File size: 17,900 Bytes
3647b28 cf1fee6 0b78609 cf1fee6 da9dd79 cf1fee6 a4c77d9 cf1fee6 a4c77d9 cf1fee6 f82b999 cf1fee6 b30ada3 cf1fee6 b30ada3 dfac116 b30ada3 cf1fee6 931dd12 917c404 cf1fee6 917c404 cf1fee6 21b887c cf1fee6 cde5754 cf1fee6 3647b28 cf1fee6 3647b28 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 |
# import streamlit as st
# from dotenv import load_dotenv
# from langchain.text_splitter import CharacterTextSplitter, RecursiveCharacterTextSplitter
# from langchain.vectorstores import FAISS
# from langchain.embeddings import HuggingFaceEmbeddings # General embeddings from HuggingFace models.
# from langchain.memory import ConversationBufferMemory
# from langchain.chains import ConversationalRetrievalChain
# from htmlTemplates import css, bot_template, user_template
# from langchain.llms import LlamaCpp
# import json
# from pathlib import Path
# from pprint import pprint
# from langchain.document_loaders import PyPDFLoader, TextLoader, JSONLoader, CSVLoader
# import tempfile # μμ νμΌμ μμ±νκΈ° μν λΌμ΄λΈλ¬λ¦¬μ
λλ€.
# import os
# from huggingface_hub import hf_hub_download # Hugging Face Hubμμ λͺ¨λΈμ λ€μ΄λ‘λνκΈ° μν ν¨μμ
λλ€.
# # PDF λ¬Έμλ‘λΆν° ν
μ€νΈλ₯Ό μΆμΆνλ ν¨μμ
λλ€.
# def get_pdf_text(pdf_docs):
# temp_dir = tempfile.TemporaryDirectory() # μμ λλ ν 리λ₯Ό μμ±ν©λλ€.
# temp_filepath = os.path.join(temp_dir.name, pdf_docs.name) # μμ νμΌ κ²½λ‘λ₯Ό μμ±ν©λλ€.
# with open(temp_filepath, "wb") as f: # μμ νμΌμ λ°μ΄λ리 μ°κΈ° λͺ¨λλ‘ μ½λλ€.
# f.write(pdf_docs.getvalue()) # PDF λ¬Έμμ λ΄μ©μ μμ νμΌμ μλλ€.
# pdf_loader = PyPDFLoader(temp_filepath) # PyPDFLoaderλ₯Ό μ¬μ©ν΄ PDFλ₯Ό λ‘λν©λλ€.
# pdf_doc = pdf_loader.load() # ν
μ€νΈλ₯Ό μΆμΆν©λλ€.
# return pdf_doc # μΆμΆν ν
μ€νΈλ₯Ό λ°νν©λλ€.
# # κ³Όμ
# # μλ ν
μ€νΈ μΆμΆ ν¨μλ₯Ό μμ±
# def get_text_file(text_docs):
# temp_dir = tempfile.TemporaryDirectory() # μμ λλ ν 리λ₯Ό μμ±ν©λλ€.
# temp_filepath = os.path.join(temp_dir.name, text_docs.name) # μμ νμΌ κ²½λ‘λ₯Ό μμ±ν©λλ€.
# with open(temp_filepath, "wb") as f: # μμ νμΌμ ν
μ€νΈ μ°κΈ° λͺ¨λλ‘ μ½λλ€.
# f.write(text_docs.getvalue()) # ν
μ€νΈ λ¬Έμμ λ΄μ©μ μμ νμΌμ μλλ€.
# text_loader = TextLoader(temp_filepath) # TextLoaderλ₯Ό μ¬μ©ν΄ ν
μ€νΈ λ¬Έμλ₯Ό λ‘λν©λλ€.
# text_doc = text_loader.load() # ν
μ€νΈλ₯Ό μΆμΆν©λλ€.
# return text_doc # μΆμΆλ ν
μ€νΈλ₯Ό λ°νν©λλ€.
# def get_csv_file(csv_docs):
# temp_dir = tempfile.TemporaryDirectory() # μμ λλ ν 리λ₯Ό μμ±ν©λλ€.
# temp_filepath = os.path.join(temp_dir.name, csv_docs.name) # μμ νμΌ κ²½λ‘λ₯Ό μμ±ν©λλ€.
# with open(temp_filepath, "wb") as f: # μμ νμΌμ λ°μ΄λ리 μ°κΈ° λͺ¨λλ‘ μ½λλ€.
# f.write(csv_docs.getvalue()) # CSV λ¬Έμμ λ΄μ©μ μμ νμΌμ μλλ€.
# csv_loader = CSVLoader(temp_filepath) # CSVLoaderλ₯Ό μ¬μ©ν΄ CSV λ¬Έμλ₯Ό λ‘λν©λλ€.
# csv_doc = csv_loader.load() # ν
μ€νΈλ₯Ό μΆμΆν©λλ€.
