Update app.py

app.py

@@ -5,7 +5,6 @@ from langchain_groq import ChatGroq
 from langchain_community.utilities import ArxivAPIWrapper, WikipediaAPIWrapper
 from langchain_community.tools import ArxivQueryRun, WikipediaQueryRun, DuckDuckGoSearchRun
 from langchain.agents import initialize_agent, AgentType
-from langchain.callbacks import StreamlitCallbackHandler
 import os
 import requests
 import pandas as pd
@@ -17,12 +16,21 @@ load_dotenv()
 # Constants for Basic Agent Evaluation
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# Initialize search tools
-api_wrapper_arxiv = ArxivAPIWrapper(top_k_results=1, doc_content_chars_max=250)
-arxiv = ArxivQueryRun(api_wrapper=api_wrapper_arxiv)
-api_wrapper_wiki = WikipediaAPIWrapper(top_k_results=1, doc_content_chars_max=250)
-wiki = WikipediaQueryRun(api_wrapper=api_wrapper_wiki)
-search = DuckDuckGoSearchRun(name="Search")
+# Initialize search tools (with warm-up)
+@st.cache_resource
+def load_tools():
+    with st.spinner("Initializing tools (first time may take a few seconds)..."):
+        api_wrapper_arxiv = ArxivAPIWrapper(top_k_results=1, doc_content_chars_max=250)
+        arxiv = ArxivQueryRun(api_wrapper=api_wrapper_arxiv)
+        api_wrapper_wiki = WikipediaAPIWrapper(top_k_results=1, doc_content_chars_max=250)
+        wiki = WikipediaQueryRun(api_wrapper=api_wrapper_wiki)
+        search = DuckDuckGoSearchRun(name="Search")
+        # Warm up tools
+        arxiv.run("machine learning")
+        wiki.run("machine learning")
+        return [search, arxiv, wiki]
+
+tools = load_tools()
 
 # Streamlit app layout
 st.title("Langchain - Chat with Search & Evaluation")
@@ -50,14 +58,11 @@ if prompt := st.chat_input(placeholder="What is machine learning?"):
         st.error("Please enter your Groq API key in the sidebar.")
         st.stop()
 
-    llm = ChatGroq(groq_api_key=api_key, model_name="Llama3-8b-8192", streaming=True)
-    tools = [search, arxiv, wiki]
-
+    llm = ChatGroq(groq_api_key=api_key, model_name="Llama3-8b-8192")
     search_agent = initialize_agent(tools, llm, agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION, handle_parsing_errors=True)
 
     with st.chat_message("assistant"):
-        st_cb = StreamlitCallbackHandler(st.container(), expand_new_thoughts=False)
-        response = search_agent.run(st.session_state.messages, callbacks=[st_cb])
+        response = search_agent.run(st.session_state.messages)
         st.session_state.messages.append({'role': 'assistant', "content": response})
         st.write(response)
 
@@ -65,84 +70,96 @@ if prompt := st.chat_input(placeholder="What is machine learning?"):
 st.sidebar.title("Basic Agent Evaluation")
 
 def run_evaluation():
-    """Function to run the evaluation using the current Groq-powered agent"""
+    """Function to run the evaluation with progress updates"""
     if not api_key:
         st.error("Please enter your Groq API key in the sidebar.")
         return "API key required", pd.DataFrame()
 
-    username = "streamlit_user"  # Default username for Streamlit
+    # Setup progress tracking
+    progress_bar = st.sidebar.progress(0)
+    status_text = st.sidebar.empty()
+    results_container = st.empty()
+    
+    username = "streamlit_user"
     api_url = DEFAULT_API_URL
     questions_url = f"{api_url}/questions"
     submit_url = f"{api_url}/submit"
     space_id = os.getenv("SPACE_ID", "local")
     agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" if space_id != "local" else "local_execution"
 
