Spaces:
Sleeping
Sleeping
import streamlit as st | |
from dotenv import load_dotenv | |
from PyPDF2 import PdfReader | |
from langchain.text_splitter import CharacterTextSplitter, RecursiveCharacterTextSplitter | |
from langchain.embeddings import OpenAIEmbeddings, HuggingFaceInstructEmbeddings | |
from langchain.vectorstores import FAISS, Chroma | |
from langchain.embeddings import HuggingFaceEmbeddings # General embeddings from HuggingFace models. | |
from langchain.chat_models import ChatOpenAI | |
from langchain.memory import ConversationBufferMemory | |
from langchain.chains import ConversationalRetrievalChain | |
from htmlTemplates import css, bot_template, user_template | |
from langchain.llms import HuggingFaceHub, LlamaCpp, CTransformers # For loading transformer models. | |
from langchain.document_loaders import PyPDFLoader, TextLoader, JSONLoader, CSVLoader | |
import tempfile # ์์ ํ์ผ์ ์์ฑํ๊ธฐ ์ํ ๋ผ์ด๋ธ๋ฌ๋ฆฌ์ ๋๋ค. | |
import os | |
# PDF ๋ฌธ์๋ก๋ถํฐ ํ ์คํธ๋ฅผ ์ถ์ถํ๋ ํจ์์ ๋๋ค. | |
def get_pdf_text(pdf_docs): | |
temp_dir = tempfile.TemporaryDirectory() # ์์ ๋๋ ํ ๋ฆฌ๋ฅผ ์์ฑํฉ๋๋ค. | |
temp_filepath = os.path.join(temp_dir.name, pdf_docs.name) # ์์ ํ์ผ ๊ฒฝ๋ก๋ฅผ ์์ฑํฉ๋๋ค. | |
with open(temp_filepath, "wb") as f: # ์์ ํ์ผ์ ๋ฐ์ด๋๋ฆฌ ์ฐ๊ธฐ ๋ชจ๋๋ก ์ฝ๋๋ค. | |
f.write(pdf_docs.getvalue()) # PDF ๋ฌธ์์ ๋ด์ฉ์ ์์ ํ์ผ์ ์๋๋ค. | |
pdf_loader = PyPDFLoader(temp_filepath) # PyPDFLoader๋ฅผ ์ฌ์ฉํด PDF๋ฅผ ๋ก๋ํฉ๋๋ค. | |
pdf_doc = pdf_loader.load() # ํ ์คํธ๋ฅผ ์ถ์ถํฉ๋๋ค. | |
return pdf_doc # ์ถ์ถํ ํ ์คํธ๋ฅผ ๋ฐํํฉ๋๋ค. | |
# ๊ณผ์ | |
# ์๋ ํ ์คํธ ์ถ์ถ ํจ์๋ฅผ ์์ฑ | |
def get_text_file(text_docs): | |
temp_dir2 = tempfile.TemporaryDirectory() # ์์ ๋๋ ํ ๋ฆฌ๋ฅผ ์์ฑํฉ๋๋ค. | |
temp_filepath2 = os.path.join(temp_dir2.name, text_docs.name) # ์์ ํ์ผ ๊ฒฝ๋ก๋ฅผ ์์ฑํฉ๋๋ค. | |
with open(temp_filepath2, "wb") as f: # ์์ ํ์ผ์ ๋ฐ์ด๋๋ฆฌ ์ฐ๊ธฐ ๋ชจ๋๋ก ์ฝ๋๋ค. | |
f.write(text_docs.getvalue()) # text ๋ฌธ์์ ๋ด์ฉ์ ์์ ํ์ผ์ ์๋๋ค. | |
text_loader = TextLoader(temp_filepath2) # TextLoader๋ฅผ ์ฌ์ฉํด text๋ฅผ ๋ก๋ํฉ๋๋ค. | |
text_doc = text_loader.load() # ํ ์คํธ๋ฅผ ์ถ์ถํฉ๋๋ค. | |
return text_doc # ์ถ์ถํ ํ ์คํธ๋ฅผ ๋ฐํํฉ๋๋ค. | |
def get_csv_file(csv_docs): | |
temp_dir3 = tempfile.TemporaryDirectory() # ์์ ๋๋ ํ ๋ฆฌ๋ฅผ ์์ฑํฉ๋๋ค. | |