# return csv_doc # μΆμΆλ ν
μ€νΈλ₯Ό λ°νν©λλ€.
# def get_json_file(json_docs):
# temp_dir = tempfile.TemporaryDirectory() # μμ λλ ν 리λ₯Ό μμ±ν©λλ€.
# temp_filepath = os.path.join(temp_dir.name, json_docs.name) # μμ νμΌ κ²½λ‘λ₯Ό μμ±ν©λλ€.
# with open(temp_filepath, "wb") as f: # μμ νμΌμ ν
μ€νΈ μ°κΈ° λͺ¨λλ‘ μ½λλ€.
# f.write(json_docs.getvalue()) # JSON λ¬Έμμ λ΄μ©μ μμ νμΌμ μλλ€.
# json_loader = JSONLoader(temp_filepath) # JSONLoaderλ₯Ό μ¬μ©ν΄ JSON λ¬Έμλ₯Ό λ‘λν©λλ€.
# json_doc = json_loader.load() # ν
μ€νΈλ₯Ό μΆμΆν©λλ€.
# return json_doc # μΆμΆλ ν
μ€νΈλ₯Ό λ°νν©λλ€.
# # def get_text_file(text_docs):
# #
# # pass
# #
# # def get_csv_file(csv_docs):
# # pass
# #
# # def get_json_file(json_docs):
# #
# #
# # pass
# # λ¬Έμλ€μ μ²λ¦¬νμ¬ ν
μ€νΈ μ²ν¬λ‘ λλλ ν¨μμ
λλ€.
# def get_text_chunks(documents):
# text_splitter = RecursiveCharacterTextSplitter(
# chunk_size=1000, # μ²ν¬μ ν¬κΈ°λ₯Ό μ§μ ν©λλ€.
# chunk_overlap=200, # μ²ν¬ μ¬μ΄μ μ€λ³΅μ μ§μ ν©λλ€.
# length_function=len # ν
μ€νΈμ κΈΈμ΄λ₯Ό μΈ‘μ νλ ν¨μλ₯Ό μ§μ ν©λλ€.
# )
# documents = text_splitter.split_documents(documents) # λ¬Έμλ€μ μ²ν¬λ‘ λλλλ€.
# return documents # λλ μ²ν¬λ₯Ό λ°νν©λλ€.
# # ν
μ€νΈ μ²ν¬λ€λ‘λΆν° λ²‘ν° μ€ν μ΄λ₯Ό μμ±νλ ν¨μμ
λλ€.
# def get_vectorstore(text_chunks):
# # μνλ μλ² λ© λͺ¨λΈμ λ‘λν©λλ€.
# embeddings = HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L12-v2',
# model_kwargs={'device': 'cpu'}) # μλ² λ© λͺ¨λΈμ μ€μ ν©λλ€.
# vectorstore = FAISS.from_documents(text_chunks, embeddings) # FAISS λ²‘ν° μ€ν μ΄λ₯Ό μμ±ν©λλ€.
# return vectorstore # μμ±λ λ²‘ν° μ€ν μ΄λ₯Ό λ°νν©λλ€.
# def get_conversation_chain(vectorstore):
# model_name_or_path = 'TheBloke/Llama-2-7B-chat-GGUF'
# model_basename = 'llama-2-7b-chat.Q2_K.gguf'
# model_path = hf_hub_download(repo_id=model_name_or_path, filename=model_basename)
# llm = LlamaCpp(model_path=model_path,
# n_ctx=8192,
# input={"temperature": 0.75, "max_length": 2000, "top_p": 1},
# verbose=True, )
# # λν κΈ°λ‘μ μ μ₯νκΈ° μν λ©λͺ¨λ¦¬λ₯Ό μμ±ν©λλ€.
# memory = ConversationBufferMemory(
# memory_key='chat_history', return_messages=True)
# # λν κ²μ 체μΈμ μμ±ν©λλ€.
# conversation_chain = ConversationalRetrievalChain.from_llm(
# llm=llm,
# retriever=vectorstore.as_retriever(),
# memory=memory
# )
# return conversation_chain # μμ±λ λν 체μΈμ λ°νν©λλ€.
# # μ¬μ©μ μ
λ ₯μ μ²λ¦¬νλ ν¨μμ
λλ€.