-    # 1. Instantiate Agent
-    try:
-        llm = ChatGroq(groq_api_key=api_key, model_name="Llama3-8b-8192")
-        tools = [search, arxiv, wiki]
-        agent = initialize_agent(tools, llm, agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION, handle_parsing_errors=True)
-    except Exception as e:
-        return f"Error initializing agent: {e}", pd.DataFrame()
-
-    # 2. Fetch Questions
     try:
+        # 1. Fetch Questions
+        status_text.text("📡 Fetching questions...")
         response = requests.get(questions_url, timeout=15)
         response.raise_for_status()
         questions_data = response.json()
+        total_questions = len(questions_data)
+        status_text.text(f"✅ Found {total_questions} questions")
+
         if not questions_data:
-            return "Fetched questions list is empty or invalid format.", pd.DataFrame()
-    except Exception as e:
-        return f"Error fetching questions: {e}", pd.DataFrame()
-
-    # 3. Run Agent
-    results_log = []
-    answers_payload = []
-    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:
-            continue
-        try:
-            submitted_answer = agent.run(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:
-            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
-
-    if not answers_payload:
-        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
-
-    # 4. Prepare and Submit
-    submission_data = {
-        "username": username,
-        "agent_code": agent_code,
-        "answers": answers_payload
-    }
+            return "No questions found", pd.DataFrame()
 
-    try:
+        # 2. Initialize Agent (reuse tools from cache)
+        llm = ChatGroq(groq_api_key=api_key, model_name="Llama3-8b-8192")
+        agent = initialize_agent(tools, llm, agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION, handle_parsing_errors=True)
+
+        # 3. Process Questions
+        results_log = []
+        answers_payload = []
+        
+        for i, item in enumerate(questions_data):
+            progress = (i + 1) / total_questions
+            progress_bar.progress(progress)
+            status_text.text(f"🔍 Processing question {i+1}/{total_questions}...")
+            
+            task_id = item.get("task_id")
+            question_text = item.get("question")
+            if not task_id or not question_text:
+                continue
+
+            try:
+                submitted_answer = agent.run(question_text)
+                answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
+                results_log.append({"Task ID": task_id, "Question": question_text[:100] + "..." if len(question_text) > 100 else question_text, 
+                                 "Submitted Answer": submitted_answer[:200] + "..." if len(submitted_answer) > 200 else submitted_answer})
+                
+                # Update results table progressively
+                if (i + 1) % 3 == 0 or (i + 1) == total_questions:  # Update every 3 questions or at end
+                    results_container.dataframe(pd.DataFrame(results_log))
+            except Exception as e:
+                results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"❌ Error: {str(e)}"})
+
+        # 4. Submit Answers
+        status_text.text("📤 Submitting answers...")
+        submission_data = {"username": username, "agent_code": agent_code, "answers": answers_payload}
         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.')}"
+            f"✅ Submission Successful!\n"
+            f"📊 Score: {result_data.get('score', 'N/A')}%\n"
+            f"📝 Correct: {result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')}\n"
+            f"💬 Message: {result_data.get('message', 'No message')}"
         )
         return final_status, pd.DataFrame(results_log)
+
     except Exception as e:
-        return f"Submission Failed: {e}", pd.DataFrame(results_log)
+        return f"❌ Failed: {str(e)}", pd.DataFrame(results_log if 'results_log' in locals() else [])
+    
+    finally:
+        progress_bar.empty()
+        status_text.empty()
 
 # Evaluation button in sidebar
-if st.sidebar.button("Run Evaluation & Submit Answers"):
-    with st.spinner("Running evaluation and submitting answers..."):
+if st.sidebar.button("🚀 Run Evaluation & Submit Answers"):
+    with st.spinner("Starting evaluation..."):
         status, results = run_evaluation()
-        
-    st.sidebar.text_area("Evaluation Status", value=status, height=150)
+    
+    st.sidebar.success("Evaluation completed!")
+    st.sidebar.text_area("Results", value=status, height=150)
     
     if not results.empty:
-        st.subheader("Evaluation Results")
-        st.dataframe(results)
+        st.subheader("📋 Detailed Results")
+        st.dataframe(results, use_container_width=True)