Update app.py

app.py

@@ -1,196 +1,226 @@
 import os
 import gradio as gr
 import requests
-import inspect
 import pandas as pd
-
-# (Keep Constants as is)
-# --- Constants ---
+import re
+import tempfile
+import pytesseract
+from PIL import Image
+from typing import Dict, List, Optional, TypedDict, Annotated
+from langgraph.graph import StateGraph, END
+from langgraph.checkpoint.memory import MemorySaver
+from langchain_core.messages import HumanMessage, SystemMessage, AnyMessage
+from langchain_openai import ChatOpenAI
+from langgraph.prebuilt import ToolNode, tools_condition
+from langchain_community.tools.tavily_search import TavilySearchResults
+from youtube_transcript_api import YouTubeTranscriptApi
+import yt_dlp
+import cv2
+import numpy as np
+import speech_recognition as sr
+
+# ================ Configuración Global ================
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
-
-# --- Basic Agent Definition ---
-# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
-class BasicAgent:
+SYSTEM_PROMPT = SYSTEM_PROMPT = """You are a precision research assistant for the GAIA benchmark. Your mission is EXTREME ACCURACY.
+CRITICAL ANSWER FORMAT RULES:
+# - ALWAYS end with: FINAL ANSWER: [answer]
+# - READ THE QUESTION CAREFULLY - answer EXACTLY what is asked for, nothing more, nothing less
+SPECIFIC FORMATTING BY QUESTION TYPE:
+# - Numbers: ONLY the number, no units, no text
+# Example: "FINAL ANSWER: 2" NOT "FINAL ANSWER: 2 albums"
+# - First name only: ONLY the first name
+# Example: If person is "John Smith" → "FINAL ANSWER: John"
+# - Country codes, IOC codes, abbreviations, symbols: ONLY the code/abbreviation, no country name or brackets
+# Example: If asked for IOC country code → "FINAL ANSWER: PHI" NOT "FINAL ANSWER: PHILIPPINES [PHI]"
+# - When asked for a specific type of identifier (code, abbreviation, symbol):
+#   Give ONLY that identifier, strip all explanatory text, brackets, or full names
+# - Lists/Sets: Exactly as requested format
+# Example: "FINAL ANSWER: a, b, d, e" (comma-separated, alphabetical order)
+CRITICAL TOOL SELECTION:
+# - Wikipedia questions → wikipedia_tool ONLY
+# - File questions → file_analyzer_tool FIRST to inspect contents, then reason based on structure
+# - Current events → web_search_tool ONLY
+# - Mathematical analysis/calculations → wolfram_alpha_tool or python_repl_tool ONLY
+# - Tables, matrices, systematic checking → python_repl_tool ONLY
+FOR MATHEMATICAL PROBLEMS:
+# ALWAYS use python_repl_tool when:
+# - Analyzing mathematical tables or matrices
+# - Checking properties like commutativity, associativity
+# - Systematic verification of mathematical statements
+# - Complex calculations that need precision
+# - ANY problem involving tables, sets, or systematic checking
+MATHEMATICAL ANALYSIS PROCESS:
+# 1. Use python_repl_tool to parse data systematically
+# 2. Write code to check ALL cases (don't rely on manual inspection)
+# 3. Collect results programmatically
+# 4. Verify your logic with multiple approaches
+# 5. Format answer exactly as requested
+# Example for commutativity checking:
+# - Parse the operation table into a data structure
+# - Check ALL pairs (x,y) to see if x*y = y*x
+# - Collect ALL elements involved in ANY counter-example
+# - Return in requested format (e.g., comma-separated, alphabetical)
+FILE HANDLING:
+# - You HAVE the ability to read and analyze uploaded files
+# - ALWAYS use file_analyzer_tool when questions mention files
+# - The tool automatically finds and analyzes Excel, CSV, images, and audio files
+# - For Excel/CSV: Returns columns, data types, sample rows, and numeric totals
+# - NEVER say "I can't access files" - you CAN access them via file_analyzer_tool
+# - Example: "The attached Excel file..." → Use file_analyzer_tool immediately
+SPECIAL CASES TO HANDLE:
+# - If the question appears reversed or encoded, decode it first.
+# - If the question includes an instruction (e.g., "write the opposite of..."), follow the instruction precisely.
+# - DO NOT repeat or paraphrase the question in your answer.
+# - NEVER answer with the full sentence unless explicitly asked to.
+# - If the decoded question asks for a word, give ONLY the word, in the required format.
+REASONING PROCESS:
+# 1. Carefully read what the question is asking for
+# 2. Identify if it needs systematic/mathematical analysis
+# 3. Use appropriate tool (python_repl_tool for math problems)
+# 4. Extract ONLY the specific part requested
+# 5. Format according to the rules above
+# 6. For file questions:
+# a. First use file_analyzer_tool to inspect column names, types, and sample data
+# b. Identify relevant columns based on the question
+# c. Reason using the data (e.g., by counting, filtering, or identifying patterns)
+# d. Only use python_repl_tool if additional computation is necessary
+# 7. If the Wikipedia tool is used but fails to provide an answer (no relevant entry or content), automatically attempt a web search using the same query or a refined version of it
+"""
+
+# ================ Clase del Agente ================
+class GaiaAgent:
     def __init__(self):
-        print("BasicAgent initialized.")
+        self.tools = self._initialize_tools()
+        self.agent_runner = self._create_agent_runner()
+        self.recognizer = sr.Recognizer()
