Spaces:
Sleeping
Sleeping
File size: 8,470 Bytes
13b9071 6332bb2 13b9071 98a0661 13b9071 38c660f 13b9071 38c660f 13b9071 38c660f 13b9071 38c660f 13b9071 38c660f 13b9071 6332bb2 13b9071 |
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 |
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 # For loading transformer models.
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(docs):
temp_dir = tempfile.TemporaryDirectory() # μμ λλ ν 리λ₯Ό μμ±ν©λλ€.
temp_filepath = os.path.join(temp_dir.name, 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(docs.getvalue()) # ν
μ€νΈ νμΌμ λ΄μ©μ μμ νμΌμ μλλ€.
text_loader = TextLoader(temp_filepath) # TextLoaderλ₯Ό μ¬μ©ν΄ ν
μ€νΈλ₯Ό λ‘λν©λλ€.
text_doc = text_loader.load() # ν
μ€νΈλ₯Ό μΆμΆν©λλ€.
return text_doc # μΆμΆν ν
μ€νΈλ₯Ό λ°νν©λλ€.
def get_csv_text(csv_file):
# μμ νμΌμ μμ±νκ³ , μ
λ‘λλ CSV νμΌμ λ΄μ©μ μλλ€.
with tempfile.NamedTemporaryFile(delete=False, suffix=".csv") as temp_file:
temp_file.write(csv_file.read())
temp_file_path = temp_file.name # μμ νμΌ κ²½λ‘λ₯Ό μ μ₯ν©λλ€.
# CSV νμΌμ μ΄κ³ λ΄μ©μ μ½μ΅λλ€.
with open(temp_file_path, mode='r', encoding='utf-8') as file:
csv_reader = csv.reader(file)
csv_text = "\n".join([", ".join(row) for row in csv_reader]) # κ° νμ λ¬Έμμ΄λ‘ λ³νν©λλ€.
os.remove(temp_file_path) # μμ νμΌμ μμ ν©λλ€.
return csv_text # CSV νμΌμ ν
μ€νΈλ₯Ό λ°νν©λλ€.
def get_json_file(docs):
temp_dir = tempfile.TemporaryDirectory() # μμ λλ ν 리λ₯Ό μμ±ν©λλ€.
temp_filepath = os.path.join(temp_dir.name, 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(docs.getvalue()) # JSON νμΌμ λ΄μ©μ μμ νμΌμ μλλ€.
json_loader = JSONLoader(temp_filepath) # JSONLoaderλ₯Ό μ¬μ©ν΄ JSONμ λ‘λν©λλ€.
json_doc = json_loader.load() # ν
μ€νΈλ₯Ό μΆμΆν©λλ€.
return json_doc # μΆμΆν ν
μ€νΈλ₯Ό λ°νν©λλ€.
# 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="bert-base-multilingual-cased",
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=4086,
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.title("λ¬Έμμ λν΄ μ§λ¬Ένλ μ±ν
")
st.sidebar.title("λ¬Έμ μ
λ‘λ")
st.sidebar.text("μ¬κΈ°μ μ¬λ¬λΆμ λ¬Έμλ₯Ό μ
λ‘λνμΈμ.")
st.sidebar.text("PDF, TXT, CSV, JSON νμΌ νμμ μ§μν©λλ€.")
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("λ¬Έμμ λν΄ μ§λ¬Ένλ μ±ν
")
user_question = st.text_input("λ¬Έμ μ λν΄μ μ§λ¬Έμ ν΄λ³΄μΈμ")
if user_question:
handle_userinput(user_question)
with st.sidebar:
st.subheader("Your documents")
docs = st.file_uploader("μ¬κΈ°μ λ¬Έμλ₯Ό μ
λ‘λνκ³ 'μ²λ¦¬'λ₯Ό ν΄λ¦νμΈμ", accept_multiple_files=True)
if st.button("μ²λ¦¬"):
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()
|