temp_filepath3 = os.path.join(temp_dir3.name, csv_docs.name) # ์์ ํ์ผ ๊ฒฝ๋ก๋ฅผ ์์ฑํฉ๋๋ค. | |
with open(temp_filepath3, "wb") as f: # ์์ ํ์ผ์ ๋ฐ์ด๋๋ฆฌ ์ฐ๊ธฐ ๋ชจ๋๋ก ์ฝ๋๋ค. | |
f.write(csv_docs.getvalue()) # csv ๋ฌธ์์ ๋ด์ฉ์ ์์ ํ์ผ์ ์๋๋ค. | |
csv_loader = CSVLoader(temp_filepath3) # CSVLoader๋ฅผ ์ฌ์ฉํด csv๋ฅผ ๋ก๋ํฉ๋๋ค. | |
csv_doc = csv_loader.load() # ํ ์คํธ๋ฅผ ์ถ์ถํฉ๋๋ค. | |
return csv_doc # ์ถ์ถํ ํ ์คํธ๋ฅผ ๋ฐํํฉ๋๋ค. | |
def get_json_file(json_docs): | |
temp_dir4 = tempfile.TemporaryDirectory() # ์์ ๋๋ ํ ๋ฆฌ๋ฅผ ์์ฑํฉ๋๋ค. | |
temp_filepath4 = os.path.join(temp_dir4.name, json_docs.name) # ์์ ํ์ผ ๊ฒฝ๋ก๋ฅผ ์์ฑํฉ๋๋ค. | |
with open(temp_filepath4, "wb") as f: # ์์ ํ์ผ์ ๋ฐ์ด๋๋ฆฌ ์ฐ๊ธฐ ๋ชจ๋๋ก ์ฝ๋๋ค. | |
f.write(json_docs.getvalue()) # json ๋ฌธ์์ ๋ด์ฉ์ ์์ ํ์ผ์ ์๋๋ค. | |
json_loader = JSONLoader(temp_filepath4, jq_schema=<your_json_schema>) # JSONLoader๋ฅผ ์ฌ์ฉํด json๋ฅผ ๋ก๋ํฉ๋๋ค. | |
json_doc = json_loader.load() # ํ ์คํธ๋ฅผ ์ถ์ถํฉ๋๋ค. | |
return json_doc # ์ถ์ถํ ํ ์คํธ๋ฅผ ๋ฐํํฉ๋๋ค. | |
# ๋ฌธ์๋ค์ ์ฒ๋ฆฌํ์ฌ ํ ์คํธ ์ฒญํฌ๋ก ๋๋๋ ํจ์์ ๋๋ค. | |
def get_text_chunks(documents): | |
text_splitter = RecursiveCharacterTextSplitter( | |
chunk_size=1000, # ์ฒญํฌ์ ํฌ๊ธฐ๋ฅผ ์ง์ ํฉ๋๋ค. | |
chunk_overlap=200, # ์ฒญํฌ ์ฌ์ด์ ์ค๋ณต์ ์ง์ ํฉ๋๋ค. | |
length_function=len # ํ ์คํธ์ ๊ธธ์ด๋ฅผ ์ธก์ ํ๋ ํจ์๋ฅผ ์ง์ ํฉ๋๋ค. | |
) | |
documents = text_splitter.split_documents(documents) # ๋ฌธ์๋ค์ ์ฒญํฌ๋ก ๋๋๋๋ค | |
if not documents: | |
raise ValueError("์ ๋ก๋๋ ๋ฌธ์๊ฐ ์๊ฑฐ๋ ๋ชจ๋ ๋ฌธ์๊ฐ ๋น์ด ์์ต๋๋ค.") | |
return documents # ๋๋ ์ฒญํฌ๋ฅผ ๋ฐํํฉ๋๋ค. | |
# ํ ์คํธ ์ฒญํฌ๋ค๋ก๋ถํฐ ๋ฒกํฐ ์คํ ์ด๋ฅผ ์์ฑํ๋ ํจ์์ ๋๋ค. | |
def get_vectorstore(text_chunks): | |
# OpenAI ์๋ฒ ๋ฉ ๋ชจ๋ธ์ ๋ก๋ํฉ๋๋ค. (Embedding models - Ada v2) | |
embeddings = OpenAIEmbeddings() | |
vectorstore = FAISS.from_documents(text_chunks, embeddings) # FAISS ๋ฒกํฐ ์คํ ์ด๋ฅผ ์์ฑํฉ๋๋ค. | |
return vectorstore # ์์ฑ๋ ๋ฒกํฐ ์คํ ์ด๋ฅผ ๋ฐํํฉ๋๋ค. | |
def get_conversation_chain(vectorstore): | |
gpt_model_name = 'gpt-3.5-turbo' | |
llm = ChatOpenAI(model_name = gpt_model_name) #gpt-3.5 ๋ชจ๋ธ ๋ก๋ | |
# ๋ํ ๊ธฐ๋ก์ ์ ์ฅํ๊ธฐ ์ํ ๋ฉ๋ชจ๋ฆฌ๋ฅผ ์์ฑํฉ๋๋ค. | |
memory = ConversationBufferMemory( | |
memory_key='chat_history', return_messages=True) | |
