Spaces:
Sleeping
Sleeping
Back to Gradio
Browse files
app.py
CHANGED
|
@@ -1,6 +1,5 @@
|
|
| 1 |
import os
|
| 2 |
-
|
| 3 |
-
# import gradio as gr
|
| 4 |
import requests
|
| 5 |
import inspect
|
| 6 |
import pandas as pd
|
|
@@ -79,8 +78,7 @@ class BasicAgent:
|
|
| 79 |
return final_answer
|
| 80 |
|
| 81 |
|
| 82 |
-
|
| 83 |
-
async def run_and_submit_all():
|
| 84 |
"""
|
| 85 |
Fetches all questions, runs the BasicAgent on them, submits all answers,
|
| 86 |
and displays the results.
|
|
@@ -88,12 +86,12 @@ async def run_and_submit_all():
|
|
| 88 |
# --- Determine HF Space Runtime URL and Repo URL ---
|
| 89 |
space_id = os.getenv("SPACE_ID") # Get the SPACE_ID for sending link to the code
|
| 90 |
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 97 |
username = "benjosaur"
|
| 98 |
|
| 99 |
api_url = DEFAULT_API_URL
|
|
@@ -226,181 +224,90 @@ async def run_and_submit_all():
|
|
| 226 |
|
| 227 |
|
| 228 |
# # --- Build Gradio Interface using Blocks ---
|
| 229 |
-
|
| 230 |
-
|
| 231 |
-
|
| 232 |
-
|
| 233 |
-
|
| 234 |
-
|
| 235 |
-
# 1. Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
|
| 236 |
-
# 2. Log in to your Hugging Face account using the button below. This uses your HF username for submission.
|
| 237 |
-
# 3. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
|
| 238 |
-
|
| 239 |
-
# ---
|
| 240 |
-
# **Disclaimers:**
|
| 241 |
-
# 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).
|
| 242 |
-
# 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.
|
| 243 |
-
# """
|
| 244 |
-
# )
|
| 245 |
-
|
| 246 |
-
# gr.LoginButton()
|
| 247 |
-
|
| 248 |
-
# run_button = gr.Button("Run Evaluation & Submit All Answers")
|
| 249 |
-
|
| 250 |
-
# status_output = gr.Textbox(
|
| 251 |
-
# label="Run Status / Submission Result", lines=5, interactive=False
|
| 252 |
-
# )
|
| 253 |
-
# # Removed max_rows=10 from DataFrame constructor
|
| 254 |
-
# results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
|
| 255 |
-
|
| 256 |
-
# run_button.click(fn=run_and_submit_all, outputs=[status_output, results_table])
|
| 257 |
-
|
| 258 |
-
# async def main():
|
| 259 |
-
# agent = BasicAgent()
|
| 260 |
-
# api_url = DEFAULT_API_URL
|
| 261 |
-
# questions_url = f"{api_url}/questions"
|
| 262 |
-
# print(f"Fetching questions from: {questions_url}")
|
| 263 |
-
|
| 264 |
-
# response = requests.get(questions_url, timeout=15)
|
| 265 |
-
# response.raise_for_status()
|
| 266 |
-
# questions_data = response.json()
|
| 267 |
-
|
| 268 |
-
# # 3. Run your Agent
|
| 269 |
-
# results_log = []
|
| 270 |
-
# answers_payload = []
|
| 271 |
-
# print(f"Running agent on {len(questions_data)} questions...")
|
| 272 |
-
# item = questions_data[0]
|
| 273 |
-
# task_id = item.get("task_id")
|
| 274 |
-
# question_text = item.get("question")
|
| 275 |
-
# file_name = item.get("file_name")
|
| 276 |
-
# submitted_answer = await agent(question_text, task_id, file_name)
|
| 277 |
-
# answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
|
| 278 |
-
