Create app.py

app.py

@@ -0,0 +1,56 @@
+import streamlit as st
+from langchain.document_loaders import PyPDFLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.vectorstores import Chroma
+from langchain.embeddings import HuggingFaceEmbeddings
+from transformers import pipeline
+
+# Page setup
+st.title("Simple Q&A Assistant")
+
+# Load and process PDF
+@st.cache_resource
+def initialize_system():
+    # Load PDF
+    data = PyPDFLoader("ai_buddy.pdf").load()
+    
+    # Split into chunks
+    splitter = RecursiveCharacterTextSplitter(chunk_size=750, chunk_overlap=150)
+    splits = splitter.split_documents(data)
+    
+    # Create embeddings and vector store
+    embeddings = HuggingFaceEmbeddings(
+        model_name="sentence-transformers/all-MiniLM-L6-v2"
+    )
+    vector_db = Chroma.from_documents(documents=splits, embedding=embeddings)
+    
+    # Setup QA pipeline
+    qa_model = pipeline("question-answering", model="deepset/roberta-base-squad2")
+    
+    return vector_db, qa_model
+
+# Initialize the system
+if 'vector_db' not in st.session_state:
+    st.session_state.vector_db, st.session_state.qa_model = initialize_system()
+
+# Function to answer questions
+def get_answer(question):
+    # Get relevant documents
+    docs = st.session_state.vector_db.as_retriever().get_relevant_documents(question)
+    
+    if not docs:
+        return "Sorry, I couldn't find any relevant information."
+    
+    # Combine document contents
+    context = " ".join([doc.page_content for doc in docs])
+    
+    # Get answer
+    response = st.session_state.qa_model(question=question, context=context)
+    return response['answer']
+
+# Simple input/output interface
+question = st.text_input("Ask your question:")
+if question:
+    with st.spinner("Finding answer..."):
+        answer = get_answer(question)
+        st.write("Answer:", answer)