Update app.py

app.py

@@ -1,34 +1,207 @@
 import os
 import gradio as gr
 import requests
-import inspect
 import pandas as pd
+from smolagents import (
+    CodeAgent,
+    DuckDuckGoSearchTool,
+    WebSearchTool,
+    VisitWebpageTool,
+    InferenceClientModel,
+    tool,
+)
 
-# (Keep Constants as is)
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# --- Basic Agent Definition ---
-# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
+
+# =============================================
+# CUSTOM TOOLS
+# =============================================
+
+@tool
+def download_file_from_api(task_id: str) -> str:
+    """Downloads a file associated with a GAIA task and returns its text content.
+    Use this tool when a question mentions an attached file, spreadsheet, image,
+    audio, or any document that you need to read.
+
+    Args:
+        task_id: The task_id string for the question that has an associated file.
+    """
+    import tempfile
+    api_url = "https://agents-course-unit4-scoring.hf.space"
+    url = f"{api_url}/files/{task_id}"
+    try:
+        response = requests.get(url, timeout=30)
+        response.raise_for_status()
+
+        content_type = response.headers.get("Content-Type", "")
+
+        # If it's a text-based file, return the text directly
+        if "text" in content_type or "json" in content_type or "csv" in content_type:
+            return response.text[:10000]  # Limit to avoid context overflow
+
+        # For Excel files
+        if "spreadsheet" in content_type or "excel" in content_type:
+            import openpyxl
+            import io
+            wb = openpyxl.load_workbook(io.BytesIO(response.content))
+            result = []
+            for sheet_name in wb.sheetnames:
+                ws = wb[sheet_name]
+                result.append(f"--- Sheet: {sheet_name} ---")
+                for row in ws.iter_rows(values_only=True):
+                    result.append("\t".join([str(c) if c is not None else "" for c in row]))
+            return "\n".join(result)[:10000]
+
+        # For PDF files
+        if "pdf" in content_type:
+            try:
+                import PyPDF2
+                import io
+                reader = PyPDF2.PdfReader(io.BytesIO(response.content))
+                text = ""
+                for page in reader.pages:
+                    text += page.extract_text() or ""
+                return text[:10000] if text else "PDF found but could not extract text."
+            except ImportError:
+                return "PDF file detected but PyPDF2 not installed."
+
+        # For other binary files, save and report
+        suffix = ""
+        if "image" in content_type:
+            suffix = ".png"
+        elif "audio" in content_type:
+            suffix = ".mp3"
+
+        with tempfile.NamedTemporaryFile(delete=False, suffix=suffix) as f:
+            f.write(response.content)
+            return f"File downloaded to {f.name} (type: {content_type}). File size: {len(response.content)} bytes."
+
+    except Exception as e:
+        return f"Error downloading file for task {task_id}: {str(e)}"
+
+
+@tool
+def read_local_file(file_path: str) -> str:
+    """Reads the content of a local file and returns it as a string.
+
+    Args:
+        file_path: The path to the file to read.
+    """
+    try:
+        with open(file_path, "r", encoding="utf-8", errors="ignore") as f:
+            return f.read()[:10000]
+    except Exception as e:
+        return f"Error reading file: {str(e)}"
+
+
+# =============================================
+# AGENT CLASS
+# =============================================
+
 class BasicAgent:
+    """An agent that uses smolagents CodeAgent with web search and file tools
+    to answer GAIA benchmark questions. Uses Hugging Face Inference API (no GPU needed)."""
+
     def __init__(self):
-        print("BasicAgent initialized.")
-    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):
+        print("Initializing SmolAgent for GAIA benchmark...")
+
+        # Use HF Inference API - completely free, no GPU needed
+        model = InferenceClientModel(
+            model_id="Qwen/Qwen2.5-Coder-32B-Instruct",
+            token=os.getenv("HF_TOKEN"),
+            max_tokens=2096,
+            temperature=0.1,  # Low temperature for more precise answers
+        )
+
+        # System prompt tailored for GAIA exact-match scoring
+        system_prompt = """You are a precise AI assistant solving GAIA benchmark questions.
+
+CRITICAL RULES:
+1. Your final answer must be ONLY the answer itself - no explanations, no prefixes like "The answer is", no extra text.
+2. If the answer is a number, give just the number (e.g., "42" not "The answer is 42").
+3. If the answer is a name, give just the name (e.g., "Paris" not "The answer is Paris").
+4. If the answer is a list, use comma-separated values as requested.
+5. Be precise and factual. Use tools to verify information.
+6. If a question mentions a file or attachment, use the download_file_from_api tool with the task_id.
+7. Always search the web when you need current or specific factual information.
+8. Think step by step, but your FINAL output must be ONLY the answer.
+"""
+
+        self.agent = CodeAgent(
+            model=model,
+            tools=[
+                DuckDuckGoSearchTool(),
+                VisitWebpageTool(),
+                download_file_from_api,
+                read_local_file,
+            ],
+            max_steps=8,
+            verbosity_level=1,
+            additional_authorized_imports=[
+                "json", "re", "math", "datetime", "collections",
+                "csv", "io", "os", "tempfile",
+            ],
+        )
+
+        # Override system prompt
+        self.agent.system_prompt = system_prompt + "\n\n" + self.agent.system_prompt
+
+        print("SmolAgent initialized successfully!")
+
+    def __call__(self, question: str, task_id: str = None) -> str:
+        print(f"Agent processing question: {question[:80]}...")
+
+        # Build the prompt with task_id context if available
+        prompt = question
+        if task_id:
+            prompt = f"""Answer this question. If you need to download an attached file, use download_file_from_api with task_id="{task_id}".
+
+Question: {question}
+
+Remember: respond with ONLY the final answer, nothing else."""
+        else:
+            prompt = f"""Answer this question precisely.
+
+Question: {question}
+
+Remember: respond with ONLY the final answer, nothing else."""
+
+        try:
+            result = self.agent.run(prompt)
+            # Clean up the answer
+            answer = str(result).strip()
+            # Remove common prefixes that models add
+            for prefix in [
+                "The answer is ", "The answer is: ", "Answer: ", "FINAL ANSWER: ",
+                "Final answer: ", "The final answer is ", "The final answer is: ",
+            ]:
+                if answer.lower().startswith(prefix.lower()):
+                    answer = answer[len(prefix):].strip()
+            # Remove trailing periods if the answer is not a sentence
+            if answer.endswith(".") and len(answer.split()) <= 5:
+                answer = answer[:-1].strip()
+            print(f"Agent answer: {answer}")
+            return answer
+        except Exception as e:
+            print(f"Agent error: {e}")
+            return "Unable to determine the answer."
+
+
+# =============================================
+# SUBMISSION LOGIC 
+# =============================================
+
+def run_and_submit_all(profile: gr.OAuthProfile | None):
     """
-    Fetches all questions, runs the BasicAgent on them, submits all answers,
+    Fetches all questions, runs the agent 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
+    space_id = os.getenv("SPACE_ID")
 