# results_log.append(
|
| 279 |
-
# {
|
| 280 |
-
# "Task ID": task_id,
|
| 281 |
-
# "Question": question_text,
|
| 282 |
-
# "Submitted Answer": submitted_answer,
|
| 283 |
-
# }
|
| 284 |
-
# )
|
| 285 |
|
| 286 |
-
|
| 287 |
-
|
| 288 |
-
|
| 289 |
-
# space_host_startup = os.getenv("SPACE_HOST")
|
| 290 |
-
# space_id_startup = os.getenv("SPACE_ID") # Get SPACE_ID at startup
|
| 291 |
-
|
| 292 |
-
# if space_host_startup:
|
| 293 |
-
# print(f"✅ SPACE_HOST found: {space_host_startup}")
|
| 294 |
-
# print(f" Runtime URL should be: https://{space_host_startup}.hf.space")
|
| 295 |
-
# else:
|
| 296 |
-
# print("ℹ️ SPACE_HOST environment variable not found (running locally?).")
|
| 297 |
-
|
| 298 |
-
# if space_id_startup: # Print repo URLs if SPACE_ID is found
|
| 299 |
-
# print(f"✅ SPACE_ID found: {space_id_startup}")
|
| 300 |
-
# print(f" Repo URL: https://huggingface.co/spaces/{space_id_startup}")
|
| 301 |
-
# print(
|
| 302 |
-
# f" Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main"
|
| 303 |
-
# )
|
| 304 |
-
# else:
|
| 305 |
-
# print(
|
| 306 |
-
# "ℹ️ SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined."
|
| 307 |
-
# )
|
| 308 |
-
|
| 309 |
-
# print("-" * (60 + len(" App Starting ")) + "\n")
|
| 310 |
-
|
| 311 |
-
# print("Launching Gradio Interface for Basic Agent Evaluation...")
|
| 312 |
-
# demo.launch(debug=True, share=False)
|
| 313 |
-
# asyncio.run(run_and_submit_all())
|
| 314 |
|
| 315 |
-
|
| 316 |
-
|
| 317 |
-
|
| 318 |
-
|
| 319 |
-
|
| 320 |
-
|
| 321 |
-
|
| 322 |
-
|
| 323 |
-
|
| 324 |
-
|
| 325 |
-
|
| 326 |
-
|
| 327 |
-
|
| 328 |
-
|
| 329 |
-
|
| 330 |
-
|
| 331 |
-
},
|
| 332 |
-
{
|
| 333 |
-
"task_id": "6f37996b-2ac7-44b0-8e68-6d28256631b4",
|
| 334 |
-
"submitted_answer": "a, b, c",
|
| 335 |
-
},
|
| 336 |
-
{
|
| 337 |
-
"task_id": "9d191bce-651d-4746-be2d-7ef8ecadb9c2",
|
| 338 |
-
"submitted_answer": "Extremely",
|
| 339 |
-
},
|
| 340 |
-
{
|
| 341 |
-
"task_id": "cabe07ed-9eca-40ea-8ead-410ef5e83f91",
|
| 342 |
-
"submitted_answer": "Undetermined",
|
| 343 |
-
},
|
| 344 |
-
{
|
| 345 |
-
"task_id": "3cef3a44-215e-4aed-8e3b-b1e3f08063b7",
|
| 346 |
-
"submitted_answer": "broccoli, celery, corn, green beans, lettuce, sweet potatoes, zucchini",
|
| 347 |
-
},
|
| 348 |
-
{
|
| 349 |
-
"task_id": "99c9cc74-fdc8-46c6-8f8d-3ce2d3bfeea3",
|
| 350 |
-
"submitted_answer": "cornstarch, freshly squeezed lemon juice, granulated sugar, pure vanilla extract, ripe strawberries",
|
| 351 |
-
},
|
| 352 |
-
{
|
| 353 |
-
"task_id": "305ac316-eef6-4446-960a-92d80d542f82",
|
| 354 |
-
"submitted_answer": "Piotr",
|
| 355 |
-
},
|
| 356 |
-
{
|
| 357 |
-
"task_id": "3f57289b-8c60-48be-bd80-01f8099ca449",
|
| 358 |
-
"submitted_answer": "525",
|
| 359 |
-
},
|
| 360 |
-
{
|
| 361 |
-
"task_id": "1f975693-876d-457b-a649-393859e79bf3",
|
| 362 |
-
"submitted_answer": "132, 133, 134, 197, 245",
|
| 363 |
-
},
|
| 364 |
-
{
|
| 365 |
-
"task_id": "840bfca7-4f7b-481a-8794-c560c340185d",
|
| 366 |
-
"submitted_answer": "unknown",
|
| 367 |
