import gradio as gr
import tensorflow as tf
import numpy as np
import gdown
from PIL import Image

input_shape = (32, 32, 3)
resized_shape = (224, 224, 3)
num_classes = 10
labels = {
    0: "plane",
    1: "car",
    2: "bird",
    3: "cat",
    4: "deer",
    5: "dog",
    6: "frog",
    7: "horse",
    8: "ship",
    9: "truck",
}

# Download the model file
def download_model():
    url = "https://drive.google.com/uc?id=12700bE-pomYKoVQ214VrpBoJ7akXcTpL"
    output = "modelV2Lmixed.keras"
    gdown.download(url, output, quiet=False)
    return output

model_file = download_model()

# Load the model
model = tf.keras.models.load_model(model_file)

# Perform image classification
def predict_class(image):
    img = tf.cast(image, tf.float32)
    img = tf.image.resize(img, [input_shape[0], input_shape[1]])
    img = np.expand_dims(img, axis=0)
    prediction = model.predict(img)
    return prediction[0]

# UI Design
def classify_image(image):
    pred = predict_class(image)
    class_names = [
        "plane",
        "car",
        "bird",
        "cat",
        "deer",
        "dog",
        "frog",
        "horse",
        "ship",
        "truck",
    ]

    probabilities = tf.nn.softmax(pred)
    top_indices = tf.argsort(probabilities, direction='DESCENDING')
    top_classes = [class_names[idx] for idx in top_indices]
    top_probs = [probabilities[idx] for idx in top_indices]

    output = "<h3>Top 3 Predictions:</h3>"
    for i in range(3):
        output += f"<p>{top_classes[i]}: {top_probs[i]*100:.2f}%</p>"

    return output

inputs = gr.inputs.Image(label="Upload an image")
outputs = gr.outputs.HTML()

title = "<h1 style='text-align: center;'>Image Classifier</h1>"
description = "Upload an image and get the top 3 predictions."

gr.Interface(fn=classify_image, inputs=inputs, outputs=outputs, title=title, description=description).launch()