app.py

from UNet import *
import torch # ; print('Using torch version -', torch.__version__)
if torch.cuda.is_available():
    device = 'cuda'
else:
    device = 'cpu'
from torch.nn import Module, Conv2d, ReLU, ModuleList, MaxPool2d, ConvTranspose2d, BCELoss, BCEWithLogitsLoss, functional as F
from torch.optim import Adam
from torchvision import transforms
from torchvision.transforms import CenterCrop
from torch.utils.data import Dataset, DataLoader
import cv2
import gradio as gr

def getoutput(input_img):
    unet = UNet().to(device)
    unet = torch.load("unet_06_07_2022_17_13_42_256_256.pth").to(device)
    output_img = make_predictions(unet, input_img, threshold=0.5)
    return output_img

demo = gr.Interface(getoutput, gr.Image(shape=(200, 200)), "image")

demo.launch(share=True)