-
},
|
| 368 |
-
{
|
| 369 |
-
"task_id": "bda648d7-d618-4883-88f4-3466eabd860e",
|
| 370 |
-
"submitted_answer": "Saint Petersburg",
|
| 371 |
-
},
|
| 372 |
-
{
|
| 373 |
-
"task_id": "cf106601-ab4f-4af9-b045-5295fe67b37d",
|
| 374 |
-
"submitted_answer": "LIE",
|
| 375 |
-
},
|
| 376 |
-
{
|
| 377 |
-
"task_id": "a0c07678-e491-4bbc-8f0b-07405144218f",
|
| 378 |
-
"submitted_answer": "Itoh, Unknown",
|
| 379 |
-
},
|
| 380 |
-
{
|
| 381 |
-
"task_id": "7bd855d8-463d-4ed5-93ca-5fe35145f733",
|
| 382 |
-
"submitted_answer": "56973.00",
|
| 383 |
-
},
|
| 384 |
-
{
|
| 385 |
-
"task_id": "5a0c1adf-205e-4841-a666-7c3ef95def9d",
|
| 386 |
-
"submitted_answer": "Claus",
|
| 387 |
-
},
|
| 388 |
-
]
|
| 389 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 390 |
api_url = DEFAULT_API_URL
|
| 391 |
questions_url = f"{api_url}/questions"
|
| 392 |
-
|
| 393 |
-
|
| 394 |
-
|
| 395 |
-
"agent_code": "https://huggingface.co/spaces/Benjosaur/Final_Assignment_Template/tree/main",
|
| 396 |
-
"answers": answers_payload,
|
| 397 |
-
}
|
| 398 |
-
response = requests.post(submit_url, json=submission_data, timeout=60)
|
| 399 |
-
response.raise_for_status()
|
| 400 |
-
if not response.ok:
|
| 401 |
-
print("Error submitting results!")
|
| 402 |
-
print("Status code:", response.status_code)
|
| 403 |
-
print("Response text:", response.text)
|
| 404 |
response.raise_for_status()
|
| 405 |
-
|
| 406 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
import os
|
| 2 |
+
import gradio as gr
|
|
|
|
| 3 |
import requests
|
| 4 |
import inspect
|
| 5 |
import pandas as pd
|
|
|
|
| 78 |
return final_answer
|
| 79 |
|
| 80 |
|
| 81 |
+
async def run_and_submit_all(profile: gr.OAuthProfile | None):
|
|
|
|
| 82 |
"""
|
| 83 |
Fetches all questions, runs the BasicAgent on them, submits all answers,
|
| 84 |
and displays the results.
|
|
|
|
| 86 |
# --- Determine HF Space Runtime URL and Repo URL ---
|
| 87 |
space_id = os.getenv("SPACE_ID") # Get the SPACE_ID for sending link to the code
|
| 88 |
|
| 89 |
+
if profile:
|
| 90 |
+
username = f"{profile.username}"
|
| 91 |
+
print(f"User logged in: {username}")
|
| 92 |
+
else:
|
| 93 |
+
print("User not logged in.")
|
| 94 |
+
return "Please Login to Hugging Face with the button.", None
|
| 95 |
username = "benjosaur"
|
| 96 |
|
| 97 |
api_url = DEFAULT_API_URL
|
|
|
|
| 224 |
|
| 225 |
|
| 226 |
# # --- Build Gradio Interface using Blocks ---
|
| 227 |
+
with gr.Blocks() as demo:
|
| 228 |
+
gr.Markdown("# Basic Agent Evaluation Runner")
|
| 229 |
+
gr.Markdown(
|
| 230 |
+
"""
|
| 231 |
+
**Instructions:**
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 232 |
|
| 233 |
+
1. Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
|
| 234 |
+
2. Log in to your Hugging Face account using the button below. This uses your HF username for submission.
|
| 235 |
+
3. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 236 |
|
| 237 |
+
---
|
| 238 |
+
**Disclaimers:**
|
| 239 |
+
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).
