adding RAG feature

app.py

@@ -2,6 +2,8 @@ from langchain.document_loaders import HuggingFaceDatasetLoader
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 from langchain.embeddings import HuggingFaceEmbeddings
 from langchain.vectorstores import FAISS
+
+from transformers import AutoTokenizer, pipeline
 import gradio as gr
 
 
@@ -24,12 +26,24 @@ embeddings = HuggingFaceEmbeddings(
 db = FAISS.from_documents(docs, embeddings)
 
 # Set up retrievers
-retriever = db.as_retriever()
+retriever = db.as_retriever(search_kwargs={"k": 4})
+
+# Load the tokenizer associated with the specified model
+tokenizer = AutoTokenizer.from_pretrained("Intel/dynamic_tinybert", padding=True, truncation=True, max_length=512)
 
+# Define a question-answering pipeline using the model and tokenizer
+question_answerer = pipeline(
+    "question-answering",
+    model="Intel/dynamic_tinybert",
+    tokenizer=tokenizer,
+    return_tensors='pt'
+)
 
-def generate(input):
-    docs = retriever.get_relevant_documents(input)
-    return docs[0].page_content
+def generate(question):
+    docs = retriever.get_relevant_documents(question)
+    context = docs[0].page_content
+    squad_ex = question_answerer(question=question, context=context)
+    return squad_ex['answer']
 
 
 def respond(message, chat_history):