import gradio as gr
import tensorflow as tf
from PIL import Image
import numpy as np
 
labels = ['barrel_jellyfish','blue_jellyfish','compass_jellyfish','lions_mane_jellyfish','mauve_stinger_jellyfish','Moon_jellyfish']
 
def predict_jellyfish_type(uploaded_file):
    
    if uploaded_file is None:
        return "No file uploaded."
   
    model = tf.keras.models.load_model('Jellyfish_transferlearning.keras')
    # Load the image from the file path
    with Image.open(uploaded_file) as img:
        img = img.resize((150, 150)).convert('RGB')  # Convert image to RGB
        img_array = np.array(img)
        
 
        prediction = model.predict(np.expand_dims(img_array, axis=0))
        confidences = {labels[i]: np.round(float(prediction[0][i]), 2) for i in range(len(labels))}
 
        return confidences
 
 
# Define the Gradio interface
iface = gr.Interface(
    fn=predict_jellyfish_type,  # Function to process the input
    inputs=gr.File(label="Upload File"),  # File upload widget
    outputs="text",  # Output type
    title="Jellyfish Classifier",  # Title of the interface
    examples=["images/barrel_jellyfish.jpeg", "images/blue_jellyfish.jpeg", "images/compass_jellyfish.jpeg", "images/lions_mane_jellyfish.jpeg", "images/mauve_stinger_jellyfish.jpeg", "images/Moon_jellyfish.jpeg"],   
    description="Upload a picture of a Jellyfish (barrel Gefährlichkeit: Niedrig, blue Gefährlichkeit: Moderat, compass Gefährlichkeit: Moderat, lions mane Gefährlichkeit: Hoch, mauve stinger Gefährlichkeit: Moderat bis Hoch, Moon Gefährlichkeit: Niedrig) "  # Description of the interface
)
 
# Launch the interface
iface.launch()