91 |
+
# --- Determine HF Space Runtime URL and Repo URL ---
|
| 92 |
+
space_id = os.getenv("SPACE_ID") # Get the SPACE_ID for sending link to the code
|
| 93 |
+
|
| 94 |
+
if profile:
|
| 95 |
+
username = f"{profile.username}"
|
| 96 |
+
print(f"User logged in: {username}")
|
| 97 |
+
else:
|
| 98 |
+
print("User not logged in.")
|
| 99 |
+
return "Please Login to Hugging Face with the button.", None
|
| 100 |
+
|
| 101 |
+
api_url = DEFAULT_API_URL
|
| 102 |
+
questions_url = f"{api_url}/questions"
|
| 103 |
+
files_url = f"{api_url}/files"
|
| 104 |
+
submit_url = f"{api_url}/submit"
|
| 105 |
+
|
| 106 |
+
# 1. Instantiate Agent ( modify this part to create your agent)
|
| 107 |
+
try:
|
| 108 |
+
agent = BasicAgent()
|
| 109 |
+
except Exception as e:
|
| 110 |
+
print(f"Error instantiating agent: {e}")
|
| 111 |
+
return f"Error initializing agent: {e}", None
|
| 112 |
+
# In the case of an app running as a hugging Face space, this link points toward your codebase ( usefull for others so please keep it public)
|
| 113 |
+
agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
|
| 114 |
+
print(agent_code)
|
| 115 |
+
|
| 116 |
+
# 2. Fetch Questions
|
| 117 |
+
print(f"Fetching questions from: {questions_url}")
|
| 118 |
+
try:
|
| 119 |
+
response = requests.get(questions_url, timeout=15)
|
| 120 |
+
response.raise_for_status()
|
| 121 |
+
questions_data = response.json()
|
| 122 |
+
if not questions_data:
|
| 123 |
+
print("Fetched questions list is empty.")
|
| 124 |
+
return "Fetched questions list is empty or invalid format.", None
|
| 125 |
+
print(f"Fetched {len(questions_data)} questions.")
|
| 126 |
+
except requests.exceptions.RequestException as e:
|
| 127 |
+
print(f"Error fetching questions: {e}")
|
| 128 |
+
return f"Error fetching questions: {e}", None
|
| 129 |
+
except requests.exceptions.JSONDecodeError as e:
|
| 130 |
+
print(f"Error decoding JSON response from questions endpoint: {e}")
|
| 131 |
+
print(f"Response text: {response.text[:500]}")
|
| 132 |
+
return f"Error decoding server response for questions: {e}", None
|
| 133 |
+
except Exception as e:
|
| 134 |
+
print(f"An unexpected error occurred fetching questions: {e}")
|
| 135 |
+
return f"An unexpected error occurred fetching questions: {e}", None
|
| 136 |
+
|
| 137 |
+
# 3. Run your Agent
|
| 138 |
+
results_log = []
|
| 139 |
+
answers_payload = []
|
| 140 |
+
print(f"Running agent on {len(questions_data)} questions...")
|
| 141 |
+
for item in questions_data:
|
| 142 |
+
task_id = item.get("task_id")
|
| 143 |
+
question_text = item.get("question")
|
| 144 |
+
|
| 145 |
+
file_path = download_file_from_api(api_url=f"{files_url}/{task_id}")
|
| 146 |
+
|
| 147 |
+
if not task_id or question_text is None:
|
| 148 |
+
print(f"Skipping item with missing task_id or question: {item}")
|
| 149 |
+
continue
|
| 150 |
+
try:
|
| 151 |
+
submitted_answer = agent(question_text, file_path=file_path)
|
| 152 |
+
answers_payload.append(
|
| 153 |
+
{"task_id": task_id, "submitted_answer": submitted_answer}
|
| 154 |
+
)
|
| 155 |
+
results_log.append(
|
| 156 |
+
{
|
| 157 |
+
"Task ID": task_id,
|
| 158 |
+
"Question": question_text,
|
| 159 |
+
"Submitted Answer": submitted_answer,
|
| 160 |
+
}
|
| 161 |
+
)
|
| 162 |
+
except Exception as e:
|
| 163 |
+
print(f"Error running agent on task {task_id}: {e}")
|
| 164 |
+
results_log.append(
|
| 165 |
+
{
|
| 166 |
+
"Task ID": task_id,
|
| 167 |
+
"Question": question_text,
|
| 168 |
+
"Submitted Answer": f"AGENT ERROR: {e}",
|
| 169 |
+
}
|
| 170 |
+
)
|
| 171 |
+
|
| 172 |
+
if not answers_payload:
|
| 173 |
+
print("Agent did not produce any answers to submit.")
