app.py

import gradio as gr
from transformers import pipeline

# Load the sentiment classification model
classifier = pipeline("text-classification", model="distilbert-base-uncased-finetuned-sst-2-english")

# Define the function for Gradio interface
def classify_sentiment(text):
    if not text.strip():
        return "⚠️ Please enter a sentence to analyze.", ""
    result = classifier(text)[0]
    label = result['label']
    confidence = round(result['score'] * 100, 2)
    return f"**Prediction:** {label}", f"**Confidence:** {confidence}%"

# Create Gradio interface
with gr.Blocks() as demo:
    gr.Markdown("# 🔍 Real-Time Sentiment Classifier")
    gr.Markdown("""
    ### 💡 Model Instruction:
    This model is trained for binary sentiment classification — **Positive** and **Negative** only.  
    Neutral or mixed opinions may be interpreted as leaning toward one side.  
    For best results, input clearly positive or negative sentences.
    """)

    with gr.Row():
        with gr.Column():
            text_input = gr.Textbox(
                label="Enter your sentence below 👇",
                placeholder='Example: "I love the features of this app!" or "The update ruined the experience."',
                lines=3
            )
            analyze_button = gr.Button("Analyze Sentiment")

        with gr.Column():
            prediction_output = gr.Markdown()
            confidence_output = gr.Markdown()

    analyze_button.click(fn=classify_sentiment, inputs=text_input, outputs=[prediction_output, confidence_output])

# Launch the app
demo.launch()