Update app.py

app.py

@@ -1,28 +1,38 @@
+import os
+from typing import List, Optional
+
+from langchain_community.document_loaders import TextLoader
 from langchain_community.embeddings import HuggingFaceEmbeddings
 from langchain_community.vectorstores import Chroma
 from langchain.text_splitter import RecursiveCharacterTextSplitter
-from langchain.document_loaders import TextLoader
 from langchain.chains import RetrievalQA
 from langchain.llms.base import LLM
-from typing import List, Optional
 from groq import Groq
 import gradio as gr
-import os
 
-# Load and split the document
+# ✅ Create sample file if missing
+if not os.path.exists("sample_readme.txt"):
+    with open("sample_readme.txt", "w") as f:
+        f.write(
+            "# Sample Project\n\nThis project demonstrates an example of a LangChain-powered RAG pipeline. "
+            "It uses FAISS for vector search and a GROQ-hosted LLaMA3 model for response generation.\n\n"
+            "## Features\n- Document embedding\n- Vector similarity search\n- LLM-based QA over documents"
+        )
+
+# Load & split
 loader = TextLoader("sample_readme.txt")
 documents = loader.load()
 text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
 docs = text_splitter.split_documents(documents)
 
-# Generate embeddings and persist them
+# Embed & store vectors
 embedding = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
 vectorstore = Chroma.from_documents(docs, embedding, persist_directory="rag_chroma_groq")
 
-# Define custom Groq LLM wrapper
+# Groq LLM
 class GroqLLM(LLM):
     model: str = "llama3-8b-8192"
-    api_key: str = os.environ.get("GROQ_API_KEY")  # Store securely in Hugging Face secrets
+    api_key: str = os.environ.get("GROQ_API_KEY")
     temperature: float = 0.0
 
     def _call(self, prompt: str, stop: Optional[List[str]] = None) -> str:
@@ -42,7 +52,7 @@ class GroqLLM(LLM):
     def _llm_type(self) -> str:
         return "groq-llm"
 
-# Initialize chain
+# QA setup
 retriever = vectorstore.as_retriever()
 groq_llm = GroqLLM()
 qa_chain = RetrievalQA.from_chain_type(
@@ -51,17 +61,20 @@ qa_chain = RetrievalQA.from_chain_type(
     return_source_documents=True
 )
 
-# Gradio Interface
+# Gradio UI
 def ask_question(query):
     response = qa_chain(query)
     answer = response["result"]
-    sources = "\n".join([doc.metadata.get("source", "unknown") for doc in response["source_documents"]])
-    return f"**Answer:** {answer}\n\n**Sources:**\n{sources}"
+    sources = "\n".join(
+        [doc.metadata.get("source", "sample_readme.txt") for doc in response["source_documents"]]
+    )
+    return f"### Answer:\n{answer}\n\n### Sources:\n{sources}"
 
 gr.Interface(
     fn=ask_question,
-    inputs=gr.Textbox(label="Ask a question"),
+    inputs=gr.Textbox(label="Ask something about the README"),
     outputs=gr.Markdown(),
-    title="RAG Chatbot with Groq LLaMA3",
-    description="Ask questions based on the content of a sample README document using LangChain + Chroma + Groq-hosted LLaMA3."
+    title="📄 RAG Chatbot with Groq LLaMA3",
+    description="Ask questions about a README file using a LangChain + Groq LLaMA3-powered chatbot.",
+    theme="soft"
 ).launch()