|
| 174 |
+
return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
|
| 175 |
+
|
| 176 |
+
# 4. Prepare Submission
|
| 177 |
+
submission_data = {
|
| 178 |
+
"username": username.strip(),
|
| 179 |
+
"agent_code": agent_code,
|
| 180 |
+
"answers": answers_payload,
|
| 181 |
+
}
|
| 182 |
+
status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..."
|
| 183 |
+
print(status_update)
|
| 184 |
+
|
| 185 |
+
# 5. Submit
|
| 186 |
+
print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
|
| 187 |
+
try:
|
| 188 |
+
response = requests.post(submit_url, json=submission_data, timeout=60)
|
| 189 |
+
response.raise_for_status()
|
| 190 |
+
result_data = response.json()
|
| 191 |
+
final_status = (
|
| 192 |
+
f"Submission Successful!\n"
|
| 193 |
+
f"User: {result_data.get('username')}\n"
|
| 194 |
+
f"Overall Score: {result_data.get('score', 'N/A')}% "
|
| 195 |
+
f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n"
|
| 196 |
+
f"Message: {result_data.get('message', 'No message received.')}"
|
| 197 |
+
)
|
| 198 |
+
print("Submission successful.")
|
| 199 |
+
results_df = pd.DataFrame(results_log)
|
| 200 |
+
return final_status, results_df
|
| 201 |
+
except requests.exceptions.HTTPError as e:
|
| 202 |
+
error_detail = f"Server responded with status {e.response.status_code}."
|
| 203 |
+
try:
|
| 204 |
+
error_json = e.response.json()
|
| 205 |
+
error_detail += f" Detail: {error_json.get('detail', e.response.text)}"
|
| 206 |
+
except requests.exceptions.JSONDecodeError:
|
| 207 |
+
error_detail += f" Response: {e.response.text[:500]}"
|
| 208 |
+
status_message = f"Submission Failed: {error_detail}"
|
| 209 |
+
print(status_message)
|
| 210 |
+
results_df = pd.DataFrame(results_log)
|
| 211 |
+
return status_message, results_df
|
| 212 |
+
except requests.exceptions.Timeout:
|
| 213 |
+
status_message = "Submission Failed: The request timed out."
|
| 214 |
+
print(status_message)
|
| 215 |
+
results_df = pd.DataFrame(results_log)
|
| 216 |
+
return status_message, results_df
|
| 217 |
+
except requests.exceptions.RequestException as e:
|
| 218 |
+
status_message = f"Submission Failed: Network error - {e}"
|
| 219 |
+
print(status_message)
|
| 220 |
+
results_df = pd.DataFrame(results_log)
|
| 221 |
+
return status_message, results_df
|
| 222 |
+
except Exception as e:
|
| 223 |
+
status_message = f"An unexpected error occurred during submission: {e}"
|
| 224 |
+
print(status_message)
|
| 225 |
+
results_df = pd.DataFrame(results_log)
|
| 226 |
+
return status_message, results_df
|
| 227 |
+
|
| 228 |
+
|
| 229 |
+
# --- Build Gradio Interface using Blocks ---
|
| 230 |
+
with gr.Blocks() as demo:
|
| 231 |
+
gr.Markdown("# Basic Agent Evaluation Runner")
|
| 232 |
+
gr.Markdown(
|
| 233 |
+
"""