     if profile:
-        username= f"{profile.username}"
+        username = f"{profile.username}"
         print(f"User logged in: {username}")
     else:
         print("User not logged in.")
@@ -38,13 +211,13 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     questions_url = f"{api_url}/questions"
     submit_url = f"{api_url}/submit"
 
-    # 1. Instantiate Agent ( modify this part to create your agent)
+    # 1. Instantiate 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)
 
@@ -55,51 +228,62 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         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("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
+    # 3. Run Agent on each question
     results_log = []
     answers_payload = []
     print(f"Running agent on {len(questions_data)} questions...")
-    for item in questions_data:
+    for i, item in enumerate(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
+        print(f"\n{'='*50}")
+        print(f"Question {i+1}/{len(questions_data)} - Task: {task_id}")
+        print(f"{'='*50}")
         try:
-            submitted_answer = agent(question_text)
+            submitted_answer = agent(question_text, task_id=task_id)
             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})
+            results_log.append({
+                "Task ID": task_id,
+                "Question": question_text,
+                "Submitted Answer": submitted_answer
+            })
         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}"})
+            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}"
+            })
 
     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)
 
-    # 4. Prepare Submission 
-    submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}
+    # 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)
 
     # 5. Submit
     print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
     try:
-        response = requests.post(submit_url, json=submission_data, timeout=60)
+        response = requests.post(submit_url, json=submission_data, timeout=120)
         response.raise_for_status()
         result_data = response.json()
         final_status = (
@@ -142,28 +326,23 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
 
 # --- Build Gradio Interface using Blocks ---
 with gr.Blocks() as demo:
-    gr.Markdown("# Basic Agent Evaluation Runner")
+    gr.Markdown("# 🤖 GAIA Agent - Final Assignment")
     gr.Markdown(
         """
-        **Instructions:**
+        **Agent**: SmolAgent (CodeAgent) with Qwen2.5-Coder-32B via HF Inference API
 
-        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.
+        **Tools**: Web Search, Webpage Visitor, File Downloader
 
-        ---
-        **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.
+        **Instructions:**
+        1. Log in to your Hugging Face account using the button below.
+        2. Click 'Run Evaluation & Submit All Answers' to start.
+        3. Wait for the agent to process all 20 questions (this may take several minutes).
         """
     )
 
     gr.LoginButton()
-
-    run_button = gr.Button("Run Evaluation & Submit All Answers")
-
+    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(
@@ -173,9 +352,8 @@ with gr.Blocks() as demo:
 
 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
+    space_id_startup = os.getenv("SPACE_ID")
 
     if space_host_startup:
         print(f"✅ SPACE_HOST found: {space_host_startup}")
@@ -183,14 +361,12 @@ if __name__ == "__main__":
     else:
         print("ℹ️  SPACE_HOST environment variable not found (running locally?).")
 
-    if space_id_startup: # Print repo URLs if SPACE_ID is found
+    if space_id_startup:
         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("ℹ️  SPACE_ID environment variable not found (running locally?).")
 
-    print("Launching Gradio Interface for Basic Agent Evaluation...")
+    print("-"*60 + "\n")
+    print("Launching Gradio Interface...")
     demo.launch(debug=True, share=False)