# def handle_userinput(user_question):
# print('user_question => ', user_question)
# # λν 체μΈμ μ¬μ©νμ¬ μ¬μ©μ μ§λ¬Έμ λν μλ΅μ μμ±ν©λλ€.
# response = st.session_state.conversation({'question': user_question})
# # λν κΈ°λ‘μ μ μ₯ν©λλ€.
# 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)
# text_chunks = []
# def initialize_conversation_chain():
# # Add the necessary code to initialize the conversation_chain
# # This may include loading the LlamaCpp model and creating the conversation_chain
# vectorstore = get_vectorstore(text_chunks) # Replace this with the appropriate code
# return get_conversation_chain(vectorstore)
# def main():
# load_dotenv()
# st.set_page_config(page_title="Chat with multiple Files",
# page_icon=":books:")
# st.write(css, unsafe_allow_html=True)
# # λν 체μΈμ΄ μΈμ
μνμ μκ±°λ NoneμΈ κ²½μ° μ΄κΈ°νν©λλ€.
# if "conversation" not in st.session_state or st.session_state.conversation is None:
# # μ μ ν λ°μ΄ν°λ‘ text_chunksλ₯Ό μ μν΄μΌ ν©λλ€.
# st.session_state.conversation = initialize_conversation_chain(text_chunks)
# # 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)
# if user_question:
# # Ensure that conversation_chain is initialized before calling handle_userinput
# if st.session_state.conversation is None:
# st.session_state.conversation = initialize_conversation_chain()
# handle_userinput(user_question)
# with st.sidebar:
# 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()
import streamlit as st
from dotenv import load_dotenv
from langchain.text_splitter import CharacterTextSplitter, RecursiveCharacterTextSplitter
from langchain.vectorstores import FAISS
from langchain.embeddings import HuggingFaceEmbeddings # General embeddings from HuggingFace models.
from langchain.memory import ConversationBufferMemory
from langchain.chains import ConversationalRetrievalChain
from htmlTemplates import css, bot_template, user_template
from langchain.llms import LlamaCpp
import json
from pathlib import Path
from pprint import pprint
from langchain.document_loaders import PyPDFLoader, TextLoader, JSONLoader, CSVLoader
import tempfile # μμ νμΌμ μμ±νκΈ° μν λΌμ΄λΈλ¬λ¦¬μ
λλ€.
import os
from huggingface_hub import hf_hub_download # Hugging Face Hubμμ λͺ¨λΈμ λ€μ΄λ‘λνκΈ° μν ν¨μμ
λλ€.
# PDF λ¬Έμλ‘λΆν° ν
μ€νΈλ₯Ό μΆμΆνλ ν¨μμ
λλ€.
def get_pdf_text(pdf_docs):
temp_dir = tempfile.TemporaryDirectory() # μμ λλ ν 리λ₯Ό μμ±ν©λλ€.
temp_filepath = os.path.join(temp_dir.name, pdf_docs.name) # μμ νμΌ κ²½λ‘λ₯Ό μμ±ν©λλ€.
with open(temp_filepath, "wb") as f: # μμ νμΌμ λ°μ΄λ리 μ°κΈ° λͺ¨λλ‘ μ½λλ€.
f.write(pdf_docs.getvalue()) # PDF λ¬Έμμ λ΄μ©μ μμ νμΌμ μλλ€.
pdf_loader = PyPDFLoader(temp_filepath) # PyPDFLoaderλ₯Ό μ¬μ©ν΄ PDFλ₯Ό λ‘λν©λλ€.
pdf_doc = pdf_loader.load() # ν
μ€νΈλ₯Ό μΆμΆν©λλ€.
return pdf_doc # μΆμΆν ν
μ€νΈλ₯Ό λ°νν©λλ€.
def get_text_file(text_docs):
temp_dir = tempfile.TemporaryDirectory() # μμ λλ ν 리λ₯Ό μμ±ν©λλ€.
temp_filepath = os.path.join(temp_dir.name, text_docs.name) # μμ νμΌ κ²½λ‘λ₯Ό μμ±ν©λλ€.
with open(temp_filepath, "wb") as f: # μμ νμΌμ ν
μ€νΈ μ°κΈ° λͺ¨λλ‘ μ½λλ€.
f.write(text_docs.getvalue()) # ν
μ€νΈ λ¬Έμμ λ΄μ©μ μμ νμΌμ μλλ€.
text_loader = TextLoader(temp_filepath) # TextLoaderλ₯Ό μ¬μ©ν΄ ν
μ€νΈ λ¬Έμλ₯Ό λ‘λν©λλ€.
text_doc = text_loader.load() # ν
μ€νΈλ₯Ό μΆμΆν©λλ€.
return text_doc # μΆμΆλ ν
μ€νΈλ₯Ό λ°νν©λλ€.
def get_csv_file(csv_docs):
temp_dir = tempfile.TemporaryDirectory() # μμ λλ ν 리λ₯Ό μμ±ν©λλ€.
temp_filepath = os.path.join(temp_dir.name, csv_docs.name) # μμ νμΌ κ²½λ‘λ₯Ό μμ±ν©λλ€.
