|
|
import streamlit as st |
|
|
from dotenv import load_dotenv |
|
|
|
|
|
|
|
|
|
|
|
from langchain.memory import ConversationBufferMemory |
|
|
from langchain.chains import ConversationalRetrievalChain |
|
|
from htmlTemplates import css, bot_template, user_template |
|
|
|
|
|
|
|
|
|
|
|
from langchain_text_splitters import CharacterTextSplitter, RecursiveCharacterTextSplitter |
|
|
|
|
|
|
|
|
from langchain_community.vectorstores import FAISS |
|
|
from langchain_community.embeddings import HuggingFaceEmbeddings |
|
|
|
|
|
|
|
|
from langchain_community.document_loaders.pdf import PyPDFLoader |
|
|
from langchain_community.document_loaders.text import TextLoader |
|
|
from langchain_community.document_loaders.csv_loader import CSVLoader |
|
|
from langchain_community.document_loaders.json_loader import JSONLoader |
|
|
import tempfile |
|
|
import os |
|
|
import json |
|
|
from langchain.docstore.document import Document |
|
|
from langchain_groq import ChatGroq |
|
|
|
|
|
|
|
|
def get_pdf_text(pdf_docs): |
|
|
temp_dir = tempfile.TemporaryDirectory() |
|
|
temp_filepath = os.path.join(temp_dir.name, pdf_docs.name) |
|
|
with open(temp_filepath, "wb") as f: |
|
|
f.write(pdf_docs.getvalue()) |
|
|
pdf_loader = PyPDFLoader(temp_filepath) |
|
|
pdf_doc = pdf_loader.load() |
|
|
return pdf_doc |
|
|
|
|
|
|
|
|
def get_text_file(docs): |
|
|
|
|
|
temp_dir = tempfile.TemporaryDirectory() |
|
|
temp_filepath = os.path.join(temp_dir.name, docs.name) |
|
|
with open(temp_filepath, "wb") as f: |
|
|
f.write(docs.getvalue()) |
|
|
|
|
|
text_loader = TextLoader(temp_filepath, encoding="utf-8") |
|
|
text_doc = text_loader.load() |
|
|
return text_doc |
|
|
|
|
|
|
|
|
def get_csv_file(docs): |
|
|
|
|
|
temp_dir = tempfile.TemporaryDirectory() |
|
|
temp_filepath = os.path.join(temp_dir.name, docs.name) |
|
|
with open(temp_filepath, "wb") as f: |
|
|
f.write(docs.getvalue()) |
|
|
|
|
|
csv_loader = CSVLoader(temp_filepath, encoding="utf-8") |
|
|
csv_doc = csv_loader.load() |
|
|
return csv_doc |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def get_json_file(file) -> list[Document]: |
|
|
|
|
|
raw = file.getvalue().decode("utf-8", errors="ignore") |
|
|
data = json.loads(raw) |
|
|
|
|
|
docs = [] |
|
|
|
|
|
|
|
|
|
|
|
def add_doc(x): |
|
|
docs.append(Document(page_content=json.dumps(x, ensure_ascii=False))) |
|
|
|
|
|
if isinstance(data, dict) and "scans" in data and isinstance(data["scans"], list): |
|
|
for s in data["scans"]: |
|
|
rels = s.get("relationships", []) |
|
|
if isinstance(rels, list) and rels: |
|
|
for r in rels: |
|
|
add_doc(r) |
|
|
if not docs: |
|
|
add_doc(data) |
|
|
elif isinstance(data, list): |
|
|
for item in data: |
|
|
add_doc(item) |
|
|
else: |
|
|
add_doc(data) |
|
|
|
|
|
return docs |
|
|
|
|
|
|
|
|
def get_text_chunks(documents): |
|
|
text_splitter = RecursiveCharacterTextSplitter( |
|
|
chunk_size=1000, |
|
|
chunk_overlap=200, |
|
|
length_function=len |
|
|
) |
|
|
|
|
|
documents = text_splitter.split_documents(documents) |
|
|
return documents |
|
|
|
|
|
|
|
|
|
|
|
def get_vectorstore(text_chunks): |
|
|
|
|
|
embeddings = HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L12-v2', |
|
|
model_kwargs={'device': 'cpu'}) |
|
|
vectorstore = FAISS.from_documents(text_chunks, embeddings) |
|
|
return vectorstore |
|
|
|
|
|
|
|
|
def get_conversation_chain(vectorstore): |
|
|
|
|
|
llm = ChatGroq( |
|
|
groq_api_key=os.environ.get("GROQ_API_KEY"), |
|
|
model_name="llama-3.1-8b-instant", |
|
|
temperature=0.75, |
|
|
max_tokens=512 |
|
|
) |
|
|
|
|
|
memory = ConversationBufferMemory( |
|
|
memory_key="chat_history", |
|
|
return_messages=True |
|
|
) |
|
|
retriever = vectorstore.as_retriever(search_kwargs={"k": 3}) |
|
|
|
|
|
