Modified

Integrated Langchain and replaced the respond() function with prompt templating.

app.py

@@ -1,13 +1,77 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from langchain.prompts import ChatPromptTemplate
+from langchain_community.llms import HuggingFaceHub
+from langchain.schema import HumanMessage, SystemMessage
+import os
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# Set up the LangChain model (using the same Zephyr model)
+llm = HuggingFaceHub(
+    repo_id="HuggingFaceH4/zephyr-7b-beta",
+    model_kwargs={"temperature": 0.7, "max_length": 512}
+)
+
+# Create different prompt templates for different subjects
+math_template = ChatPromptTemplate.from_messages([
+    ("system", """You are an expert math tutor. For every math problem:
+    1. Break it down step-by-step
+    2. Explain the reasoning behind each step
+    3. Show all work clearly
+    4. Check your answer
+    5. Ask if the student has questions about any step"""),
+    ("human", "{question}")
+])
+
+research_template = ChatPromptTemplate.from_messages([
+    ("system", """You are a research skills mentor. Help students with:
+    - Finding reliable sources
+    - Evaluating source credibility  
+    - Proper citation formats
+    - Research strategies
+    - Academic writing techniques
+    Always provide specific, actionable advice."""),
+    ("human", "{question}")
+])
+
+study_template = ChatPromptTemplate.from_messages([
+    ("system", """You are a study skills coach. Help students with:
+    - Effective study methods
+    - Time management
+    - Memory techniques
+    - Test preparation strategies
+    - Learning style optimization
+    Provide personalized, practical advice."""),
+    ("human", "{question}")
+])
 
+general_template = ChatPromptTemplate.from_messages([
+    ("system", """You are EduBot, a friendly AI learning assistant. You help students with:
+    📐 Math problems (step-by-step solutions)
+    🔍 Research skills (finding sources, citations)
+    📚 Study strategies (effective learning techniques)
+    🛠️ Educational tools (learning resources)
+    
+    Always be encouraging, patient, and thorough in your explanations."""),
+    ("human", "{question}")
+])
 
-def respond(
+def detect_subject(message):
+    """Determine which prompt template to use based on the message"""
+    message_lower = message.lower()
+    
+    math_keywords = ['math', 'solve', 'calculate', 'equation', 'formula', 'algebra', 'geometry', 'calculus', 'derivative', 'integral']
+    research_keywords = ['research', 'source', 'citation', 'bibliography', 'reference', 'academic', 'paper', 'essay', 'thesis']
+    study_keywords = ['study', 'memorize', 'exam', 'test', 'quiz', 'review', 'learn', 'remember']
+    
+    if any(keyword in message_lower for keyword in math_keywords):
+        return math_template, "🧮 Math Mode"
+    elif any(keyword in message_lower for keyword in research_keywords):
+        return research_template, "🔍 Research Mode"
+    elif any(keyword in message_lower for keyword in study_keywords):
+        return study_template, "📚 Study Mode"
+    else:
+        return general_template, "🎓 General Mode"
+
+def respond_with_langchain(
     message,
     history: list[tuple[str, str]],
     system_message,
@@ -15,40 +79,57 @@ def respond(
     temperature,
     top_p,
 ):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
+    # Select the appropriate template
+    template, mode = detect_subject(message)
+    
+    # Format the prompt
+    formatted_prompt = template.format_messages(question=message)
+    
+    # Get response from LangChain
+    try:
+        # Convert to string format for the HuggingFace model
+        prompt_text = f"{formatted_prompt[0].content}\n\nHuman: {formatted_prompt[1].content}\n\nAssistant:"
+        
+        response = llm.invoke(prompt_text)
+        
+        # Add mode indicator to response
+        full_response = f"*{mode}*\n\n{response}"
+        
+        # Yield the response (for streaming effect)
+        yield full_response
+        
+    except Exception as e:
+        yield f"Sorry, I encountered an error: {str(e)}"
 
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# Create the Gradio interface
 demo = gr.ChatInterface(
-    respond,
+    respond_with_langchain,
+    title="🎓 EduBot - AI Learning Assistant",
+    description="""
+    Your personal AI tutor powered by LangChain! I automatically adapt my teaching style based on your question:
+    
+    • 🧮 **Math Mode** - Step-by-step problem solving
+    • 🔍 **Research Mode** - Source finding and citation help  
+    • 📚 **Study Mode** - Learning strategies and test prep
+    • 🎓 **General Mode** - All-around educational support
+    
+    Try asking me anything about learning!
+    """,
+    examples=[
+        "Solve the equation 3x + 7 = 22 step by step",
+        "How do I find reliable sources for my history paper?",
+        "What's the best way to study for a biology exam?",
+        "Explain the Pythagorean theorem with an example",
+        "How do I cite a website in MLA format?"
+    ],
     additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
+        gr.Textbox(
+            value="You are EduBot, an expert AI learning assistant powered by LangChain prompt templates.",
+            label="System message",
+            visible=False  
+        ),
+        gr.Slider(minimum=1, maximum=1024, value=512, step=1, label="Max new tokens"),
+        gr.Slider(minimum=0.1, maximum=2.0, value=0.7, step=0.1, label="Temperature"),
         gr.Slider(
             minimum=0.1,
             maximum=1.0,
@@ -59,6 +140,5 @@ demo = gr.ChatInterface(
     ],
 )
 
-
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()