import streamlit as st
from transformers import BertForSequenceClassification, BertTokenizerFast
from emotion_utils import predict # Custom module for prediction
# Load the BERT model and tokenizer
model_path = "./model/"
model = BertForSequenceClassification.from_pretrained(model_path)
tokenizer = BertTokenizerFast.from_pretrained(model_path)
# Function to update sentiment analysis
def analyze_sentiment(text):
if text.strip():
probs, _, label = predict(text, model, tokenizer)
score = probs.max().item() # Get the highest probability score
return label, score
else:
return None, None
# Function to get emoji based on emotion
def get_emoji(label):
if label == "Anger":
return "😠"
elif label == "Astonished":
return "😲"
elif label == "Optimistic":
return "😊"
elif label == "Sadness":
return "😢"
else:
return "🙂"
# Streamlit app configuration
st.set_page_config(
page_title="G-Bert: Emotion Analysis",
page_icon="😊",
layout="centered"
)
# Custom CSS for a modern UI
st.markdown("""
""", unsafe_allow_html=True)
# Title and description
st.title("🌟 G-Bert: Emotion Analysis")
st.markdown("""
G-Bert is a Bangla sentiment analysis tool that uses a pre-trained BERT model to analyze the emotion of any Bengali or religious (Gita) text.
It can detect emotions like Anger, Astonished, Optimistic, and Sadness with a confidence score.
""")
# Text input
st.write("Enter some text below, and G-Bert will analyze its emotion for you!")
text = st.text_area("Input Text", height=150, placeholder="Type your text here...")
# Analyze button
# Analyze button
if st.button("✨ Analyze Emotion ✨"):
if text.strip():
label, score = analyze_sentiment(text)
if label and score:
emoji = get_emoji(label)
st.markdown(f"""
{emoji} Emotion: {label} {emoji}
Confidence Score: {score:.2f}
""", unsafe_allow_html=True)
# Add issue reporting section
st.markdown("""
🤔 Didn't get the expected result?
If you believe the emotion detected is incorrect, please let us know!
Your feedback will help us improve our model. Click the button below to report the issue:
""", unsafe_allow_html=True)
else:
st.error("🚨 Something went wrong with the analysis.")
else:
st.warning("⚠️ Please enter some text to analyze.")
# Footer with authorship
st.markdown("---")
st.markdown("""
""", unsafe_allow_html=True)