app.py

import gradio as gr
from fastai.vision.all import *
from os.path import file_exists
import requests

model_fn = 'quick_224px'
if not file_exists(model_fn):
  with requests.get(url, stream=True) as r:
    r.raise_for_status()
    with open(model_fn, 'wb') as f:
      for chunk in r.iter_content(chunk_size=8192):
        f.write(chunk)

# Load the model
def open_img(fn:Path): return Image.open(fn).convert('RGB').copy()
def get_images(path):
  return get_image_files(path/'train') + get_image_files(path/'test')
def get_x(item):
  return np.array(open_img(item).crop((0, 0, 512, 512)))
def get_y(item):
  return np.array(open_img(item).crop((512, 0, 1024, 512)))
sketch_model = load_learner(model_fn)


def sketchify(image_path):
   pred = sketch_model.predict(image_path)
   np_im = pred[0].permute(1, 2, 0).numpy()
   return np_im
    
    

iface = gr.Interface(fn=sketchify, 
  inputs=[gr.inputs.Image(shape=(512, 512), type="filepath")], 
  outputs=[gr.outputs.Image(type="numpy", label="Output Image")]
)
iface.launch()