# ๋ํ ๊ฒ์ ์ฒด์ธ์ ์์ฑํฉ๋๋ค. | |
conversation_chain = ConversationalRetrievalChain.from_llm( | |
llm=llm, | |
retriever=vectorstore.as_retriever(), | |
memory=memory | |
) | |
return conversation_chain | |
# ์ฌ์ฉ์ ์ ๋ ฅ์ ์ฒ๋ฆฌํ๋ ํจ์์ ๋๋ค. | |
def handle_userinput(user_question): | |
# ๋ํ ์ฒด์ธ์ ์ฌ์ฉํ์ฌ ์ฌ์ฉ์ ์ง๋ฌธ์ ๋ํ ์๋ต์ ์์ฑํฉ๋๋ค. | |
response = st.session_state.conversation({'question': user_question}) | |
# ๋ํ ๊ธฐ๋ก์ ์ ์ฅํฉ๋๋ค. | |
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="Chat with multiple Files", | |
page_icon=":books:") | |
st.write(css, unsafe_allow_html=True) | |
if "conversation" not in st.session_state or st.session_state.conversation is None: | |
st.session_state.conversation = None | |
st.session_state.chat_history = None | |
st.header("Chat with multiple Files :") | |
user_question = st.text_input("Ask a question about your documents:") | |
# "Send" ๋ฒํผ | |
if st.button("Send"): | |
if user_question: | |
handle_userinput(user_question) | |
if user_question: | |
handle_userinput(user_question) | |
with st.sidebar: | |
openai_key = st.text_input("Paste your OpenAI API key (sk-...)") | |
if openai_key: | |
os.environ["OPENAI_API_KEY"] = openai_key | |
st.subheader("Your documents") | |
docs = st.file_uploader( | |
"Upload your PDFs, TEXTfiles, CSVfiles, JSONfiles here and click on 'Process'", accept_multiple_files=True) | |
if st.button("Process"): | |
with st.spinner("Processing"): | |
# get pdf text | |
doc_list = [] | |
for file in docs: | |
print('file - type : ', file.type) | |
if file.type == 'text/plain': | |
# file is .txt | |
doc_list.extend(get_text_file(file)) | |
elif file.type in ['application/octet-stream', 'application/pdf']: | |
# file is .pdf | |
doc_list.extend(get_pdf_text(file)) | |
elif file.type == 'text/csv': | |
# file is .csv | |
doc_list.extend(get_csv_file(file)) | |
elif file.type == 'application/json': | |
# file is .json | |
doc_list.extend(get_json_file(file)) | |
# get the text chunks | |
text_chunks = get_text_chunks(doc_list) | |
# create vector store | |
vectorstore = get_vectorstore(text_chunks) | |
# create conversation chain | |
st.session_state.conversation = get_conversation_chain( | |
vectorstore) | |
if __name__ == '__main__': | |
main() |