Spaces:
Runtime error
Runtime error
import streamlit as st | |
from transformers import pipeline | |
# Load the sentiment classifier model | |
distilled_student_sentiment_classifier = pipeline( | |
model="lxyuan/distilbert-base-multilingual-cased-sentiments-student", | |
return_all_scores=True | |
) | |
# Define the Streamlit app | |
def main(): | |
# Add a title to the app | |
st.title("DistilBERT Sentiment Analysis") | |
# Add a text input field for user input | |
user_input = st.text_area("Enter text:", height=100) | |
# Perform sentiment analysis when the user submits input | |
if st.button("Analyze"): | |
# Check if the input text is empty | |
if not user_input.strip(): | |
st.error("Please enter some text.") | |
else: | |
# Display a loading message while performing analysis | |
with st.spinner("Analyzing..."): | |
# Perform sentiment analysis on the input text | |
result = distilled_student_sentiment_classifier(user_input) | |
# Print the type and content of the result for debugging | |
st.write("Result Type:", type(result)) | |
st.write("Result Content:", result) | |
# Check if the result is a dictionary | |
if isinstance(result, dict): | |
# Access the 'scores' dictionary and check if sentiment is negative | |
negative = result.get('scores', {}).get('label') == 'negative' | |
# If sentiment is negative, print "Negative" | |
if negative: | |
st.write("Sentiment Analysis Result: Negative") | |
else: | |
st.error("Unexpected data format in result.") | |
# Run the Streamlit app | |
if __name__ == "__main__": | |
main() | |