import streamlit as st | |
from transformers import pipeline | |
# Load the sentiment analysis model | |
model_name = "poom-sci/WangchanBERTa-finetuned-sentiment" | |
sentiment_analyzer = pipeline('sentiment-analysis', model=model_name) | |
# Streamlit app | |
st.title("Thai Sentiment Analysis App") | |
# Input text | |
text_input = st.text_area("Enter Thai text for sentiment analysis", "ขอความเห็นหน่อย... ") | |
# Button to trigger analysis | |
if st.button("Analyze Sentiment"): | |
# Analyze sentiment using the model | |
results = sentiment_analyzer([text_input]) | |
# Extract sentiment and score | |
sentiment = results[0]['label'] | |
score = results[0]['score'] | |
# Display result as progress bars | |
st.subheader("Sentiment Analysis Result:") | |
if sentiment == 'pos': | |
st.success(f"Positive Sentiment (Score: {score:.2f})") | |
st.progress(score) | |
elif sentiment == 'neg': | |
st.error(f"Negative Sentiment (Score: {score:.2f})") | |
st.progress(score) | |
else: | |
st.warning(f"Neutral Sentiment (Score: {score:.2f})") | |
st.progress(score) | |