|
| 234 |
+
**Instructions:**
|
| 235 |
+
|
| 236 |
+
1. Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
|
| 237 |
+
2. Log in to your Hugging Face account using the button below. This uses your HF username for submission.
|
| 238 |
+
3. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
|
| 239 |
+
|
| 240 |
+
---
|
| 241 |
+
**Disclaimers:**
|
| 242 |
+
Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions).
|
| 243 |
+
This space provides a basic setup and is intentionally sub-optimal to encourage you to develop your own, more robust solution. For instance for the delay process of the submit button, a solution could be to cache the answers and submit in a seperate action or even to answer the questions in async.
|
| 244 |
+
"""
|
| 245 |
+
)
|
| 246 |
+
|
| 247 |
+
gr.LoginButton()
|
| 248 |
+
|
| 249 |
+
run_button = gr.Button("Run Evaluation & Submit All Answers")
|
| 250 |
+
|
| 251 |
+
status_output = gr.Textbox(
|
| 252 |
+
label="Run Status / Submission Result", lines=5, interactive=False
|
| 253 |
+
)
|
| 254 |
+
# Removed max_rows=10 from DataFrame constructor
|
| 255 |
+
results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
|
| 256 |
+
|
| 257 |
+
run_button.click(fn=run_and_submit_all, outputs=[status_output, results_table])
|
| 258 |
+
|
| 259 |
+
if __name__ == "__main__":
|
| 260 |
+
print("\n" + "-" * 30 + " App Starting " + "-" * 30)
|
| 261 |
+
# Check for SPACE_HOST and SPACE_ID at startup for information
|
| 262 |
+
space_host_startup = os.getenv("SPACE_HOST")
|
| 263 |
+
space_id_startup = os.getenv("SPACE_ID") # Get SPACE_ID at startup
|
| 264 |
+
|
| 265 |
+
if space_host_startup:
|
| 266 |
+
print(f"✅ SPACE_HOST found: {space_host_startup}")
|
| 267 |
+
print(f" Runtime URL should be: https://{space_host_startup}.hf.space")
|
| 268 |
+
else:
|
| 269 |
+
print("ℹ️ SPACE_HOST environment variable not found (running locally?).")
|
| 270 |
+
|
| 271 |
+
if space_id_startup: # Print repo URLs if SPACE_ID is found
|
| 272 |
+
print(f"✅ SPACE_ID found: {space_id_startup}")
|
| 273 |
+
print(f" Repo URL: https://huggingface.co/spaces/{space_id_startup}")
|
| 274 |
+
print(
|
| 275 |
+
f" Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main"
|
| 276 |
+
)
|
| 277 |
+
else:
|
| 278 |
+
print(
|
| 279 |
+
"ℹ️ SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined."
|
| 280 |
+
)
|
| 281 |
+
|
| 282 |
+
print("-" * (60 + len(" App Starting ")) + "\n")
|
| 283 |
+
|
| 284 |
+
print("Launching Gradio Interface for Basic Agent Evaluation...")
