Update app.py

app.py

@@ -1,160 +1,118 @@
 import os
 import gradio as gr
 import requests
+import time
 import pandas as pd
-from transformers import pipeline
+from dotenv import load_dotenv
+
+# Load environment variables
+load_dotenv()
 
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
+HF_API_URL = "https://api-inference.huggingface.co/models/mistralai/Mistral-7B-Instruct-v0.2"
+HUGGINGFACE_TOKEN = os.getenv("HUGGINGFACE_TOKEN")
 
-# --- Basic Agent Definition ---
+# --- Agent Definition ---
 class BasicAgent:
     def __init__(self):
-        print("Loading FLAN-T5 model...")
-        self.model = pipeline("text2text-generation", model="google/flan-t5-base", max_length=256)
-        print("Model loaded.")
-
+        self.headers = {"Authorization": f"Bearer {HUGGINGFACE_TOKEN}"}
+    
     def __call__(self, question: str) -> str:
-        prompt = f"Answer the following question accurately and concisely:\n{question.strip()}"
-        result = self.model(prompt)
-        answer = result[0]['generated_text'].strip()
-        print(f"Answer generated: {answer}")
-        return answer
-
-# --- Helper Functions ---
-def fetch_questions(api_url: str):
-    """
-    Fetches the questions from the provided API URL.
-    Returns a list of questions or raises an error.
-    """
-    questions_url = f"{api_url}/questions"
-    try:
-        response = requests.get(questions_url, timeout=15)
-        response.raise_for_status()
-        questions_data = response.json()
-        if not questions_data:
-            print("Fetched questions list is empty.")
-            return None
-        print(f"Fetched {len(questions_data)} questions.")
-        return questions_data
-    except requests.exceptions.RequestException as e:
-        print(f"Error fetching questions: {e}")
-        return 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 None
-
-def submit_answers(api_url: str, submission_data: dict):
-    """
-    Submits the answers to the leaderboard API.
-    """
-    submit_url = f"{api_url}/submit"
-    try:
-        response = requests.post(submit_url, json=submission_data, timeout=60)
-        response.raise_for_status()
-        result_data = response.json()
-        return result_data
-    except requests.exceptions.RequestException as e:
-        print(f"Submission failed: {e}")
-        return None
-
-# --- Main Execution Function ---
+        prompt = f"""Answer the question concisely and factually. Provide only the answer in the exact format required by the question, with no additional text, explanations, or formatting. Do not use quotation marks, bullet points, or any other formatting. Ensure correct casing and pluralization as specified.
+
+Question: {question}
+Answer: """
+        payload = {
+            "inputs": prompt,
+            "parameters": {
+                "max_new_tokens": 100,
+                "return_full_text": False
+            }
+        }
+        max_retries = 3
+        for _ in range(max_retries):
+            try:
+                response = requests.post(HF_API_URL, headers=self.headers, json=payload, timeout=30)
+                response.raise_for_status()
+                answer = response.json()[0]['generated_text'].strip()
+                # Clean up answer
+                answer = answer.strip().rstrip('.').strip()
+                return answer
+            except requests.exceptions.HTTPError as e:
+                if e.response.status_code == 503:
+                    time.sleep(10)
+                else:
+                    print(f"HTTP Error: {e.response.text}")
+                    break
+            except requests.exceptions.RequestException as e:
+                print(f"Request failed: {e}, retrying...")
+                time.sleep(5)
+            except Exception as e:
+                print(f"Error: {e}")
+                break
+        return "Error: Unable to generate answer."
+
+# --- Submission Logic ---
 def run_and_submit_all(profile: gr.OAuthProfile | None):
-    """
-    Fetches all questions, runs the agent on them, and submits all answers.
-    """
-    # --- Check Hugging Face Profile ---
     if profile:
-        username = profile.username
-        print(f"User logged in: {username}")
+        username = f"{profile.username}"
     else:
-        print("User not logged in.")
-        return "Please log in to Hugging Face with the button.", None
+        return "Please Login to Hugging Face with the button.", None
+
+    agent = BasicAgent()
+    space_id = os.getenv("SPACE_ID")
+    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" if space_id else ""
 
-    # --- Initialize Agent ---
     try:
-        agent = BasicAgent()
+        response = requests.get(f"{DEFAULT_API_URL}/questions", timeout=15)
+        response.raise_for_status()
+        questions_data = response.json()
     except Exception as e:
-        print(f"Error initializing agent: {e}")
-        return f"Error initializing agent: {e}", None
-
-    # --- Fetch Questions ---
-    questions_data = fetch_questions(DEFAULT_API_URL)
-    if not questions_data:
-        return "Failed to fetch questions.", None
+        return f"Error fetching questions: {e}", None
 
-    # --- Process Questions ---
     answers_payload = []
     results_log = []
     for item in 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}")
+        if not task_id or not question_text:
             continue
         try:
-            submitted_answer = agent.call(question_text)
+            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})
         except Exception as e:
-            print(f"Error processing task {task_id}: {e}")
-            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
-
-    # --- Prepare Submission Data ---
-    if not answers_payload:
-        return "No answers to submit.", pd.DataFrame(results_log)
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"Error: {e}"})
 
     submission_data = {
         "username": username.strip(),
-        "agent_code": f"https://huggingface.co/spaces/{os.getenv('SPACE_ID')}/tree/main",
+        "agent_code": agent_code,
         "answers": answers_payload
     }
 
-    # --- Submit Answers ---
-    result_data = submit_answers(DEFAULT_API_URL, submission_data)
-    if not result_data:
-        return "Submission failed.", pd.DataFrame(results_log)
-
-    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.')}"
-    )
-
-    results_df = pd.DataFrame(results_log)
-    return final_status, results_df
+    try:
+        response = requests.post(f"{DEFAULT_API_URL}/submit", json=submission_data, timeout=60)
+        response.raise_for_status()
+        result_data = response.json()
+        final_status = f"Success! Score: {result_data.get('score', 'N/A')}%"
+        return final_status, pd.DataFrame(results_log)
+    except Exception as e:
+        return f"Submission failed: {e}", pd.DataFrame(results_log)
 
 # --- Gradio Interface ---
 with gr.Blocks() as demo:
-    gr.Markdown("# Basic Agent Evaluation Runner")
-    gr.Markdown(
-    """
+    gr.Markdown("# GAIA Evaluation Agent")
+    gr.Markdown("""
     **Instructions:**
-    1. Clone this space and modify the code to define your agent's logic and tools.
-    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 the score.
-    
-    ---
-    **Disclaimers:**
-    Clicking "submit" might take some time to process all questions and submit answers.
-    """
-    )
-
+    1. Log in to your Hugging Face account using the button below.
+    2. Click 'Run Evaluation & Submit Answers' to fetch questions, run your agent, and submit answers.
+    """)
     gr.LoginButton()
+    run_button = gr.Button("Run Evaluation & Submit Answers")
+    status_output = gr.Textbox(label="Status")
+    results_table = gr.DataFrame(label="Results")
+    run_button.click(run_and_submit_all, outputs=[status_output, results_table])
 
-    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]
-    )
-
-# --- Run the Interface ---
 if __name__ == "__main__":
-    print("\n" + "-"*30 + " App Starting " + "-"*30)
-    demo.launch(debug=True, share=False)
+    demo.launch()