mmmm

app.py

@@ -1,96 +1,153 @@
+import os
+import gradio as gr
 import requests
-import json
-
-# Configuration
-HF_USERNAME = "AdanAtif"  # Replace with your Hugging Face username
-AGENT_CODE_URL = f"https://huggingface.co/spaces/{HF_USERNAME}/YOUR_SPACE_NAME/tree/main"
-API_BASE_URL = "https://agents-course-unit4-scoring.hf.space"
-
-def fetch_questions():
-    """Retrieves the 20 GAIA level 1 validation questions from the API."""
-    print("Fetching questions from the API...")
-    response = requests.get(f"{API_BASE_URL}/questions")
-    if response.status_code == 200:
-        return response.json()
-    else:
-        raise Exception(f"Failed to fetch questions: {response.text}")
-
-def download_task_file(task_id):
-    """Downloads associated files if the task requires it."""
-    # The API documentation specifies: GET /files/{task_id}
-    # Use this if your agent needs to read PDFs, images, or CSVs locally
-    url = f"{API_BASE_URL}/files/{task_id}"
-    response = requests.get(url)
-    if response.status_code == 200:
-        file_path = f"./{task_id}_file" # You'll want to parse the header for exact extension
-        with open(file_path, "wb") as f:
-            f.write(response.content)
-        return file_path
-    return None
-
-def dummy_agent_fallback(question_text):
-    """
-    Placeholder for your real agent call.
-    REPLACE THIS with your actual LLM/Agent call!
-    """
-    # Strict prompt constraint to force exact matches
-    system_prompt = (
-        "You are a precise QA bot. Answer the question using your tools. "
-        "Output ONLY the final string or number representing the answer. "
-        "Do not include 'The answer is...', do not include 'FINAL ANSWER', "
-        "and do not include any conversational filler."
-    )
-    # Your agent logic goes here (e.g., agent.run(question_text))
-    return "example_answer" 
-
-def submit_answers(submission_payload):
-    """Submits the payload to the final leaderboard API."""
-    print("Submitting answers to the leaderboard...")
-    headers = {"Content-Type": "application/json"}
-    response = requests.post(f"{API_BASE_URL}/submit", json=submission_payload, headers=headers)
-    
-    if response.status_code == 200:
-        print("🎉 Submission Successful!")
-        print("API Response:", response.json())
-    else:
-        print(f"❌ Submission Failed ({response.status_code}):", response.text)
+import inspect
+import pandas as pd
+# Import smolagents components
+from smolagents import CodeAgent, HfApiClient, DuckDuckGoSearchTool, HfDropInEngine
 
-def main():
-    questions = fetch_questions()
-    submitted_answers = []
-    
-    # Loop through all 20 questions
-    for item in questions:
-        task_id = item.get("task_id")
-        question = item.get("question")
-        has_file = item.get("has_file", False)
+# --- Constants ---
+DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
+
+# --- Smart Agent Definition ---
+# This overrides the original BasicAgent with a functioning tool-using agent loop
+class BasicAgent:
+    def __init__(self):
+        print("Initializing smart CodeAgent...")
+        # 1. Setup the LLM engine (Llama-3.3-70B is great for reasoning tasks)
+        # Note: Ensure HUGGINGFACE_TOKEN is set in your Space Secrets
+        self.engine = HfDropInEngine(model_id="meta-llama/Llama-3.3-70B-Instruct")
         
-        print(f"\nProcessing Task ID: {task_id}")
+        # 2. Build the agent and equip it with a web search tool
+        self.agent = CodeAgent(
+            tools=[DuckDuckGoSearchTool()], 
+            engine=self.engine,
+            max_steps=5  # Gives the agent steps to search, think, and fix errors
+        )
+        print("Smart Agent initialized successfully.")
+
+    def __call__(self, question: str) -> str:
+        print(f"\n[Agent Processing] Received question: {question[:100]}...")
         