|
| 240 |
+
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.
|
| 241 |
+
"""
|
| 242 |
+
)
|
| 243 |
+
|
| 244 |
+
gr.LoginButton()
|
| 245 |
+
|
| 246 |
+
run_button = gr.Button("Run Evaluation & Submit All Answers")
|
| 247 |
+
|
| 248 |
+
status_output = gr.Textbox(
|
| 249 |
+
label="Run Status / Submission Result", lines=5, interactive=False
|
| 250 |
+
)
|
| 251 |
+
# Removed max_rows=10 from DataFrame constructor
|
| 252 |
+
results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 253 |
|
| 254 |
+
run_button.click(fn=run_and_submit_all, outputs=[status_output, results_table])
|
| 255 |
+
|
| 256 |
+
|
| 257 |
+
async def main():
|
| 258 |
+
agent = BasicAgent()
|
| 259 |
api_url = DEFAULT_API_URL
|
| 260 |
questions_url = f"{api_url}/questions"
|
| 261 |
+
print(f"Fetching questions from: {questions_url}")
|
| 262 |
+
|
| 263 |
+
response = requests.get(questions_url, timeout=15)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 264 |
response.raise_for_status()
|
| 265 |
+
questions_data = response.json()
|
| 266 |
+
|
| 267 |
+
# 3. Run your Agent
|
| 268 |
+
results_log = []
|
| 269 |
+
answers_payload = []
|
| 270 |
+
print(f"Running agent on {len(questions_data)} questions...")
|
| 271 |
+
item = questions_data[0]
|
| 272 |
+
task_id = item.get("task_id")
|
| 273 |
+
question_text = item.get("question")
|
| 274 |
+
file_name = item.get("file_name")
|
| 275 |
+
submitted_answer = await agent(question_text, task_id, file_name)
|
| 276 |
+
answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
|
| 277 |
+
results_log.append(
|
| 278 |
+
{
|
| 279 |
+
"Task ID": task_id,
|
| 280 |
+
"Question": question_text,
|
| 281 |
+
"Submitted Answer": submitted_answer,
|
| 282 |
+
}
|
| 283 |
+
)
|
| 284 |
+
|
| 285 |
+
|
| 286 |
+
if __name__ == "__main__":
|
| 287 |
+
print("\n" + "-" * 30 + " App Starting " + "-" * 30)
|
| 288 |
+
# Check for SPACE_HOST and SPACE_ID at startup for information
|
| 289 |
+
space_host_startup = os.getenv("SPACE_HOST")
|
| 290 |
+
space_id_startup = os.getenv("SPACE_ID") # Get SPACE_ID at startup
|
| 291 |
+
|
| 292 |
+
if space_host_startup:
|
| 293 |
+
print(f"✅ SPACE_HOST found: {space_host_startup}")
|
| 294 |
+
print(f" Runtime URL should be: https://{space_host_startup}.hf.space")
|
| 295 |
+
else:
|
| 296 |
+
print("ℹ️ SPACE_HOST environment variable not found (running locally?).")
|
| 297 |
+
|
| 298 |
+
if space_id_startup: # Print repo URLs if SPACE_ID is found
|
| 299 |
+
print(f"✅ SPACE_ID found: {space_id_startup}")
|
| 300 |
+
print(f" Repo URL: https://huggingface.co/spaces/{space_id_startup}")
|
| 301 |
+
print(
|
| 302 |
+
f" Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main"
|
| 303 |
+
)
|
| 304 |
+
else:
|
| 305 |
+
print(
|
| 306 |
+
"ℹ️ SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined."
|
| 307 |
+
)
|
| 308 |
+
|
| 309 |
+
print("-" * (60 + len(" App Starting ")) + "\n")
|
| 310 |
+
|
| 311 |
+
print("Launching Gradio Interface for Basic Agent Evaluation...")
|
| 312 |
+
demo.launch(debug=True, share=False)
|
| 313 |
+
# asyncio.run(run_and_submit_all())
|