Update app.py

app.py

@@ -5,197 +5,689 @@ import inspect
 import pandas as pd
 import spaces
 
+from smolagents import (
+    CodeAgent,
+    InferenceClientModel,
+    DuckDuckGoSearchTool,
+    VisitWebpageTool,
+)
+
+
+# ---------------------------------------------------------
+# ZeroGPU startup requirement
+# ---------------------------------------------------------
+
 @spaces.GPU(duration=1)
 def _zerogpu_probe():
     return True
 
-# (Keep Constants as is)
-# --- Constants ---
+
+# ---------------------------------------------------------
+# Constants
+# ---------------------------------------------------------
+
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# --- Basic Agent Definition ---
-# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
+
+# ---------------------------------------------------------
+# GAIA Agent
+# ---------------------------------------------------------
+
 class BasicAgent:
     def __init__(self):
-        print("BasicAgent initialized.")
+        print("Initializing GAIA agent...")
+
+        # Read the Hugging Face token securely from Space Secrets.
+        token = os.getenv("HF_TOKEN")
+
+        if not token:
+            raise RuntimeError(
+                "HF_TOKEN secret is not configured. "
+                "Add HF_TOKEN under Space Settings -> Variables and secrets."
+            )
+
+        # Small reasoning model available through HF Inference Providers.
+        self.model = InferenceClientModel(
+            model_id="Qwen/Qwen3-4B-Thinking-2507",
+            provider="auto",
+            token=token,
+            temperature=0.0,
+            max_tokens=2048,
+            timeout=120,
+        )
+
+        # CodeAgent can reason, execute Python, search the web,
+        # and inspect webpages.
+        self.agent = CodeAgent(
+            tools=[
+                DuckDuckGoSearchTool(
+                    max_results=8,
+                ),
+                VisitWebpageTool(
+                    max_output_length=30000,
+                ),
+            ],
+            model=self.model,
+            max_steps=6,
+            additional_authorized_imports=[
+                "math",
+                "re",
+                "json",
+                "datetime",
+                "statistics",
+                "collections",
+                "itertools",
+            ],
+            instructions="""
+You are solving GAIA benchmark questions.
+
+Your job is to determine the exact answer to the user's question.
+
+Work carefully and verify factual information.
+
+Use web search when the question requires current, historical,
+geographical, biographical, scientific, cultural, or other factual
+information that should be verified.
+
+Use the webpage-reading tool when search results point to useful sources.
+
+Use Python when calculations, counting, date arithmetic, conversions,
+logical operations, or other computational reasoning would improve
+accuracy.
+
+You may perform multiple searches or calculations before answering.
+
+IMPORTANT:
+The evaluation system uses exact-match scoring.
+
+Your final response MUST contain only the requested answer.
+
+Do not explain your reasoning.
+Do not provide citations.
+Do not add introductory text.
+Do not add concluding text.
+Do not write "FINAL ANSWER".
+Do not write "Final answer".
+Do not write "The answer is".
+Do not surround the answer with Markdown formatting.
+Do not use a code block unless the question explicitly asks for code.
+
+Follow the exact format requested by the question.
+
+Preserve appropriate:
+- capitalization
+- punctuation
+- spelling
+- names
+- dates
+- units
+- decimal places
+- comma-separated lists
+- ordering
+
+If the question requests multiple items, return only those items in the
+requested format.
+
+Before producing the final response, verify that it directly answers
+the question and contains nothing unnecessary.
+""",
+        )
+
+        print("GAIA agent initialized successfully.")
+
     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
+        print("\n" + "=" * 80)
+        print("Agent received question:")
+        print(question)
+        print("=" * 80)
+
+        try:
+            result = self.agent.run(question)
+
+            answer = str(result).strip()
+
+            # -------------------------------------------------
+            # Defensive cleanup for exact-match scoring
+            # -------------------------------------------------
+
+            prefixes = [
+                "FINAL ANSWER:",
+                "FINAL ANSWER :",
+                "Final Answer:",
+                "Final Answer :",
+                "Final answer:",
+                "Final answer :",
+                "Answer:",
+                "Answer :",
+                "The answer is ",
+                "The answer is: ",
+            ]
 