+        
+    def _initialize_tools(self):
+        return [
+            self.wikipedia_tool,
+            self.youtube_transcript_tool,
+            self.file_analyzer_tool,
+            self.web_search_tool
+        ]
+    
+    def _create_agent_runner(self):
+        llm = ChatOpenAI(model="gpt-4-turbo", temperature=0.0)
+        model_with_tools = llm.bind_tools(self.tools)
+
+        def agent_node(state):
+            messages = state['messages']
+            if not messages or not isinstance(messages[0], SystemMessage):
+                messages = [SystemMessage(content=SYSTEM_PROMPT)] + messages
+            
+            response = model_with_tools.invoke(messages)
+            return {"messages": [response]}
+
+        tool_node = ToolNode(self.tools)
+
+        builder = StateGraph(AgentState)
+        builder.add_node("agent", agent_node)
+        builder.add_node("tools", tool_node)
+        builder.add_edge("tools", "agent")
+        builder.add_conditional_edges("agent", tools_condition, {"tools": "tools", END: END})
+        
+        return builder.compile(checkpointer=MemorySaver())
+
+    # ================ Herramientas ================
+    def wikipedia_tool(self, query: str) -> str:
+        try:
+            import wikipedia
+            wikipedia.set_lang("en")
+            return wikipedia.summary(query, sentences=3)
+        except Exception as e:
+            return f"Wikipedia error: {str(e)}"
+
+    def youtube_transcript_tool(self, url: str, question: str) -> str:
+        try:
+            video_id = re.findall(r'(?:v=|\/)([0-9A-Za-z_-]{11})', url)[0]
+            transcript = YouTubeTranscriptApi.get_transcript(video_id)
+            return " ".join([entry['text'] for entry in transcript[:5]])
+        except Exception as e:
+            return f"Transcript error: {str(e)}"
+
+    def file_analyzer_tool(self, file_description: str = "") -> str:
+        try:
+            img = Image.open("temp_file.png")
+            text = pytesseract.image_to_string(img)
+            return f"OCR Text: {text[:500]}..." if text else "No text found"
+        except:
+            return "File analysis not available"
+
+    def web_search_tool(self, query: str) -> str:
+        try:
+            tavily = TavilySearchResults(max_results=3)
+            results = tavily.invoke(query)
+            return "\n".join([f"{res['title']}: {res['content']}" for res in results])
+        except Exception as e:
+            return f"Search error: {str(e)}"
+
+    # ================ Procesamiento Principal ================
     def __call__(self, question: str) -> str:
-        print(f"Agent received question (first 50 chars): {question[:50]}...")
-        fixed_answer = "This is a default answer."
-        print(f"Agent returning fixed answer: {fixed_answer}")
-        return fixed_answer
-
-def run_and_submit_all( profile: gr.OAuthProfile | None):
-    """
-    Fetches all questions, runs the BasicAgent on them, submits all answers,
-    and displays the results.
-    """
-    # --- Determine HF Space Runtime URL and Repo URL ---
-    space_id = os.getenv("SPACE_ID") # Get the SPACE_ID for sending link to the code
-
-    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"
-
-    # 1. Instantiate Agent ( modify this part to create your agent)
-    try:
-        agent = BasicAgent()
-    except Exception as e:
-        print(f"Error instantiating agent: {e}")
-        return f"Error initializing agent: {e}", None
-    # 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)
-    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
-    print(agent_code)
-
-    # 2. Fetch Questions
-    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 requests.exceptions.RequestException as e:
-        print(f"Error fetching questions: {e}")
-        return f"Error fetching questions: {e}", None
-    except requests.exceptions.JSONDecodeError as e:
-         print(f"Error decoding JSON response from questions endpoint: {e}")
-         print(f"Response text: {response.text[:500]}")
-         return f"Error decoding server response for questions: {e}", None
-    except Exception as e:
-        print(f"An unexpected error occurred fetching questions: {e}")
-        return f"An unexpected error occurred fetching questions: {e}", None
-
-    # 3. Run your Agent
-    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:
-            print(f"Skipping item with missing task_id or question: {item}")
-            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})
+            events = self.agent_runner.stream(
+                {"messages": [HumanMessage(content=question)]},
+                config={"configurable": {"thread_id": "gaia_thread"}},
+                stream_mode="values"
+            )
+            
+            for event in events:
+                if event['messages']:
+                    last_msg = event['messages'][-1]
+                    if hasattr(last_msg, 'content'):
+                        return self._extract_final_answer(last_msg.content)
+            return "No answer generated"
         except Exception as e:
-             print(f"Error running agent on task {task_id}: {e}")
-             results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
+            return f"Agent Error: {str(e)}"
 