with open(temp_filepath, "wb") as f: # μμ νμΌμ λ°μ΄λ리 μ°κΈ° λͺ¨λλ‘ μ½λλ€.
f.write(csv_docs.getvalue()) # CSV λ¬Έμμ λ΄μ©μ μμ νμΌμ μλλ€.
csv_loader = CSVLoader(temp_filepath) # CSVLoaderλ₯Ό μ¬μ©ν΄ CSV λ¬Έμλ₯Ό λ‘λν©λλ€.
csv_doc = csv_loader.load() # ν
μ€νΈλ₯Ό μΆμΆν©λλ€.
return csv_doc # μΆμΆλ ν
μ€νΈλ₯Ό λ°νν©λλ€.
def get_json_file(json_docs):
temp_dir = tempfile.TemporaryDirectory() # μμ λλ ν 리λ₯Ό μμ±ν©λλ€.
temp_filepath = os.path.join(temp_dir.name, json_docs.name) # μμ νμΌ κ²½λ‘λ₯Ό μμ±ν©λλ€.
with open(temp_filepath, "wb") as f: # μμ νμΌμ λ°μ΄λ리 μ°κΈ° λͺ¨λλ‘ μ½λλ€.
f.write(json_docs.getvalue()) # JSON λ¬Έμμ λ΄μ©μ μμ νμΌμ μλλ€.
json_loader = JSONLoader(file_path=temp_filepath, jq_schema='.messages[].content',text_content=False)
json_doc = json_loader.load()
return json_doc
# λ¬Έμλ€μ μ²λ¦¬νμ¬ ν
μ€νΈ μ²ν¬λ‘ λλλ ν¨μμ
λλ€.
def get_text_chunks(documents):
text_splitter = RecursiveCharacterTextSplitter(
chunk_size=1000, # μ²ν¬μ ν¬κΈ°λ₯Ό μ§μ ν©λλ€.
chunk_overlap=200, # μ²ν¬ μ¬μ΄μ μ€λ³΅μ μ§μ ν©λλ€.
length_function=len # ν
μ€νΈμ κΈΈμ΄λ₯Ό μΈ‘μ νλ ν¨μλ₯Ό μ§μ ν©λλ€.
)
documents = text_splitter.split_documents(documents) # λ¬Έμλ€μ μ²ν¬λ‘ λλλλ€.
return documents # λλ μ²ν¬λ₯Ό λ°νν©λλ€.
# ν
μ€νΈ μ²ν¬λ€λ‘λΆν° λ²‘ν° μ€ν μ΄λ₯Ό μμ±νλ ν¨μμ
λλ€.
def get_vectorstore(text_chunks):
# μνλ μλ² λ© λͺ¨λΈμ λ‘λν©λλ€.
embeddings = HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L12-v2',
model_kwargs={'device': 'cpu'})
vectorstore = FAISS.from_documents(text_chunks, embeddings) # FAISS λ²‘ν° μ€ν μ΄λ₯Ό μμ±ν©λλ€.
return vectorstore # μμ±λ λ²‘ν° μ€ν μ΄λ₯Ό λ°νν©λλ€.
# λν κΈ°λ‘μ μ μ₯νκΈ° μν λ©λͺ¨λ¦¬λ₯Ό μμ±ν©λλ€.
def get_conversation_chain(vectorstore):
model_name_or_path = 'TheBloke/Llama-2-7B-chat-GGUF'
model_basename = 'llama-2-7b-chat.Q2_K.gguf'
model_path = hf_hub_download(repo_id=model_name_or_path, filename=model_basename)
llm = LlamaCpp(model_path=model_path,
n_ctx=9000,
input={"temperature": 0.75, "max_length": 2000, "top_p": 1},
verbose=True, )
# λν κΈ°λ‘μ μ μ₯νκΈ° μν λ©λͺ¨λ¦¬λ₯Ό μμ±ν©λλ€.
memory = ConversationBufferMemory(
memory_key='chat_history', return_messages=True)
# λν κ²μ 체μΈμ μμ±ν©λλ€.
conversation_chain = ConversationalRetrievalChain.from_llm(
llm=llm,
retriever=vectorstore.as_retriever(),
memory=memory
)
return conversation_chain # μμ±λ λν 체μΈμ λ°νν©λλ€.
# μ¬μ©μ μ
λ ₯μ μ²λ¦¬νλ ν¨μμ
λλ€.
def handle_userinput(user_question):
print('user_question => ', user_question)
# λν 체μΈμ μ¬μ©νμ¬ μ¬μ©μ μ§λ¬Έμ λν μλ΅μ μμ±ν©λλ€.
response = st.session_state.conversation({'question': user_question})
# λν κΈ°λ‘μ μ μ₯ν©λλ€.
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:
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()
|