conversation_chain = ConversationalRetrievalChain.from_llm( |
|
|
llm=llm, |
|
|
retriever=retriever, |
|
|
memory=memory, |
|
|
) |
|
|
return conversation_chain |
|
|
|
|
|
|
|
|
def handle_userinput(user_question): |
|
|
print('user_question => ', user_question) |
|
|
|
|
|
response = st.session_state.conversation({'question': user_question}) |
|
|
|
|
|
st.session_state.chat_history = response['chat_history'] |
|
|
|
|
|
for i, message in enumerate(st.session_state.chat_history): |
|
|
if i % 2 == 0: |
|
|
st.write(user_template.replace( |
|
|
"{{MSG}}", message.content), unsafe_allow_html=True) |
|
|
else: |
|
|
st.write(bot_template.replace( |
|
|
"{{MSG}}", message.content), unsafe_allow_html=True) |
|
|
|
|
|
|
|
|
def main(): |
|
|
load_dotenv() |
|
|
st.set_page_config(page_title="Basic_RAG_AI_Chatbot_with_Llama", |
|
|
page_icon=":books:") |
|
|
st.write(css, unsafe_allow_html=True) |
|
|
|
|
|
if "conversation" not in st.session_state: |
|
|
st.session_state.conversation = None |
|
|
if "chat_history" not in st.session_state: |
|
|
st.session_state.chat_history = None |
|
|
|
|
|
st.header("Basic_RAG_AI_Chatbot_with_Llama3 :books:") |
|
|
user_question = st.text_input("Ask a question about your documents:") |
|
|
if user_question: |
|
|
handle_userinput(user_question) |
|
|
|
|
|
with st.sidebar: |
|
|
st.subheader("Your documents") |
|
|
docs = st.file_uploader( |
|
|
"Upload your Files here and click on 'Process'", accept_multiple_files=True) |
|
|
if st.button("Process[PDF]"): |
|
|
with st.spinner("Processing"): |
|
|
|
|
|
doc_list = [] |
|
|
for file in docs: |
|
|
print('file - type : ', file.type) |
|
|
if file.type in ['application/octet-stream', 'application/pdf']: |
|
|
|
|
|
doc_list.extend(get_pdf_text(file)) |
|
|
else: |
|
|
st.error("PDF ํ์ผ์ด ์๋๋๋ค.") |
|
|
if not doc_list: |
|
|
st.error("์ฒ๋ฆฌ ๊ฐ๋ฅํ ๋ฌธ์๋ฅผ ์ฐพ์ง ๋ชปํ์ต๋๋ค.") |
|
|
st.stop() |
|
|
|
|
|
text_chunks = get_text_chunks(doc_list) |
|
|
vectorstore = get_vectorstore(text_chunks) |
|
|
st.session_state.conversation = get_conversation_chain(vectorstore) |
|
|
|
|
|
|
|
|
if st.button("Process[TXT]"): |
|
|
with st.spinner("Processing"): |
|
|
doc_list = [] |
|
|
for file in docs: |
|
|
if file.type == 'text/plain': |
|
|
doc_list.extend(get_text_file(file)) |
|
|
else: |
|
|
st.error("TXT ํ์ผ์ด ์๋๋๋ค.") |
|
|
|
|
|
if not doc_list: |
|
|
st.error("์ฒ๋ฆฌ ๊ฐ๋ฅํ TXT ๋ฌธ์๋ฅผ ์ฐพ์ง ๋ชปํ์ต๋๋ค.") |
|
|
st.stop() |
|
|
|
|
|
text_chunks = get_text_chunks(doc_list) |
|
|
vectorstore = get_vectorstore(text_chunks) |
|
|
st.session_state.conversation = get_conversation_chain(vectorstore) |
|
|
|
|
|
if st.button("Process[CSV]"): |
|
|
with st.spinner("Processing"): |
|
|
doc_list = [] |
|
|
for file in docs: |
|
|
if file.type == 'text/csv': |
|
|
doc_list.extend(get_csv_file(file)) |
|
|
else: |
|
|
st.error("CSV ํ์ผ์ด ์๋๋๋ค.") |
|
|
|
|
|
if not doc_list: |
|
|
st.error("์ฒ๋ฆฌ ๊ฐ๋ฅํ CSV ๋ฌธ์๋ฅผ ์ฐพ์ง ๋ชปํ์ต๋๋ค.") |
|
|
st.stop() |
|
|
|
|
|
text_chunks = get_text_chunks(doc_list) |
|
|
vectorstore = get_vectorstore(text_chunks) |
|
|
st.session_state.conversation = get_conversation_chain(vectorstore) |
|
|
|
|
|
if st.button("Process[JSON]"): |
|
|
with st.spinner("Processing"): |
|
|
|
|
|
doc_list = [] |
|
|
for file in docs: |
|
|
print('file - type : ', file.type) |
|
|
if file.type == 'application/json': |
|
|
|
|
|
doc_list.extend(get_json_file(file)) |
|
|
else: |
|
|
st.error("JSON ํ์ผ์ด ์๋๋๋ค.") |
|
|
if not doc_list: |
|
|
st.error("์ฒ๋ฆฌ ๊ฐ๋ฅํ ๋ฌธ์๋ฅผ ์ฐพ์ง ๋ชปํ์ต๋๋ค.") |
|
|
st.stop() |
|
|
|
|
|
text_chunks = get_text_chunks(doc_list) |
|
|
vectorstore = get_vectorstore(text_chunks) |
|
|
st.session_state.conversation = get_conversation_chain(vectorstore) |
|
|
|
|
|
|
|
|
if __name__ == '__main__': |
|
|
main() |