app.py

import gradio as gr
from transformers import pipeline

# Load the model pipeline
pipe = pipeline("image-classification", "dima806/medicinal_plants_image_detection")

# Define the image classification function
def image_classifier(image):
    # Perform image classification
    outputs = pipe(image)
    # Get the label of the first result
    output_text = outputs[0]['label']
    return output_text

# Define app title and description with HTML formatting
title = "<h1 style='text-align: center; color: #4CAF50;'>Image Classification</h1>"
description = "<p style='text-align: center; font-size: 18px;'>This application serves to classify Medicinal Plants </p>"

# Define custom CSS styles for the Gradio app
custom_css = """
            .gradio-interface {
                max-width: 600px;
                margin: auto;
                border-radius: 10px;
                box-shadow: 0px 0px 10px rgba(0, 0, 0, 0.1);
            }
            .title-container {
                padding: 20px;
                background-color: #f0f0f0;
                border-top-left-radius: 10px;
                border-top-right-radius: 10px;
            }
            .description-container {
                padding: 20px;
            }
            """

# Launch the Gradio interface with custom HTML and CSS
demo = gr.Interface(fn=image_classifier, inputs=gr.Image(type="pil"), outputs="textbox", title=title, description=description, 
                     theme="gstaff/sketch", css=custom_css,
                   )
demo.launch()