Update app.py

app.py

@@ -3,76 +3,91 @@ import gradio as gr
 import requests
 import pandas as pd
 import json
-from transformers import pipeline
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# --- Smart Agent Definition ---
-class ZeroShotAgent:
+# --- Smart Agent using Mistral-7B ---
+class SmartLLMAgent:
     def __init__(self):
-        print("ZeroShotAgent initialized.")
-        self.classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
+        print("🔁 Loading SmartLLMAgent with Mistral-7B...")
+        self.tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2")
+        self.model = AutoModelForCausalLM.from_pretrained(
+            "mistralai/Mistral-7B-Instruct-v0.2",
+            torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
+            device_map="auto"
+        )
 
     def __call__(self, question: str) -> str:
-        print(f"Received question: {question[:100]}")
-        labels = ["Yes", "No", "Not Enough Information"]
-        result = self.classifier(question, labels)
-        best_answer = result["labels"][0] if result and "labels" in result else "Not Enough Information"
-        print(f"Answer: {best_answer}")
-        return best_answer or "Not Enough Information"
-
+        prompt = f"Answer the following question clearly and precisely:\n\nQuestion: {question}\n\nAnswer:"
+        inputs = self.tokenizer(prompt, return_tensors="pt", truncation=True, max_length=2048)
+        inputs = {k: v.to(self.model.device) for k, v in inputs.items()}
+
+        with torch.no_grad():
+            outputs = self.model.generate(
+                **inputs,
+                max_new_tokens=150,
+                do_sample=True,
+                top_p=0.9,
+                temperature=0.7,
+                pad_token_id=self.tokenizer.eos_token_id
+            )
+
+        decoded = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
+        final_answer = decoded.split("Answer:")[-1].strip()
+        print(f"✅ Generated answer: {final_answer}")
+        return final_answer or "Not Enough Information"
+
+# --- Run Agent & Submit ---
 def run_and_submit_all(profile: gr.OAuthProfile | None):
     space_id = os.getenv("SPACE_ID")
 
     if profile:
         username = f"{profile.username}"
-        print(f"User logged in: {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 = DEFAULT_API_URL
-    questions_url = f"{api_url}/questions"
-    submit_url = f"{api_url}/submit"
+    questions_url = f"{DEFAULT_API_URL}/questions"
+    submit_url = f"{DEFAULT_API_URL}/submit"
 
     try:
-        agent = ZeroShotAgent()
+        agent = SmartLLMAgent()
     except Exception as e:
-        print(f"Error instantiating agent: {e}")
-        return f"Error initializing agent: {e}", None
+        return f"❌ Error loading agent: {e}", None
 
     agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
-    print(f"Agent code URL: {agent_code}")
+    print(f"🔗 Agent Code: {agent_code}")
 
     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.")
+        questions = response.json()
+        print(f"📥 Fetched {len(questions)} questions.")
     except Exception as e:
-        return f"Error fetching questions: {e}", None
+        return f"❌ Failed to fetch questions: {e}", None
 
-    results_log = []
     answers_payload = []
-    for item in questions_data:
+    results_log = []
+
+    for item in questions:
         task_id = item.get("task_id")
         question_text = item.get("question")
-        if not task_id or question_text is None:
+
+        if not task_id or not question_text:
             continue
+
         try:
-            submitted_answer = agent(question_text)
-            if not submitted_answer:
-                submitted_answer = "Not Enough Information"
-            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})
+            answer = agent(question_text)
+            answers_payload.append({"task_id": task_id, "submitted_answer": answer})
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": answer})
         except Exception as e:
-            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"ERROR: {e}"})
 
     if not answers_payload:
-        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
+        return "❌ No answers produced.", pd.DataFrame(results_log)
 
     submission_data = {
         "username": username.strip(),
@@ -80,47 +95,43 @@ def run_and_submit_all(profile: gr.OAuthProfile | None):
         "answers": answers_payload
     }
 
-    # Debug payload
-    print("Submission Payload Preview:")
-    print(json.dumps(submission_data, indent=2)[:1000])  # Safe preview log
+    print("📤 Submission Payload:")
+    print(json.dumps(submission_data, indent=2)[:1000])
 
     try:
         response = requests.post(submit_url, json=submission_data, timeout=60)
         response.raise_for_status()
-        result_data = response.json()
-        final_status = (
+        result = response.json()
+        status = (
             f"✅ Submission Successful!\n"
-            f"👤 User: {result_data.get('username')}\n"
-            f"📊 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.')}"
+            f"👤 User: {result.get('username')}\n"
+            f"📊 Score: {result.get('score', 'N/A')}% "
+            f"({result.get('correct_count', '?')}/{result.get('total_attempted', '?')} correct)\n"
+            f"📝 Message: {result.get('message', 'No message received.')}"
         )
-        return final_status, pd.DataFrame(results_log)
-    except requests.exceptions.RequestException as e:
-        return f"❌ Submission Failed: {str(e)}", pd.DataFrame(results_log)
+        return status, pd.DataFrame(results_log)
     except Exception as e:
-        return f"❌ Unexpected Error: {e}", pd.DataFrame(results_log)
+        return f"❌ Submission Failed: {e}", pd.DataFrame(results_log)
 
 # --- Gradio UI ---
 with gr.Blocks() as demo:
-    gr.Markdown("# 🤖 Smart Agent Evaluation Runner")
+    gr.Markdown("# 🧠 Smart Agent Evaluation Runner (Mistral-7B)")
     gr.Markdown(
         """
         ✅ **Instructions:**
         1. Login to your Hugging Face account.
-        2. Click 'Run Evaluation & Submit All Answers'.
-        3. Wait ~1–2 mins for results.
-        📌 You need at least **30% correct** on Level 1 to unlock the certificate.
+        2. Click **"Run Evaluation & Submit All Answers"**.
+        3. Wait 1–2 mins. You need at least **30% correct** to pass Level 1 and get the certificate.
         """
     )
 
     gr.LoginButton()
     run_button = gr.Button("🚀 Run Evaluation & Submit All Answers")
-    status_output = gr.Textbox(label="📝 Run Status / Submission Result", lines=6, interactive=False)
-    results_table = gr.DataFrame(label="📋 Questions and Agent Answers", wrap=True)
+    status_output = gr.Textbox(label="📝 Submission Result", lines=6, interactive=False)
+    results_table = gr.DataFrame(label="📋 Agent Answers", wrap=True)
 
     run_button.click(fn=run_and_submit_all, outputs=[status_output, results_table])
 
 if __name__ == "__main__":
-    print("Launching Smart Agent Gradio UI...")
+    print("🚀 Launching Smart Agent with Mistral-7B...")
     demo.launch(debug=True, share=False)