app.py

import numpy as np
import tensorflow as tf
import gradio as gr
from huggingface_hub import from_pretrained_keras
import cv2

img_size = 28
model = from_pretrained_keras("keras-io/keras-reptile")
examples = examples = [
            ['./example0.JPG'], 
            ['./example1.JPG']
]
def inference(original_image):
    image = tf.keras.utils.img_to_array(original_image)
    image = image.astype("float32") / 255.0
    image = image[:,:,0]
    image = np.reshape(image, (28, 28, 1))
    output = model.predict(image)
    return output

'''
def read_image(image):
    image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    image = tf.image.resize(image,(28,28,1))
    return image

def infer(model, image):
    predictions = model.predict(image)
def display_result(input_image):
    image = read_image(input_image)
    prediction_label = infer(model=model, image=image)
    return prediction_label

input = gr.inputs.Image()
examples = [["./example0.JPG"], ["./example1.JPG"]] 
title = "Few shot learning"
description = "Upload an image or select from examples to classify fashion items."

'''

interface = gr.Interface(
    fn = inference,
    title = "Few shot learning with Reptile",
    description = "Keras Implementation of Reptile",
    inputs = gr.inputs.Image(shape=(28, 28)),
    outputs = "text",
    examples = examples,
    article = "Author: <a href=\"https://huggingface.co/\">Animesh Maheshwari</a>. Based on the keras example from <a href=\"\"></a> \n Model Link: https://huggingface.co/keras-io/keras-reptile",
    ).launch(enable_queue=True, debug = True)