|
import os |
|
|
|
import streamlit as st |
|
from dotenv import load_dotenv |
|
from langchain_community.llms import VLLMOpenAI |
|
|
|
from preprocessing import preprocess_pipeline |
|
|
|
load_dotenv() |
|
|
|
|
|
def disable_input(): |
|
st.session_state.translating = True |
|
|
|
|
|
def translate(prompt, top_p, temperature, repetition_penalty): |
|
llm = VLLMOpenAI( |
|
openai_api_key="EMPTY", |
|
openai_api_base=os.getenv("VISTRAL_7B_MT_SERVER"), |
|
model_name="nguyen1207/Vistral-7B-MT-0.1", |
|
temperature=temperature, |
|
top_p=top_p, |
|
frequency_penalty=repetition_penalty, |
|
) |
|
|
|
stream = llm.stream( |
|
prompt, |
|
) |
|
|
|
count = 0 |
|
|
|
for response in stream: |
|
if count < 3: |
|
count += 1 |
|
yield "" |
|
else: |
|
yield response |
|
|
|
|
|
st.set_page_config(page_title="Vietnamese to English Translation") |
|
|
|
st.title( |
|
"🇻🇳 Vietnamese to 🇺🇸 English Translation but with Teencode and Slang understanding 🤯" |
|
) |
|
|
|
st.sidebar.header("Translation Parameters") |
|
top_p = st.sidebar.slider("Top p", min_value=0.0, max_value=1.0, value=0.95) |
|
temperature = st.sidebar.slider("Temperature", min_value=0.0, max_value=2.0, value=0.3) |
|
repetition_penalty = st.sidebar.slider( |
|
"Repetition Penalty", min_value=1.0, max_value=3.0, value=1.05 |
|
) |
|
|
|
if "messages" not in st.session_state: |
|
st.session_state.messages = [] |
|
st.session_state.translating = False |
|
|
|
for message in st.session_state.messages: |
|
with st.chat_message(message["role"]): |
|
st.markdown(message["content"]) |
|
|
|
|
|
if user_input := st.chat_input( |
|
"Vietnamese text goes here... 🇻🇳", |
|
disabled=st.session_state.translating, |
|
on_submit=disable_input, |
|
): |
|
if user_input.strip() != "": |
|
st.session_state.translating = True |
|
preprocessed_input = preprocess_pipeline(user_input) |
|
|
|
st.session_state.messages.append({"role": "user", "content": user_input}) |
|
with st.chat_message("user"): |
|
st.markdown(user_input) |
|
|
|
with st.chat_message("assistant"): |
|
prompt_template = """<s> [INST] Dịch câu sau từ tiếng Việt sang tiếng Anh: |
|
|
|
Tiếng Việt: {} [/INST] """ |
|
|
|
prompt = prompt_template.format(preprocessed_input) |
|
|
|
stream = translate(prompt, top_p, temperature, repetition_penalty) |
|
|
|
translation = st.write_stream(stream) |
|
st.markdown(translation) |
|
|
|
st.session_state.messages.append({"role": "assistant", "content": translation}) |
|
|
|
|
|
st.session_state.translating = False |
|
st.rerun() |
|
|
