rag basic assignment

7_Assignment_RagBasicPdf/.env

+OPENAI_API_KEY=your-api-key-here
+

7_Assignment_RagBasicPdf/app.py

+import streamlit as st
+from utils import load_pdf_chunks, build_faiss_index, get_answer
+import os
+from dotenv import load_dotenv
+
+load_dotenv()
+st.set_page_config(page_title="PDF Q&A Bot", layout="centered")
+st.title("📄 Ask Questions to Your PDF")
+
+uploaded_file = st.file_uploader("Upload a PDF", type="pdf")
+
+if uploaded_file:
+    with open("temp.pdf", "wb") as f:
+        f.write(uploaded_file.read())
+    st.success("PDF uploaded successfully!")
+
+    with st.spinner("Reading and indexing the document..."):
+        chunks = load_pdf_chunks("temp.pdf")
+        faiss_index = build_faiss_index(chunks)
+
+    st.success("Document is ready for questions!")
+
+    query = st.text_input("Ask a question:")
+    if query:
+        with st.spinner("Thinking..."):
+            answer = get_answer(faiss_index, query)
+        st.markdown(f"**Answer:** {answer}")
+

7_Assignment_RagBasicPdf/requirements.txt

+streamlit
+langchain
+openai
+faiss-cpu
+python-dotenv
+PyPDF2
+

7_Assignment_RagBasicPdf/utils.py

+import os
+from langchain_community.document_loaders import PyPDFLoader
+#from langchain.document_loaders import PyPDFLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.vectorstores import FAISS
+from langchain.embeddings.openai import OpenAIEmbeddings
+from langchain.chains.question_answering import load_qa_chain
+from langchain.llms import OpenAI
+
+def load_pdf_chunks(pdf_path):
+    loader = PyPDFLoader(pdf_path)
+    docs = loader.load()
+    splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
+    return splitter.split_documents(docs)
+
+def build_faiss_index(chunks):
+    embeddings = OpenAIEmbeddings()
+    return FAISS.from_documents(chunks, embeddings)
+
+def get_answer(faiss_index, query):
+    docs = faiss_index.similarity_search(query, k=3)
+    llm = OpenAI(temperature=0)
+    chain = load_qa_chain(llm, chain_type="stuff")
+    return chain.run(input_documents=docs, question=query)
+