Update app.py

app.py

@@ -2,27 +2,19 @@
 import os
 import gradio as gr
 import requests
-#from smolagents.agent import Agent
-from smolagents import Tool
-
-from smolagents import Agent
+import pandas as pd
 
+#from smolagents.agent import CodeAgent
+#from smolagents.models import HfApiModel
+from smolagents import Tool
+from smolagents import  CodeAgent, HfApiModel
 
 from audio_transcriber import AudioTranscriptionTool
 from image_analyzer import ImageAnalysisTool
 from wikipedia_searcher import WikipediaSearcher
 
 
-# Hugging Face API setup
-HF_API_TOKEN = os.getenv("HF_API_TOKEN")
-HF_CHAT_MODEL_URL = "https://api-inference.huggingface.com/models/HuggingFaceH4/zephyr-7b-beta"
-
-HEADERS = {
-    "Authorization": f"Bearer {HF_API_TOKEN}",
-    "Content-Type": "application/json"
-}
-
-# Static system prompt
+# System prompt
 SYSTEM_PROMPT = """You are an agent solving the GAIA benchmark and you are required to provide exact answers.
 Rules to follow:
 1. Return only the exact requested answer: no explanation and no reasoning.
@@ -39,9 +31,11 @@ Examples of good responses:
 Never include phrases like "the answer is..." or "Based on my research".
 Only return the exact answer."""
 
-# Agent tools
+
+# Tool definitions
 audio_tool = AudioTranscriptionTool()
 image_tool = ImageAnalysisTool()
+
 wiki_tool = Tool.from_function(
     name="wikipedia_search",
     description="Search for facts using Wikipedia.",
@@ -52,68 +46,153 @@ wiki_tool = Tool.from_function(
 
 tools = [audio_tool, image_tool, wiki_tool]
 
-agent = Agent(
-    tools=tools,
-    system_prompt=SYSTEM_PROMPT
-)
 