-def run_and_submit_all( profile: gr.OAuthProfile | None):
+            for prefix in prefixes:
+                if answer.startswith(prefix):
+                    answer = answer[len(prefix):].strip()
+                    break
+
+            # Remove accidental Markdown code fences/backticks.
+            if answer.startswith("```") and answer.endswith("```"):
+                answer = answer[3:-3].strip()
+
+                # Remove a language identifier if one appears.
+                first_line, separator, remainder = answer.partition("\n")
+                if separator and first_line.strip().lower() in {
+                    "text",
+                    "plaintext",
+                    "python",
+                }:
+                    answer = remainder.strip()
+
+            answer = answer.strip("`").strip()
+
+            print(f"Final submitted answer: {answer}")
+            return answer
+
+        except Exception as e:
+            print(f"Agent execution error: {type(e).__name__}: {e}")
+            raise
+
+
+# ---------------------------------------------------------
+# Evaluation and Submission
+# ---------------------------------------------------------
+
+def run_and_submit_all(profile: gr.OAuthProfile | None):
     """
-    Fetches all questions, runs the BasicAgent on them, submits all answers,
-    and displays the results.
+    Fetches all evaluation questions, runs the agent on each question,
+    submits all generated 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
+
+    # -----------------------------------------------------
+    # Determine Hugging Face username and Space URL
+    # -----------------------------------------------------
+
+    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.")
-        return "Please Login to Hugging Face with the button.", None
+        return (
+            "Please Login to Hugging Face with the button.",
+            None,
+        )
+
+    if not space_id:
+        print("SPACE_ID environment variable is missing.")
+        return (
+            "SPACE_ID environment variable is missing.",
+            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)
+    # -----------------------------------------------------
+    # 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)
 
-    # 2. Fetch Questions
+        return (
+            f"Error initializing agent: {e}",
+            None,
+        )
+
+    # Public link to the submitted agent code.
+    agent_code = (
+        f"https://huggingface.co/spaces/"
+        f"{space_id}/tree/main"
+    )
+
+    print(f"Agent code URL: {agent_code}")
+
+    # -----------------------------------------------------
+    # 2. Fetch questions
+    # -----------------------------------------------------
+
     print(f"Fetching questions from: {questions_url}")
+
     try:
-        response = requests.get(questions_url, timeout=15)
+        response = requests.get(
+            questions_url,
+            timeout=30,
+        )
+
         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.")
+            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
+
+        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
+        print(
+            f"Error decoding JSON response "
+            f"from questions endpoint: {e}"
+        )
+
+        print(
+            f"Response text: {response.text[:500]}"
+        )
+
+        return (
+            f"Error decoding server response "
+            f"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
+        print(
+            f"Unexpected error fetching questions: {e}"
+        )
+
+        return (
+            f"An unexpected error occurred "
+            f"fetching questions: {e}",
+            None,
+        )
+
+    # -----------------------------------------------------
+    # 3. Run agent on every question
+    # -----------------------------------------------------
 
-    # 3. Run your Agent
     results_log = []
     answers_payload = []
-    print(f"Running agent on {len(questions_data)} questions...")
-    for item in questions_data:
+
+    print(
+        f"Running agent on "
+        f"{len(questions_data)} questions..."
+    )
+
+    for index, item in enumerate(
+        questions_data,
+        start=1,
+    ):
         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}")
+            print(
+                "Skipping item with missing "
+                f"task_id or question: {item}"
+            )
             continue
+
+        print("\n" + "#" * 80)
+        print(
+            f"QUESTION {index}/"
+            f"{len(questions_data)}"
+        )
+        print(f"Task ID: {task_id}")
+        print("#" * 80)
+
         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})
+            submitted_answer = agent(
+                question_text
+            )
+
+            submitted_answer = str(
+                submitted_answer
+            ).strip()
+
+            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:
-             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 "
+                f"task {task_id}: {e}"
+            )
+
+            results_log.append(
+                {
+                    "Task ID": task_id,
+                    "Question": question_text,
+                    "Submitted Answer":
+                        f"AGENT ERROR: {e}",
+                }
+            )
+
+    # -----------------------------------------------------
+    # Ensure at least one valid answer exists
+    # -----------------------------------------------------
 
     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)
+        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,
+    }
+
+    status_update = (
+        f"Agent finished. "
+        f"Submitting {len(answers_payload)} "
+        f"answers for user '{username}'..."
+    )
 
