app.py

import gradio as gr
from transformers import pipeline
from  transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("bert-base-cased")

# Load the text classification pipeline from Hugging Face
classifier = pipeline("text-classification", model="./clickbait-classifier-model-90", tokenizer=tokenizer)


def classify_text(text):
    prediction = classifier(text)[0]
    clickbait_label = "LABEL_1"  # Assuming LABEL_1 corresponds to clickbait
    non_clickbait_label = "LABEL_0"  # Assuming LABEL_0 corresponds to non-clickbait

    predicted_label = prediction["label"]
    predicted_score = prediction["score"] * 100

    clickbait_score = predicted_score if predicted_label == clickbait_label else 0
    non_clickbait_score = predicted_score if predicted_label == non_clickbait_label else 0

    return clickbait_score, non_clickbait_score

# Example clickbait headline
clickbait_example = ["You'll Never Believe What This Dog Did Next!"]

# Example non-clickbait headline
non_clickbait_example = ["Local School Board Approves New Budget"]

# Combine into a list of examples
examples = [clickbait_example, non_clickbait_example]

# Create the Gradio interface
iface = gr.Interface(
    fn=classify_text,
    inputs=[gr.Textbox(lines=2, placeholder="Enter a text headline...")],
    outputs=[
        gr.Slider(label="Clickbait", minimum=0, maximum=100, step=1),
        gr.Slider(label="Non-Clickbait", minimum=0, maximum=100, step=1),
    ],
    title="Clickbait Detector",
    examples=examples,
)

# Launch the interface
iface.launch()