File size: 2,499 Bytes
8a4a948
 
 
 
 
 
 
 
 
 
 
 
 
 
735db2e
 
8a4a948
 
1ea821b
 
 
 
 
e4bf64c
 
1ea821b
 
8a4a948
1ea821b
8a4a948
 
1ea821b
 
8a4a948
 
 
 
 
338f71e
8a4a948
 
 
 
 
735db2e
1ea821b
 
8a4a948
 
 
 
d554e69
e4bf64c
1ea821b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
import gradio as gr
import cv2
import torch
import numpy as np
from torchvision import transforms
import torch
from pytorch_lightning import seed_everything
from torchvision.utils import save_image
from model_lib.modules import MoMA_main_modal
from model_lib.utils import parse_args
import os
os.environ["CUDA_VISIBLE_DEVICES"]="0"

title = "MoMA"
description = "This model has to run on GPU. By default, we load the model with 4-bit quantization to make it fit in smaller hardwares."

article = "<p style='text-align: center'><a href='https://news.machinelearning.sg/posts/beautiful_profile_pics_remove_background_image_with_deeplabv3/'>Blog</a> | <a href='https://github.com/eugenesiow/practical-ml'>Github Repo</a></p>"

def MoMA_demo(rgb, subject, prompt, strength, seed):
    seed = int(seed) if seed else 0
    try:
        seed = int(seed)
    except ValueError:
        seed = 0
    seed = seed if not seed == 0 else np.random.randint(0,1000)
    print(f"Seed: {seed}")
    
    with torch.no_grad():
        generated_image = model.generate_images(rgb, subject, prompt, strength=strength, seed=seed)
    return generated_image

def inference(rgb, subject, prompt, strength, seed):
    result = MoMA_demo(rgb, subject, prompt, strength, seed)
    return result

seed_everything(0)
args = parse_args()
#load MoMA from HuggingFace. Auto download
model = MoMA_main_modal(args).to(args.device, dtype=torch.float16)

gr.Interface(
    inference,
    [gr.Image(type="pil", label="Input RGB"),
     gr.Textbox(lines=1, label="subject"),
     gr.Textbox(lines=1, label="Prompt"),
     gr.Slider(minimum=0.2, maximum=1.2, step=0.1,label="Strength. Recommend: 1.0 for context editing; 0.4 for texture editing",value=1.0),
     gr.Textbox(lines=1, label="Seed. Use 0 for a random seed")],
    gr.Image(type="pil", label="Output"),
    title=title,
    description=description,
    article=article,
    examples=[["example_images/newImages/3.jpg",'car','A car in autumn with falling leaves.',1.0,"6"],["example_images/newImages/3.jpg",'car','A wooden sculpture of a car on a table.',0.4,"4"],["example_images/newImages/2.jpg",'car','A car on a city road with trees and buildings.',1.0,"4"],["example_images/newImages/2.jpg",'car','A green jade of a car in a garden.',0.5,"12"],["example_images/newImages/1.jpeg",'bird','A bird in spring with flowers.',1.0,"1"],["example_images/newImages/1.jpeg",'bird','A golden statue of a bird on a table.',0.4,"1"]],
    allow_flagging='never'
).launch(debug=False)