-    # 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}")
+    # -----------------------------------------------------
+    # 5. Submit answers
+    # -----------------------------------------------------
+
+    print(
+        f"Submitting "
+        f"{len(answers_payload)} answers "
+        f"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 = (
-            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.')}"
+            "Submission Successful!\n"
+            f"User: "
+            f"{result_data.get('username')}\n"
+            f"Overall Score: "
+            f"{result_data.get('score', 'N/A')}% "
+            f"("
+            f"{result_data.get('correct_count', '?')}/"
+            f"{result_data.get('total_attempted', '?')} "
+            f"correct)\n"
+            f"Message: "
+            f"{result_data.get('message', 'No message received.')}"
         )
+
         print("Submission successful.")
-        results_df = pd.DataFrame(results_log)
-        return final_status, results_df
+        print(final_status)
+
+        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}."
+        error_detail = (
+            f"Server responded with status "
+            f"{e.response.status_code}."
+        )
+
         try:
             error_json = e.response.json()
-            error_detail += f" Detail: {error_json.get('detail', e.response.text)}"
+
+            error_detail += (
+                " Detail: "
+                f"{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}"
+            error_detail += (
+                f" Response: "
+                f"{e.response.text[:500]}"
+            )
+
+        status_message = (
+            f"Submission Failed: "
+            f"{error_detail}"
+        )
+
         print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
+
+        results_df = pd.DataFrame(
+            results_log
+        )
+
+        return (
+            status_message,
+            results_df,
+        )
+
     except requests.exceptions.Timeout:
-        status_message = "Submission Failed: The request timed out."
+        status_message = (
+            "Submission Failed: "
+            "The request timed out."
+        )
+
         print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
+
+        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}"
+        status_message = (
+            "Submission Failed: "
+            f"Network error - {e}"
+        )
+
         print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
+
+        results_df = pd.DataFrame(
+            results_log
+        )
+
+        return (
+            status_message,
+            results_df,
+        )
+
     except Exception as e:
-        status_message = f"An unexpected error occurred during submission: {e}"
+        status_message = (
+            "An unexpected error occurred "
+            f"during submission: {e}"
+        )
+
         print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
 
+        results_df = pd.DataFrame(
+            results_log
+        )
+
+        return (
+            status_message,
+            results_df,
+        )
+
+
+# ---------------------------------------------------------
+# Gradio Interface
+# ---------------------------------------------------------
 
-# --- Build Gradio Interface using Blocks ---
 with gr.Blocks() as demo:
-    gr.Markdown("# Basic Agent Evaluation Runner")
+    gr.Markdown(
+        "# GAIA 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.
+        1. Log in to your Hugging Face account using the button below.
+        2. Click **Run Evaluation & Submit All Answers**.
+        3. The agent will fetch the evaluation questions, research and
+           answer them, and submit the results to the course API.
 
         ---
-        **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.
+
+        **Important:**
+
+        The evaluation uses exact-match scoring, so the agent is
+        configured to return concise final answers without explanations.
         """
     )
 
     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)
+    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]
+        outputs=[
+            status_output,
+            results_table,
+        ],
     )
 
+
+# ---------------------------------------------------------
+# Startup
+# ---------------------------------------------------------
+
 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
+    print(
+        "\n"
+        + "-" * 30
+        + " App Starting "
+        + "-" * 30
+    )
+
+    # Check Space runtime information.
+    space_host_startup = os.getenv(
+        "SPACE_HOST"
+    )
+
+    space_id_startup = os.getenv(
+        "SPACE_ID"
+    )
 
     if space_host_startup:
-        print(f"✅ SPACE_HOST found: {space_host_startup}")
-        print(f"   Runtime URL should be: https://{space_host_startup}.hf.space")
+        print(
+            "SPACE_HOST found: "
+            f"{space_host_startup}"
+        )
+
+        print(
+            "Runtime URL should be: "
+            f"https://{space_host_startup}.hf.space"
+        )
+
     else:
-        print("ℹ️  SPACE_HOST environment variable not found (running locally?).")
+        print(
+            "SPACE_HOST environment variable "
+            "not found (running locally?)."
+        )
+
+    if space_id_startup:
+        print(
+            "SPACE_ID found: "
+            f"{space_id_startup}"
+        )
+
+        print(
+            "Repo URL: "
+            "https://huggingface.co/spaces/"
+            f"{space_id_startup}"
+        )
+
+        print(
+            "Repo Tree URL: "
+            "https://huggingface.co/spaces/"
+            f"{space_id_startup}/tree/main"
+        )
 
-    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(
+            "SPACE_ID environment variable "
+            "not found (running locally?). "
+            "Repo URL cannot be determined."
+        )
 
-    print("-"*(60 + len(" App Starting ")) + "\n")
+    print(
+        "-" * (
+            60
+            + len(" App Starting ")
+        )
+        + "\n"
+    )
+
+    print(
+        "Launching Gradio Interface "
+        "for GAIA Agent Evaluation..."
+    )
 
-    print("Launching Gradio Interface for Basic Agent Evaluation...")
-    demo.launch(debug=True, share=False)
+    demo.launch(
+        debug=True,
+        share=False,
+    )