test2

app.py

@@ -1,17 +1,20 @@
 import os
 import datetime
 import requests
-import gradio as gr
 import pandas as pd
+import gradio as gr
 from openai import OpenAI
 
+DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
+
+# -------- Tool-Enhanced Agent --------
 class ToolEnhancedAgent:
     def __init__(self):
         api_key = os.getenv("OPENAI_API_KEY")
         if not api_key:
             raise ValueError("OPENAI_API_KEY is not set.")
         self.client = OpenAI(api_key=api_key)
-        print("✅ ToolEnhancedAgent initialized with GPT-4 + CoT + Tool usage.")
+        print("✅ ToolEnhancedAgent initialized with GPT-4 + CoT + Tools.")
 
     def use_tool(self, tool_name: str, input_text: str) -> str:
         try:
@@ -38,8 +41,8 @@ class ToolEnhancedAgent:
 
     def __call__(self, question: str) -> str:
         prompt = (
-            "You are a helpful AI assistant. Use tools when needed. "
-            "Think step-by-step before answering. Respond clearly.\n\n"
+            "You are a helpful AI assistant. You can use tools (calculator, date, wikipedia). "
+            "Think step-by-step before answering.\n\n"
             f"Question: {question}\n"
             "Answer (show thinking steps):"
         )
@@ -48,7 +51,7 @@ class ToolEnhancedAgent:
             response = self.client.chat.completions.create(
                 model="gpt-4",
                 messages=[
-                    {"role": "system", "content": "You are a smart assistant that can use tools and think step-by-step."},
+                    {"role": "system", "content": "You are a smart assistant that uses tools and thinks step-by-step."},
                     {"role": "user", "content": prompt}
                 ],
                 temperature=0.3,
@@ -60,3 +63,85 @@ class ToolEnhancedAgent:
         except Exception as e:
             print(f"[Agent Error]: {e}")
             return f"[Agent Error: {e}]"
+
+# -------- Evaluation & Submission Function --------
+def run_and_submit_all(profile: gr.OAuthProfile | None):
+    space_id = os.getenv("SPACE_ID")
+    if not profile:
+        return "Please login with your Hugging Face account.", None
+    username = profile.username
+
+    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
+    questions_url = f"{DEFAULT_API_URL}/questions"
+    submit_url = f"{DEFAULT_API_URL}/submit"
+
+    try:
+        agent = ToolEnhancedAgent()
+    except Exception as e:
+        return f"Agent init error: {e}", None
+
+    try:
+        response = requests.get(questions_url, timeout=15)
+        response.raise_for_status()
+        questions = response.json()
+    except Exception as e:
+        return f"Failed to fetch questions: {e}", None
+
+    results_log = []
+    answers_payload = []
+
+    for item in questions:
+        task_id = item.get("task_id")
+        question_text = item.get("question")
+        if not task_id or not question_text:
+            continue
+        try:
+            answer = agent(question_text)
+        except Exception as e:
+            answer = f"[Agent Error: {e}]"
+        results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": answer})
+        answers_payload.append({"task_id": task_id, "submitted_answer": answer})
+
+    submission = {
+        "username": username,
+        "agent_code": agent_code,
+        "answers": answers_payload,
+    }
+
+    try:
+        response = requests.post(submit_url, json=submission, timeout=60)
+        response.raise_for_status()
+        result = response.json()
+        status = (
+            f"✅ Submission Successful!\n"
+            f"User: {result.get('username')}\n"
+            f"Score: {result.get('score')}%\n"
+            f"Correct: {result.get('correct_count')}/{result.get('total_attempted')}\n"
+            f"Message: {result.get('message')}"
+        )
+    except Exception as e:
+        status = f"❌ Submission failed: {e}"
+
+    return status, pd.DataFrame(results_log)
+
+# -------- Gradio Interface --------
+with gr.Blocks() as demo:
+    gr.Markdown("## 🤖 GAIA Agent Evaluation with ToolEnhancedAgent")
+    gr.Markdown(
+        """
+        - This Space lets you run your agent on GAIA benchmark.
+        - Login with your HF account, click "Run Evaluation", and wait for the results.
+        """
+    )
+
+    gr.LoginButton()
+    run_button = gr.Button("🚀 Run Evaluation & Submit")
+    status_output = gr.Textbox(label="Status / Score", lines=6, interactive=False)
+    table_output = gr.DataFrame(label="Agent Answers")
+
+    run_button.click(fn=run_and_submit_all, outputs=[status_output, table_output])
+
+# -------- Launch App --------
+if __name__ == "__main__":
+    print("✅ Launching GAIA Agent Evaluation App")
+    demo.launch(debug=True)