Update app.py

app.py

@@ -1,73 +1,123 @@
 import os
+import gradio as gr
 import requests
-from smolagents import CodeAgent, tool, OpenAIServerModel
-
-# ------------------------
-# Constants
-# ------------------------
-API_URL = "https://agents-course-unit4-scoring.hf.space"
-
-# ------------------------
-# Tool definitions
-# ------------------------
-@tool
-def fetch_questions() -> list:
-    """
-    Fetch the full list of GAIA evaluation questions.
-    """
-    response = requests.get(f"{API_URL}/questions", timeout=15)
-    response.raise_for_status()
-    return response.json()
+import pandas as pd
 
-@tool
-def fetch_random_question() -> dict:
-    """
-    Fetch a single random GAIA question.
-    """
-    response = requests.get(f"{API_URL}/random-question", timeout=15)
-    response.raise_for_status()
-    return response.json()
+from agent import create_agent, fetch_random_question
 
-@tool
-def fetch_file(task_id: str) -> bytes:
-    """
-    Download a file associated with a given task_id.
-    """
-    response = requests.get(f"{API_URL}/files/{task_id}", timeout=15)
-    response.raise_for_status()
-    return response.content
+# --- Constants ---
+DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-@tool
-def submit_answers(username: str, agent_code: str, answers: list) -> dict:
+
+def run_and_submit_all(profile: gr.OAuthProfile | None):
     """
-    Submit the agent's answers to GAIA and return the scoring.
+    Fetch all questions, run the SmolAgent on them, submit all answers,
+    and display the results.
     """
-    payload = {
-        "username": username,
-        "agent_code": agent_code,
-        "answers": answers
-    }
-    response = requests.post(f"{API_URL}/submit", json=payload, timeout=60)
-    response.raise_for_status()
-    return response.json()
-
-# ------------------------
-# Agent factory
-# ------------------------
-def create_agent() -> CodeAgent:
+    space_id = os.getenv("SPACE_ID")
+
+    if profile:
+        username = profile.username
+        print(f"User logged in: {username}")
+    else:
+        print("User not logged in.")
+        return "Please login to Hugging Face with the button.", None
+
+    questions_url = f"{DEFAULT_API_URL}/questions"
+    submit_url = f"{DEFAULT_API_URL}/submit"
+
+    try:
+        agent = create_agent()
+        print("SmolAgent initialized.")
+    except Exception as e:
+        print(f"Error instantiating agent: {e}")
+        return f"Error initializing agent: {e}", None
+
+    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
+    print(f"Agent code URL: {agent_code}")
+
+    try:
+        response = requests.get(questions_url, timeout=15)
+        response.raise_for_status()
+        questions = response.json()
+        if not questions:
+            return "No questions fetched.", None
+        print(f"Fetched {len(questions)} questions.")
+    except Exception as e:
+        print(f"Error fetching questions: {e}")
+        return f"Error fetching questions: {e}", None
+
+    results = []
+    payload = []
+    for q in questions:
+        tid = q.get("task_id")
+        text = q.get("question")
+        if not tid or not text:
+            continue
+        try:
+            ans = agent.run(question=text)
+        except Exception as e:
+            ans = f"ERROR: {e}"
+        payload.append({"task_id": tid, "submitted_answer": ans})
+        results.append({"Task ID": tid, "Question": text, "Answer": ans})
+
+    if not payload:
+        return "Agent returned no answers.", pd.DataFrame(results)
+
+    submission = {"username": username, "agent_code": agent_code, "answers": payload}
+    try:
+        resp = requests.post(submit_url, json=submission, timeout=60)
+        resp.raise_for_status()
+        data = resp.json()
+        status = (
+            f"Submission Successful!\n"
+            f"User: {data.get('username')}\n"
+            f"Score: {data.get('score')}% ({data.get('correct_count')}/{data.get('total_attempted')})\n"
+            f"Message: {data.get('message')}"
+        )
+    except Exception as e:
+        print(f"Submission error: {e}")
+        status = f"Submission Failed: {e}"
+
+    return status, pd.DataFrame(results)
+
+
+def test_random_question(profile: gr.OAuthProfile | None):
     """
-    Factory that returns a configured CodeAgent instance.
-    Requires OPENAI_API_KEY in environment.
+    Fetch a random GAIA question and return the agent's answer.
     """
-    # Initialize the LLM with OpenAI API
-    llm = OpenAIServerModel(
-        model_id=os.getenv("OPENAI_MODEL_ID", "gpt-3.5-turbo"),
-        api_key=os.getenv("OPENAI_API_KEY")
-    )
-    # Create agent with defined tools
-    agent = CodeAgent(
-        tools=[fetch_questions, fetch_random_question, fetch_file, submit_answers],
-        model=llm
+    if not profile:
+        return "Please login to test.", ""
+    try:
+        q = fetch_random_question()
+        agent = create_agent()
+        ans = agent.run(question=q.get("question", ""))
+        return q.get("question", ""), ans
+    except Exception as e:
+        print(f"Test error: {e}")
+        return f"Error: {e}", ""
+
+# --- Gradio Interface ---
+with gr.Blocks() as demo:
+    gr.Markdown("# SmolAgent Evaluation Runner")
+    gr.Markdown(
+        """
+        **Istruzioni:**
+        1. Clone questo space e definisci la logica in agent.py.
+        2. Effettua il login con il tuo account Hugging Face.
+        3. Usa 'Run Evaluation & Submit All Answers' o 'Test Random Question'.
+        """
     )
-    return agent
+    login = gr.LoginButton()
+    run_all = gr.Button("Run Evaluation & Submit All Answers")
+    test = gr.Button("Test Random Question")
+    status = gr.Textbox(label="Status / Risultato", lines=5, interactive=False)
+    table = gr.DataFrame(label="Risultati Completi", wrap=True)
+    qbox = gr.Textbox(label="Domanda Casuale", lines=3, interactive=False)
+    abox = gr.Textbox(label="Risposta Agente", lines=3, interactive=False)
+
+    run_all.click(fn=run_and_submit_all, inputs=[login], outputs=[status, table])
+    test.click(fn=test_random_question, inputs=[login], outputs=[qbox, abox])
 
+if __name__ == "__main__":
+    demo.launch(debug=True, share=False)