Update app.py

app.py

@@ -2,96 +2,124 @@ import os
 import gradio as gr
 import requests
 import pandas as pd
-from huggingface_hub import InferenceClient
-from duckduckgo_search import DDGS
 from datasets import load_dataset
+from duckduckgo_search import DDGS
+from llama_index.llms.huggingface import HuggingFaceLLM
+from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, ServiceContext
+from huggingface_hub import InferenceClient
+import wikipediaapi
 
+# Constants
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
-
-# Hugging Face Token (set in environment)
 HF_TOKEN = os.environ.get("HF_TOKEN")
-deepseek_model = "deepseek-ai/DeepSeek-R1"
-hf_client = InferenceClient(model=deepseek_model, token=HF_TOKEN)
 
-# Load Wikipedia dataset (small subset for efficient retrieval)
-wiki_dataset = load_dataset("wikipedia", "20220301.en", split="train[:10000]")
+# Advanced LLM via Hugging Face Inference API
+llm_model_id = "deepseek-ai/DeepSeek-R1"
+hf_client = InferenceClient(llm_model_id, token=HF_TOKEN)
+
+# Wikipedia API setup
+wiki_api = wikipediaapi.Wikipedia('en')
 
-def search_wikipedia(question):
-    results = wiki_dataset.filter(lambda x: question.lower() in x["text"].lower())
-    if len(results):
-        return results[0]["text"][:1000]  # limit to first 1000 chars
-    return "No relevant information found on Wikipedia."
+# Load Wikipedia dataset from Hugging Face
+wiki_dataset = load_dataset(
+    "wikipedia", "20220301.en", split="train[:10000]", trust_remote_code=True
+)
 
+# DuckDuckGo search function
 def duckduckgo_search(query):
     with DDGS() as ddgs:
-        results = [r["body"] for r in ddgs.text(query, max_results=3)]
-        return "\n".join(results) if results else "No results found."
+        results = [r for r in ddgs.text(query, max_results=3)]
+        if results:
+            return "\n".join([r["body"] for r in results if r.get("body")])
+        else:
+            return "No results found."
 
-def ask_deepseek(prompt, max_tokens=512):
-    try:
-        response = hf_client.text_generation(
-            prompt, max_new_tokens=max_tokens, temperature=0.2, repetition_penalty=1.1
+# Smart Agent combining multiple sources
+class SmartAgent:
+    def __init__(self):
+        service_context = ServiceContext.from_defaults(
+            llm=HuggingFaceLLM(model_name=llm_model_id, token=HF_TOKEN)
         )
-        return response
-    except Exception as e:
-        return f"DeepSeek Error: {e}"
+        docs = [doc["text"] for doc in wiki_dataset]
+        self.index = VectorStoreIndex.from_documents(
+            [SimpleDirectoryReader.input_to_document(doc) for doc in docs],
+            service_context=service_context,
+            show_progress=True
+        )
+        self.query_engine = self.index.as_query_engine()
 
-class SmartAgent:
     def __call__(self, question: str) -> str:
-        q_lower = question.lower()
-        if any(term in q_lower for term in ["current", "latest", "2024", "2025", "recent", "live", "today", "now"]):
+        question_lower = question.lower()
+
+        # Use DuckDuckGo for recent events, dates, or temporal queries
+        if any(term in question_lower for term in ["current", "latest", "2024", "2025", "recent", "today", "president"]):
             return duckduckgo_search(question)
-        deepseek_response = ask_deepseek(question)
-        if "DeepSeek Error" not in deepseek_response and deepseek_response.strip():
-            return deepseek_response
-        # fallback to Wikipedia if DeepSeek fails
-        return search_wikipedia(question)
 
+        # Check if Wikipedia page exists for topic
+        page = wiki_api.page(question)
+        if page.exists():
+            return page.summary[:1000] + "..."
+
+        # Fallback to indexed Wikipedia with RAG
+        try:
+            response = self.query_engine.query(question)
+            return str(response)
+        except Exception as e:
+            return f"LLM query error: {e}"
+
+# Run and submit evaluation
 def run_and_submit_all(profile: gr.OAuthProfile | None):
-    if not profile:
+    space_id = os.getenv("SPACE_ID")
+    if profile:
+        username = f"{profile.username}"
+    else:
         return "Please Login to Hugging Face with the button.", None
-    username = profile.username
+
     questions_url = f"{DEFAULT_API_URL}/questions"
     submit_url = f"{DEFAULT_API_URL}/submit"
-    agent_code = f"https://huggingface.co/spaces/{os.getenv('SPACE_ID')}/tree/main"
-    
+
+    # Instantiate agent
+    agent = SmartAgent()
+    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
+
+    # Fetch questions
     try:
-        agent = SmartAgent()
+        questions_data = requests.get(questions_url).json()
     except Exception as e:
-        return f"Agent Error: {e}", None
+        return f"Error fetching questions: {e}", None
 
-    questions_data = requests.get(questions_url).json()
     results_log, answers_payload = [], []
-
     for item in questions_data:
-        task_id = item.get("task_id")
-        question_text = item.get("question")
-        if task_id and question_text:
-            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})
+        task_id, question_text = item.get("task_id"), item.get("question")
+        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})
 
     submission_data = {"username": username, "agent_code": agent_code, "answers": answers_payload}
-    response = requests.post(submit_url, json=submission_data).json()
-
-    final_status = (
-        f"Submission Successful!\n"
-        f"User: {response.get('username')}\n"
-        f"Overall Score: {response.get('score', 'N/A')}%\n"
-        f"({response.get('correct_count', '?')}/{response.get('total_attempted', '?')} correct)\n"
-        f"Message: {response.get('message', 'No message received.')}"
-    )
 
-    return final_status, pd.DataFrame(results_log)
+    try:
+        result_data = requests.post(submit_url, json=submission_data).json()
+        final_status = (
+            f"Submission Successful!\n"
+            f"User: {result_data.get('username')}\n"
+            f"Overall Score: {result_data.get('score')}%\n"
+            f"({result_data.get('correct_count')}/{result_data.get('total_attempted')}) correct\n"
+            f"Message: {result_data.get('message')}"
+        )
+        results_df = pd.DataFrame(results_log)
+        return final_status, results_df
+    except Exception as e:
+        return f"Submission Failed: {e}", pd.DataFrame(results_log)
 
+# Gradio interface setup
 with gr.Blocks() as demo:
-    gr.Markdown("# Smart Agent Evaluation Runner")
+    gr.Markdown("# 🚀 Smart Multi-Source Agent Evaluation")
     gr.LoginButton()
     run_button = gr.Button("Run Evaluation & Submit All Answers")
-    status_output = gr.Textbox(label="Run Status", lines=5, interactive=False)
-    results_table = gr.DataFrame(label="Questions and Answers")
+    status_output = gr.Textbox(label="Status & Results", lines=6, interactive=False)
+    results_table = gr.DataFrame(label="Agent Answers")
 
-    run_button.click(run_and_submit_all, outputs=[status_output, results_table])
+    run_button.click(fn=run_and_submit_all, outputs=[status_output, results_table])
 
 if __name__ == "__main__":
-    demo.launch(debug=True)
+    demo.launch(debug=True, share=False)