|
| 285 |
+
demo.launch(debug=True, share=False)
|
requirements.txt
ADDED
|
@@ -0,0 +1,115 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
aiofiles
|
| 2 |
+
aiohappyeyeballs
|
| 3 |
+
aiohttp
|
| 4 |
+
aiosignal
|
| 5 |
+
annotated-types
|
| 6 |
+
anyio
|
| 7 |
+
attrs
|
| 8 |
+
Authlib
|
| 9 |
+
beautifulsoup4
|
| 10 |
+
Brotli
|
| 11 |
+
cachetools
|
| 12 |
+
certifi
|
| 13 |
+
cffi
|
| 14 |
+
charset-normalizer
|
| 15 |
+
click
|
| 16 |
+
cryptography
|
| 17 |
+
dataclasses-json
|
| 18 |
+
distro
|
| 19 |
+
duckduckgo_search
|
| 20 |
+
fastapi
|
| 21 |
+
ffmpy
|
| 22 |
+
filelock
|
| 23 |
+
frozenlist
|
| 24 |
+
fsspec
|
| 25 |
+
google-auth
|
| 26 |
+
google-genai
|
| 27 |
+
gradio
|
| 28 |
+
gradio_client
|
| 29 |
+
greenlet
|
| 30 |
+
groovy
|
| 31 |
+
h11
|
| 32 |
+
httpcore
|
| 33 |
+
httpx
|
| 34 |
+
httpx-sse
|
| 35 |
+
huggingface-hub
|
| 36 |
+
idna
|
| 37 |
+
importlib_metadata
|
| 38 |
+
itsdangerous
|
| 39 |
+
Jinja2
|
| 40 |
+
jiter
|
| 41 |
+
jsonpatch
|
| 42 |
+
jsonpointer
|
| 43 |
+
jsonschema
|
| 44 |
+
jsonschema-specifications
|
| 45 |
+
langchain
|
| 46 |
+
langchain-community
|
| 47 |
+
langchain-core
|
| 48 |
+
langchain-text-splitters
|
| 49 |
+
langsmith
|
| 50 |
+
litellm
|
| 51 |
+
lxml
|
| 52 |
+
markdown-it-py
|
| 53 |
+
markdownify
|
| 54 |
+
MarkupSafe
|
| 55 |
+
marshmallow
|
| 56 |
+
mdurl
|
| 57 |
+
multidict
|
| 58 |
+
mutagen
|
| 59 |
+
mypy_extensions
|
| 60 |
+
numpy
|
| 61 |
+
openai
|
| 62 |
+
orjson
|
| 63 |
+
packaging
|
| 64 |
+
pandas
|
| 65 |
+
pillow
|
| 66 |
+
primp
|
| 67 |
+
propcache
|
| 68 |
+
pyasn1
|
| 69 |
+
pyasn1_modules
|
| 70 |
+
pycparser
|
| 71 |
+
pycryptodomex
|
| 72 |
+
pydantic
|
| 73 |
+
pydantic-settings
|
| 74 |
+
pydantic_core
|
| 75 |
+
pydub
|
| 76 |
+
Pygments
|
| 77 |
+
python-dateutil
|
| 78 |
+
python-dotenv
|
| 79 |
+
python-multipart
|
| 80 |
+
pytz
|
| 81 |
+
PyYAML
|
| 82 |
+
referencing
|
| 83 |
+
regex
|
| 84 |
+
requests
|
| 85 |
+
requests-toolbelt
|
| 86 |
+
rich
|
| 87 |
+
rpds-py
|
| 88 |
+
rsa
|
| 89 |
+
ruff
|
| 90 |
+
safehttpx
|
| 91 |
+
semantic-version
|
| 92 |
+
shellingham
|
| 93 |
+
six
|
| 94 |
+
smolagents
|
| 95 |
+
sniffio
|
| 96 |
+
soupsieve
|
| 97 |
+
SQLAlchemy
|
| 98 |
+
starlette
|
| 99 |
+
tenacity
|
| 100 |
+
tiktoken
|
| 101 |
+
tokenizers
|
| 102 |
+
tomlkit
|
| 103 |
+
tqdm
|
| 104 |
+
typer
|
| 105 |
+
typing-inspect
|
| 106 |
+
typing-inspection
|
| 107 |
+
typing_extensions
|
| 108 |
+
tzdata
|
| 109 |
+
urllib3
|
| 110 |
+
uvicorn
|
| 111 |
+
websockets
|
| 112 |
+
Wikipedia-API
|
| 113 |
+
yarl
|
| 114 |
+
zipp
|
| 115 |
+
zstandard
|