rag.py

import pandas as pd
# from langchain.schema import Document
from langchain_core.documents import Document
# from langchain.embeddings import HuggingFaceEmbeddings
from langchain_huggingface import HuggingFaceEmbeddings
# from langchain.vectorstores import FAISS
from langchain_community.vectorstores import FAISS


def load_student_documents(csv_path: str):
    df = pd.read_csv(csv_path)
    documents = []
    for _, row in df.iterrows():
        content = (
    f"Student record. "
    f"Register number {row['Register Number']}. "
    f"Name {row['Name']}. "
    f"Email {row['EmailID']}. "
    f"Department {row['Department']}. "
    f"Year {row['Year of Study']}. "
    f"Hobby {row['Hobby']}. "
    f"Study level {row['Study Level']}. "
    f"Sleeptime {row['Sleeptime']}."
)

        documents.append(
            Document(
                page_content=content.strip(),
                metadata={"register_number": row["Register Number"]}
            )
        )
    return documents

def create_vectorstore(documents):
    embeddings = HuggingFaceEmbeddings(
        model_name="sentence-transformers/all-MiniLM-L6-v2"
    )
    vectorstore = FAISS.from_documents(
        documents=documents,
        embedding=embeddings
    )
    return vectorstore

def get_retriever(vectorstore):
    return vectorstore.as_retriever(
    search_type="mmr",
    search_kwargs={"k": 6, "lambda_mult": 0.7}
)