File size: 7,148 Bytes
5938c8c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 |
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(docs):
text_loader = TextLoader(docs) # TextLoader๋ฅผ ์ฌ์ฉํด ํ
์คํธ ํ์ผ์ ๋ก๋ํฉ๋๋ค.
text = text_loader.load() # ํ
์คํธ๋ฅผ ์ถ์ถํฉ๋๋ค.
return text # ์ถ์ถํ ํ
์คํธ๋ฅผ ๋ฐํํฉ๋๋ค.
def get_csv_file(docs):
csv_loader = CSVLoader(docs) # CSVLoader๋ฅผ ์ฌ์ฉํด CSV ํ์ผ์ ๋ก๋ํฉ๋๋ค.
csv_data = csv_loader.load() # CSV ๋ฐ์ดํฐ๋ฅผ ์ถ์ถํฉ๋๋ค.
text_column = csv_data['your_text_column'] # ํน์ ํ
์คํธ ์ปฌ๋ผ์ ์ ํํฉ๋๋ค.
return text_column.tolist() # ํ
์คํธ ์ปฌ๋ผ์ ๋ฆฌ์คํธ๋ก ๋ฐํํฉ๋๋ค.
def get_json_file(docs):
json_loader = JSONLoader(docs) # JSONLoader๋ฅผ ์ฌ์ฉํด JSON ํ์ผ์ ๋ก๋ํฉ๋๋ค.
json_data = json_loader.load() # JSON ๋ฐ์ดํฐ๋ฅผ ์ถ์ถํฉ๋๋ค.
text_field = json_data['your_text_field'] # ํน์ ํ
์คํธ ํ๋๋ฅผ ์ ํํฉ๋๋ค.
if isinstance(text_field, list):
return text_field # ํ
์คํธ ํ๋๊ฐ ๋ฆฌ์คํธ์ธ ๊ฒฝ์ฐ ๋ฐ๋ก ๋ฐํํฉ๋๋ค.
else:
return [text_field] # ํ
์คํธ ํ๋๊ฐ ๋จ์ผ ๊ฐ์ธ ๊ฒฝ์ฐ ๋ฆฌ์คํธ๋ก ๋ณํํ์ฌ ๋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
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:
st.session_state.conversation = None
if "chat_history" not in st.session_state:
st.session_state.chat_history = None
st.header("Chat with multiple Files :")
user_question = st.text_input("Ask a question about your documents:")
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 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()
|