-        # If the task has an associated file, download it first
-        if has_file:
-            file_path = download_task_file(task_id)
-            print(f"Downloaded file for task to: {file_path}")
-            
-        # Run your agent framework here
-        answer = dummy_agent_fallback(question)
-        print(f"Agent Output: {answer}")
+        # 3. Shape the prompt to force the exact match required by the leaderboard
+        strict_prompt = (
+            f"You are a precise, truth-seeking QA bot. Answer the following question using your tools:\n"
+            f"\"{question}\"\n\n"
+            "CRITICAL INSTRUCTION: Output ONLY the final answer value (e.g., a number, a specific date, or a name). "
+            "Do not write conversational filler, do not explain your steps in the final output, and DO NOT include "
+            "phrases like 'The answer is:' or 'FINAL ANSWER'."
+        )
         
-        # Format for API specification
-        submitted_answers.append({
-            "task_id": task_id,
-            "submitted_answer": str(answer).strip()
-        })
+        try:
+            # Run the agent framework
+            raw_result = self.agent.run(strict_prompt)
+            final_answer = str(raw_result).strip()
+            print(f"[Agent Success] Output: {final_answer}")
+            return final_answer
+        except Exception as e:
+            print(f"[Agent Error] Failed to process task: {e}")
+            return "Error calculating answer"
+
+# =====================================================================
+# THE REST OF YOUR GRADIO CODE REMAINS EXACTLY THE SAME BELOW
+# =====================================================================
+
+def run_and_submit_all(profile: gr.OAuthProfile | None):
+    """
+    Fetches all questions, runs the upgraded BasicAgent on them, submits all answers,
+    and displays the results.
+    """
+    space_id = os.getenv("SPACE_ID") 
+
+    if profile:
+        username = f"{profile.username}"
+        print(f"User logged in: {username}")
+    else:
+        print("User not logged in.")
+        return "Please Login to Hugging Face with the button.", None
+
+    api_url = DEFAULT_API_URL
+    questions_url = f"{api_url}/questions"
+    submit_url = f"{api_url}/submit"
+
+    try:
+        agent = BasicAgent()
+    except Exception as e:
+        print(f"Error instantiating agent: {e}")
+        return f"Error initializing agent: {e}", None
         
-    # Build final payload
-    payload = {
-        "username": HF_USERNAME,
-        "agent_code": AGENT_CODE_URL,
-        "answers": submitted_answers
-    }
+    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
+    print(agent_code)
+
+    print(f"Fetching questions from: {questions_url}")
+    try:
+        response = requests.get(questions_url, timeout=15)
+        response.raise_for_status()
+        questions_data = response.json()
+        if not questions_data:
+             print("Fetched questions list is empty.")
+             return "Fetched questions list is empty or invalid format.", None
+        print(f"Fetched {len(questions_data)} questions.")
+    except Exception as e:
+        print(f"An unexpected error occurred fetching questions: {e}")
+        return f"An unexpected error occurred fetching questions: {e}", None
+
+    results_log = []
+    answers_payload = []
+    print(f"Running agent on {len(questions_data)} questions...")
+    for item in questions_data:
+        task_id = item.get("task_id")
+        question_text = item.get("question")
+        if not task_id or question_text is None:
+            continue
+        try:
+            submitted_answer = agent(question_text)
+            answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
+        except Exception as e:
+             results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
+
+    if not answers_payload:
+        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
+
+    submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}
+    print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
     
-    # Post results
-    submit_answers(payload)
+    try:
+        response = requests.post(submit_url, json=submission_data, timeout=60)
+        response.raise_for_status()
+        result_data = response.json()
+        final_status = (
+            f"Submission Successful!\n"
+            f"User: {result_data.get('username')}\n"
+            f"Overall Score: {result_data.get('score', 'N/A')}% "
+            f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n"
+            f"Message: {result_data.get('message', 'No message received.')}"
+        )
+        return final_status, pd.DataFrame(results_log)
+    except Exception as e:
+        status_message = f"Submission Failed: {e}"
+        return status_message, pd.DataFrame(results_log)
+
+with gr.Blocks() as demo:
+    gr.Markdown("# Smart Agent Evaluation Runner")
+    gr.Markdown(
+        """
+        **Instructions:**
+        1. Log in to your Hugging Face account using the button below.
+        2. Click 'Run Evaluation & Submit All Answers' to process the GAIA benchmark.
+        """
+    )
+
+    gr.LoginButton()
+    run_button = gr.Button("Run Evaluation & Submit All Answers")
+    status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
+    results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
+
+    run_button.click(
+        fn=run_and_submit_all,
+        outputs=[status_output, results_table]
+    )
 
 if __name__ == "__main__":
-    main()
+    demo.launch(debug=True, share=False)