-    if not answers_payload:
-        print("Agent did not produce any answers to submit.")
-        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
+    def _extract_final_answer(self, text: str) -> str:
+        match = re.search(r"FINAL ANSWER:\s*(.*)", text, re.IGNORECASE)
+        return match.group(1).strip() if match else text.split("\n")[-1].strip()
 
-    # 4. Prepare Submission 
-    submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}
-    status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..."
-    print(status_update)
+# ================ Integración con Gradio ================
+def run_and_submit_all(profile: gr.OAuthProfile | None):
+    space_id = os.getenv("SPACE_ID")
+    if not profile:
+        return "Please login with Hugging Face account", None
 
-    # 5. Submit
-    print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
     try:
-        response = requests.post(submit_url, json=submission_data, timeout=60)
+        agent = GaiaAgent()
+        questions = requests.get(f"{DEFAULT_API_URL}/questions").json()
+        
+        answers = []
+        results_log = []
+        for item in questions:
+            answer = agent(item['question'])
+            answers.append({"task_id": item['task_id'], "submitted_answer": answer})
+            results_log.append({"Task": item['task_id'], "Answer": answer})
+
+        submission_data = {
+            "username": profile.username,
+            "agent_code": f"https://huggingface.co/spaces/{space_id}",
+            "answers": answers
+        }
+        
+        response = requests.post(f"{DEFAULT_API_URL}/submit", json=submission_data)
         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.')}"
-        )
-        print("Submission successful.")
-        results_df = pd.DataFrame(results_log)
-        return final_status, results_df
-    except requests.exceptions.HTTPError as e:
-        error_detail = f"Server responded with status {e.response.status_code}."
-        try:
-            error_json = e.response.json()
-            error_detail += f" Detail: {error_json.get('detail', e.response.text)}"
-        except requests.exceptions.JSONDecodeError:
-            error_detail += f" Response: {e.response.text[:500]}"
-        status_message = f"Submission Failed: {error_detail}"
-        print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
-    except requests.exceptions.Timeout:
-        status_message = "Submission Failed: The request timed out."
-        print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
-    except requests.exceptions.RequestException as e:
-        status_message = f"Submission Failed: Network error - {e}"
-        print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
+        
+        return f"Success! Score: {response.json().get('score', 0)}", pd.DataFrame(results_log)
     except Exception as e:
-        status_message = f"An unexpected error occurred during submission: {e}"
-        print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
+        return f"Error: {str(e)}", pd.DataFrame()
 
-
-# --- Build Gradio Interface using Blocks ---
+# ================ Interfaz de Usuario ================
 with gr.Blocks() as demo:
-    gr.Markdown("# Basic Agent Evaluation Runner")
-    gr.Markdown(
-        """
-        **Instructions:**
-
-        1.  Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
-        2.  Log in to your Hugging Face account using the button below. This uses your HF username for submission.
-        3.  Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
-
-        ---
-        **Disclaimers:**
-        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).
-        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.
-        """
-    )
-
+    gr.Markdown("# GAIA Agent - Hugging Face")
     gr.LoginButton()
-
-    run_button = gr.Button("Run Evaluation & Submit All Answers")
-
-    status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
-    # Removed max_rows=10 from DataFrame constructor
-    results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
-
-    run_button.click(
+    run_btn = gr.Button("Run Evaluation")
+    status = gr.Textbox(label="Status")
+    results = gr.DataFrame(label="Results")
+    
+    run_btn.click(
         fn=run_and_submit_all,
-        outputs=[status_output, results_table]
+        outputs=[status, results]
     )
 
 if __name__ == "__main__":
-    print("\n" + "-"*30 + " App Starting " + "-"*30)
-    # Check for SPACE_HOST and SPACE_ID at startup for information
-    space_host_startup = os.getenv("SPACE_HOST")
-    space_id_startup = os.getenv("SPACE_ID") # Get SPACE_ID at startup
-
-    if space_host_startup:
-        print(f"✅ SPACE_HOST found: {space_host_startup}")
-        print(f"   Runtime URL should be: https://{space_host_startup}.hf.space")
-    else:
-        print("ℹ️  SPACE_HOST environment variable not found (running locally?).")
-
-    if space_id_startup: # Print repo URLs if SPACE_ID is found
-        print(f"✅ SPACE_ID found: {space_id_startup}")
-        print(f"   Repo URL: https://huggingface.co/spaces/{space_id_startup}")
-        print(f"   Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main")
-    else:
-        print("ℹ️  SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined.")
-
-    print("-"*(60 + len(" App Starting ")) + "\n")
-
-    print("Launching Gradio Interface for Basic Agent Evaluation...")
-    demo.launch(debug=True, share=False)
+    demo.launch(debug=True)