-def query_hf_model(prompt: str) -> str:
+# Agent factory
+def MyAgent():
+    return CodeAgent(
+        tools=tools,
+        system_prompt=SYSTEM_PROMPT,
+        model=HfApiModel(
+            api_url="https://api-inference.huggingface.com/models/HuggingFaceH4/zephyr-7b-beta",
+            api_key=os.getenv("HF_API_TOKEN")
+        )
+    )
+
+
+# Main run and submission logic
+def run_and_submit_all(profile: gr.OAuthProfile | None):
+    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
+
+    api_url = os.getenv("GAIA_API_URL", "https://gaia-benchmark.com/api")
+    questions_url = f"{api_url}/questions"
+    submit_url = f"{api_url}/submit"
+
+    try:
+        agent = MyAgent()
+    except Exception as e:
+        print(f"Error initializing 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}")
+
+    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:
+            return "Fetched questions list is empty or invalid format.", None
+        print(f"Fetched {len(questions_data)} questions.")
+    except Exception as e:
+        return f"Error fetching questions: {e}", None
+
+    results_log = []
+    answers_payload = []
+    print(f"Running agent on {len(questions_data)} questions...")
+
+    for item in questions_data:
+        task_id = item.get("task_id")
+        if not task_id:
+            continue
+        try:
+            submitted_answer = agent(item)
+            answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
+            results_log.append({
+                "Task ID": task_id,
+                "Question": item.get("question", ""),
+                "Submitted Answer": submitted_answer
+            })
+        except Exception as e:
+            error_msg = f"AGENT ERROR: {e}"
+            results_log.append({
+                "Task ID": task_id,
+                "Question": item.get("question", ""),
+                "Submitted Answer": error_msg
+            })
+
+    if not answers_payload:
+        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
+
+    submission_data = {
+        "username": username.strip(),
+        "agent_code": agent_code,
+        "answers": answers_payload
+    }
+
+    print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
     try:
-        response = requests.post(
-            HF_CHAT_MODEL_URL,
-            headers=HEADERS,
-            json={
-                "inputs": {
-                    "past_user_inputs": [],
-                    "text": prompt
-                },
-                "parameters": {
-                    "max_new_tokens": 256,
-                    "return_full_text": False
-                }
-            }
+        response = requests.post(submit_url, json=submission_data, timeout=60)
+        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.')}"
         )
-        result = response.json()
-        if isinstance(result, dict) and "error" in result:
-            return f"HF API Error: {result['error']}"
-        return result[0]["generated_text"].strip()
+        results_df = pd.DataFrame(results_log)
+        return final_status, results_df
+    except requests.exceptions.HTTPError as e:
+        try:
+            detail = e.response.json().get("detail", e.response.text)
+        except Exception:
+            detail = e.response.text[:500]
+        return f"Submission Failed: {detail}", pd.DataFrame(results_log)
+    except requests.exceptions.Timeout:
+        return "Submission Failed: The request timed out.", pd.DataFrame(results_log)
     except Exception as e:
-        return f"Error querying Hugging Face model: {e}"
-
-def run_and_submit_all(question, file):
-    if file:
-        file_path = file.name
-        if file_path.endswith((".mp3", ".wav")):
-            transcript = audio_tool.forward(file_path)
-            question = f"{question}\n\nTranscription of audio: {transcript}"
-        elif file_path.endswith((".png", ".jpg", ".jpeg")):
-            image_answer = image_tool.forward(file_path, question)
-            return image_answer
-        elif file_path.endswith(".py"):
-            try:
-                with open(file_path, "r") as f:
-                    code = f.read()
-                question = f"{question}\n\nPython code:\n{code}"
-            except Exception as e:
-                return f"Error reading code file: {e}"
-        else:
-            return "Unsupported file type."
-
-    full_prompt = f"{SYSTEM_PROMPT}\nQUESTION:\n{question}"
-    return query_hf_model(full_prompt)
-
-with gr.Blocks(title="GAIA Agent with HF API") as demo:
-    gr.Markdown("### GAIA Evaluation Agent (Hugging Face-based)")
-
-    with gr.Row():
-        question_input = gr.Textbox(label="Question", placeholder="Enter your question here...", lines=3)
-        file_input = gr.File(label="Optional File (Audio, Image, or Python)", file_types=[".mp3", ".wav", ".jpg", ".jpeg", ".png", ".py"])
-
-    submit_button = gr.Button("Run Agent")
-    output_box = gr.Textbox(label="Answer")
-
-    submit_button.click(fn=run_and_submit_all, inputs=[question_input, file_input], outputs=output_box)
+        return f"An unexpected error occurred during submission: {e}", pd.DataFrame(results_log)
 
+
+# Gradio UI setup
+with gr.Blocks() as demo:
+    gr.Markdown("# Basic Agent Evaluation Runner")
+    gr.Markdown("""
+        **Instructions:**
+        1. Clone this space, modify code to define your agent's logic, tools, and packages.
+        2. Log in to your Hugging Face account using the button below.
+        3. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see your score.
+        **Note:** Submitting can take some time.
+    """)
+
+    gr.LoginButton()
+    run_button = gr.Button("Run Evaluation & Submit All Answers")
+
+    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])
+
+# App startup logs
 if __name__ == "__main__":
-    demo.launch()
+    print("\n" + "-" * 30 + " App Starting " + "-" * 30)
+    space_host = os.getenv("SPACE_HOST")
+    space_id = os.getenv("SPACE_ID")
+
+    if space_host:
+        print(f"✅ SPACE_HOST found: {space_host}")
+        print(f"   Runtime URL should be: https://{space_host}.hf.space")
+    else:
+        print("ℹ️  SPACE_HOST environment variable not found (running locally?).")
+
+    if space_id:
+        print(f"✅ SPACE_ID found: {space_id}")
+        print(f"   Repo URL: https://huggingface.co/spaces/{space_id}")
+        print(f"   Repo Tree URL: https://huggingface.co/spaces/{space_id}/tree/main")
+    else:
+        print("ℹ️  SPACE_ID environment variable not found (running locally?).")
+
+    print("-" * (60 + len(" App Starting ")) + "\n")
+    print("Launching Gradio Interface for Basic Agent Evaluation...")
+    demo.launch(debug=True, share=False)
+