# Import necessary libraries
from transformers import pipeline
import gradio as gr

# Load the Filipino sentiment analysis model
pipe = pipeline("text-classification", model="raphgonda/FilipinoShopping")

# Define the sentiment analysis function
def analyze_sentiment(text):
    try:
        # Predict sentiment using the model
        results = pipe(text)
        # Extract label and score
        label = results[0]["label"]
        score = round(results[0]["score"] * 100, 2)  # Convert score to percentage
        return label, f"{score}%"
    except Exception as e:
        return "Error", "N/A"

# Create a Gradio interface with custom UI
with gr.Blocks() as interface:
    gr.Markdown("<h1 style='text-align: center;'>Filipino Sentiment Analysis</h1>")
    gr.Markdown("<p style='text-align: center;'>Enter text in Filipino to analyze its sentiment.</p>")

    with gr.Row():
        input_text = gr.Textbox(
            label="Enter text to analyze its sentiment",
            placeholder="Type your text here...",
        )

    with gr.Row():
        submit_btn = gr.Button("Submit")
        clear_btn = gr.Button("Clear")
    
    sentiment_label = gr.Textbox(label="Sentiment Label", interactive=False, visible=True)
    
    with gr.Row():
        emotion_score = gr.Textbox(label="Emotion Score", interactive=False)

    examples = gr.Examples(
        examples=[
            ["Okay ang aesthetic"],
            ["Mabagal ang delivery"],
            ["Napakaganda ng serbisyo!"],
            ["Ang pangit ng produkto."]
        ],
        inputs=input_text,
    )

    # Define the function connection
    submit_btn.click(
        analyze_sentiment,
        inputs=[input_text],
        outputs=[sentiment_label, emotion_score],
    )
    clear_btn.click(
        lambda: ("", ""),
        inputs=[],
        outputs=[sentiment_label, emotion_score],
    )

# Launch the app
interface.launch()