implemented test on random question without submit
Browse files
app.py
CHANGED
|
@@ -19,6 +19,71 @@ class BasicAgent:
|
|
| 19 |
answer = self.agent.run(question)
|
| 20 |
print(f"Agent returning answer: {answer}")
|
| 21 |
return answer
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
|
| 23 |
def run_and_submit_all( profile: gr.OAuthProfile | None):
|
| 24 |
"""
|
|
@@ -163,6 +228,8 @@ with gr.Blocks() as demo:
|
|
| 163 |
|
| 164 |
run_button = gr.Button("Run Evaluation & Submit All Answers")
|
| 165 |
|
|
|
|
|
|
|
| 166 |
status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
|
| 167 |
# Removed max_rows=10 from DataFrame constructor
|
| 168 |
results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
|
|
@@ -172,6 +239,11 @@ with gr.Blocks() as demo:
|
|
| 172 |
outputs=[status_output, results_table]
|
| 173 |
)
|
| 174 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 175 |
if __name__ == "__main__":
|
| 176 |
print("\n" + "-"*30 + " App Starting " + "-"*30)
|
| 177 |
# Check for SPACE_HOST and SPACE_ID at startup for information
|
|
|
|
| 19 |
answer = self.agent.run(question)
|
| 20 |
print(f"Agent returning answer: {answer}")
|
| 21 |
return answer
|
| 22 |
+
|
| 23 |
+
def run_random_question(profile: gr.OAuthProfile | None):
|
| 24 |
+
space_id = os.getenv("SPACE_ID")
|
| 25 |
+
|
| 26 |
+
if profile:
|
| 27 |
+
username= f"{profile.username}"
|
| 28 |
+
print(f"User logged in: {username}")
|
| 29 |
+
else:
|
| 30 |
+
print("User not logged in.")
|
| 31 |
+
return "Please Login to Hugging Face with the button.", None
|
| 32 |
+
|
| 33 |
+
api_url = DEFAULT_API_URL
|
| 34 |
+
random_question_url = f"{api_url}//random-question"
|
| 35 |
+
|
| 36 |
+
try:
|
| 37 |
+
agent = BasicAgent()
|
| 38 |
+
except Exception as e:
|
| 39 |
+
print(f"Error instantiating agent: {e}")
|
| 40 |
+
return f"Error initializing agent: {e}", None
|
| 41 |
+
# 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)
|
| 42 |
+
agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
|
| 43 |
+
print(agent_code)
|
| 44 |
+
|
| 45 |
+
print(f"Fetching questions from: {random_question_url}")
|
| 46 |
+
try:
|
| 47 |
+
response = requests.get(random_question_url, timeout=15)
|
| 48 |
+
response.raise_for_status()
|
| 49 |
+
questions_data = response.json()
|
| 50 |
+
if not questions_data:
|
| 51 |
+
print("Fetched questions list is empty.")
|
| 52 |
+
return "Fetched questions list is empty or invalid format.", None
|
| 53 |
+
print(f"Fetched {len(questions_data)} questions.")
|
| 54 |
+
except requests.exceptions.RequestException as e:
|
| 55 |
+
print(f"Error fetching questions: {e}")
|
| 56 |
+
return f"Error fetching questions: {e}", None
|
| 57 |
+
except requests.exceptions.JSONDecodeError as e:
|
| 58 |
+
print(f"Error decoding JSON response from questions endpoint: {e}")
|
| 59 |
+
print(f"Response text: {response.text[:500]}")
|
| 60 |
+
return f"Error decoding server response for questions: {e}", None
|
| 61 |
+
except Exception as e:
|
| 62 |
+
print(f"An unexpected error occurred fetching questions: {e}")
|
| 63 |
+
return f"An unexpected error occurred fetching questions: {e}", None
|
| 64 |
+
|
| 65 |
+
results_log = []
|
| 66 |
+
answers_payload = []
|
| 67 |
+
print(f"Running agent on {len(questions_data)} questions...")
|
| 68 |
+
for item in questions_data:
|
| 69 |
+
task_id = item.get("task_id")
|
| 70 |
+
question_text = item.get("question")
|
| 71 |
+
if not task_id or question_text is None:
|
| 72 |
+
print(f"Skipping item with missing task_id or question: {item}")
|
| 73 |
+
continue
|
| 74 |
+
try:
|
| 75 |
+
submitted_answer = agent(question_text)
|
| 76 |
+
answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
|
| 77 |
+
results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
|
| 78 |
+
except Exception as e:
|
| 79 |
+
print(f"Error running agent on task {task_id}: {e}")
|
| 80 |
+
results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
|
| 81 |
+
|
| 82 |
+
if not answers_payload:
|
| 83 |
+
print("Agent did not produce any answers to submit.")
|
| 84 |
+
return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
|
| 85 |
+
|
| 86 |
+
return "Test done successfully", pd.DataFrame(results_log)
|
| 87 |
|
| 88 |
def run_and_submit_all( profile: gr.OAuthProfile | None):
|
| 89 |
"""
|
|
|
|
| 228 |
|
| 229 |
run_button = gr.Button("Run Evaluation & Submit All Answers")
|
| 230 |
|
| 231 |
+
test_button = gr.Button("Run Test on a random question")
|
| 232 |
+
|
| 233 |
status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
|
| 234 |
# Removed max_rows=10 from DataFrame constructor
|
| 235 |
results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
|
|
|
|
| 239 |
outputs=[status_output, results_table]
|
| 240 |
)
|
| 241 |
|
| 242 |
+
test_button.click(
|
| 243 |
+
fn=run_random_question,
|
| 244 |
+
outputs=[status_output, results_table]
|
| 245 |
+
)
|
| 246 |
+
|
| 247 |
if __name__ == "__main__":
|
| 248 |
print("\n" + "-"*30 + " App Starting " + "-"*30)
|
| 249 |
# Check for SPACE_HOST and SPACE_ID at startup for information
|