vector_database.py

from langchain_community.document_loaders import PDFPlumberLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_ollama import OllamaEmbeddings
from langchain_community.vectorstores import FAISS

#step1: upload and load raw PDF file

pdfs_directory= 'pdfs/'

def upload_pdf(file):
    with open(pdfs_directory + file.name, "wb") as f:
        f.write(file.getbuffer())

def load_pdf(file_path):
    loader = PDFPlumberLoader(file_path)
    documents = loader.load()
    return documents

file_path="eng.pdf"
documents = load_pdf(file_path)
# print("Pdf pages:",len(documents))  

#step2:create chunks

def create_chunks(documents):
    text_splitter=RecursiveCharacterTextSplitter(
       chunk_size=1000,
       chunk_overlap=200,
       add_start_index=True
    )

    text_chunks= text_splitter.split_documents(documents)
    return text_chunks

text_chunks= create_chunks(documents)
# print("chunks count:",len(text_chunks))  


#step3:setup embeddings model (using DeepSeek with Ollama)

ollama_model_name="deepseek-r1:8b"
def get_embeddings_model(ollama_model_name):
    embeddings = OllamaEmbeddings(model=ollama_model_name)
    return embeddings


#step4: Index Documents **Store embeddings in FAISS (vector store)
FAISS_DB_path = "vectorstore/db_faiss"
faiss_db = FAISS.from_documents(text_chunks, get_embeddings_model(ollama_model_name))
faiss_db.save_local(FAISS_DB_path)