File size: 8,065 Bytes
8e5c4a8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
import sys, argparse, glob, os
import torch
import numpy as np
from tqdm import tqdm
import gradio as gr
from PIL import Image
from omegaconf import OmegaConf
from einops import repeat, rearrange
from pytorch_lightning import seed_everything
from imwatermark import WatermarkEncoder

from scripts.txt2img import put_watermark
from ldm.models.diffusion.ddim import DDIMSampler
from ldm.models.diffusion.ddpm import LatentUpscaleDiffusion, LatentUpscaleFinetuneDiffusion
from ldm.util import exists, instantiate_from_config


torch.set_grad_enabled(False)


def load_model_from_config(config, ckpt, verbose=False):
    print(f"Loading model from {ckpt}")
    pl_sd = torch.load(ckpt, map_location="cpu")
    if "global_step" in pl_sd:
        print(f"Global Step: {pl_sd['global_step']}")
    sd = pl_sd["state_dict"]
    model = instantiate_from_config(config.model)
    m, u = model.load_state_dict(sd, strict=False)
    if len(m) > 0 and verbose:
        print("missing keys:")
        print(m)
    if len(u) > 0 and verbose:
        print("unexpected keys:")
        print(u)

    model.cuda()
    model.eval()
    return model


def make_batch_sd( image, txt, device,num_samples=1,size=(512,512)):
    image = Image.open(image).convert("RGB")
    image = image.resize(size)
    image = np.array(image)
    image = torch.from_numpy(image).to(dtype=torch.float32) / 127.5 - 1.0
    batch = {
        "lr": rearrange(image, 'h w c -> 1 c h w'),
        "txt": num_samples * [txt],
    }
    batch["lr"] = repeat(batch["lr"].to(device=device), "1 ... -> n ...", n=num_samples)
    return batch


def make_noise_augmentation(model, batch, noise_level=None):
    x_low = batch[model.low_scale_key]
    x_low = x_low.to(memory_format=torch.contiguous_format).float()
    x_aug, noise_level = model.low_scale_model(x_low, noise_level)
    return x_aug, noise_level


def paint(sampler, image, prompt, seed, scale, h, w, steps, num_samples=1, callback=None, eta=0., noise_level=None):
    device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
    model = sampler.model
    seed_everything(seed)
    prng = np.random.RandomState(seed)
    start_code = prng.randn(num_samples, model.channels, h, w)
    start_code = torch.from_numpy(start_code).to(device=device, dtype=torch.float32)

    with torch.no_grad(), torch.autocast("cuda"):
        batch = make_batch_sd(image, txt=prompt, device=device, num_samples=num_samples, size=(h, w))
        c = model.cond_stage_model.encode(batch["txt"])
        c_cat = list()
        if isinstance(model, LatentUpscaleFinetuneDiffusion):
            for ck in model.concat_keys:
                cc = batch[ck]
                if exists(model.reshuffle_patch_size):
                    assert isinstance(model.reshuffle_patch_size, int)
                    cc = rearrange(cc, 'b c (p1 h) (p2 w) -> b (p1 p2 c) h w',p1=model.reshuffle_patch_size, p2=model.reshuffle_patch_size)
                c_cat.append(cc)
            c_cat = torch.cat(c_cat, dim=1)
            # cond
            cond = {"c_concat": [c_cat], "c_crossattn": [c]}
            # uncond cond
            uc_cross = model.get_unconditional_conditioning(num_samples, "")
            uc_full = {"c_concat": [c_cat], "c_crossattn": [uc_cross]}
        elif isinstance(model, LatentUpscaleDiffusion):
            x_augment, noise_level = make_noise_augmentation(model, batch, noise_level)
            cond = {"c_concat": [x_augment], "c_crossattn": [c], "c_adm": noise_level}
            # uncond cond
            uc_cross = model.get_unconditional_conditioning(num_samples, "")
            uc_full = {"c_concat": [x_augment], "c_crossattn": [uc_cross], "c_adm": noise_level}
        else:
            raise NotImplementedError()

        shape = [model.channels, h, w]
        samples, intermediates = sampler.sample(
            steps,
            num_samples,
            shape,
            cond,
            verbose=False,
            eta=eta,
            unconditional_guidance_scale=scale,
            unconditional_conditioning=uc_full,
            x_T=start_code,
            callback=callback
        )
    with torch.no_grad():
        x_samples_ddim = model.decode_first_stage(samples)
    result = torch.clamp((x_samples_ddim + 1.0) / 2.0, min=0.0, max=1.0)
    result = result.cpu().numpy().transpose(0, 2, 3, 1) * 255
    return Image.fromarray(result.astype(np.uint8)[0])


if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("--indir", type=str, nargs="?", help="dir containing image-mask pairs (`example.png` and `example_mask.png`)",)
    parser.add_argument("--num_imgs", type=int, default=None, help="number of images to generate",)
    parser.add_argument("--steps",type=int,default=50,help="number of ddim sampling steps",)
    parser.add_argument("--config",type=str,default="/checkpoint/pfz/autoencoders/sd/stable-diffusion-x4-upscaler/x4-upscaling.yaml",help="path to config which constructs model",)
    parser.add_argument("--ckpt",type=str,default="/checkpoint/pfz/autoencoders/sd/stable-diffusion-x4-upscaler/x4-upscaler-ema.ckpt",help="path to checkpoint of model",)
    parser.add_argument("--ldm_decoder_ckpt",default=None,type=str,help="path to checkpoint of LDM decoder")
    parser.add_argument("--num_samples",default=1,type=int,help="number of samples to generate")
    parser.add_argument("--scale", default=10.0, type=float, help="scale")
    parser.add_argument("--eta", default=0.0, type=float, help="eta")
    parser.add_argument("--noise_level", default=20, type=float, help="eta")
    parser.add_argument("--output_dir",type=str,default="outputs",nargs="?",help="dir to write results to",)
    parser.add_argument("--height",type=int,default=512,help="height of output image",)
    parser.add_argument("--width",type=int,default=512,help="width of output image",)
    parser.add_argument("--seed",type=int,default=0,help="random seed",)
    opt = parser.parse_args()

    print(f'>>> Building LDM model with config {opt.config} and weights from {opt.ckpt}...')
    config = OmegaConf.load(f"{opt.config}")
    model = load_model_from_config(config, f"{opt.ckpt}")

    # Parameter None for clutil sweep
    print(f'reload decoder weights {opt.ldm_decoder_ckpt}...')
    if opt.ldm_decoder_ckpt is not None and opt.ldm_decoder_ckpt.lower() == "none":
        opt.ldm_decoder_ckpt = None
    if opt.ldm_decoder_ckpt is not None:
        state_dict = torch.load(opt.ldm_decoder_ckpt)['ldm_decoder']
        msg = model.first_stage_model.load_state_dict(state_dict, strict=False)
        print(msg)

    device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
    model = model.to(device)
    model.eval()
    sampler = DDIMSampler(model)

    os.makedirs(opt.output_dir, exist_ok=True)

    images = sorted(glob.glob(os.path.join(opt.indir, "*.png"))) + sorted(glob.glob(os.path.join(opt.indir, "*.jpg"))) + sorted(glob.glob(os.path.join(opt.indir, "*.jpeg")))
    images += sorted(glob.glob(os.path.join(opt.indir, "*.PNG"))) + sorted(glob.glob(os.path.join(opt.indir, "*.JPG"))) + sorted(glob.glob(os.path.join(opt.indir, "*.JPEG")))
    print(f"Found {len(images)} inputs.")

    counter = 0
    for image in tqdm(images):
        if opt.num_imgs is not None and counter >= opt.num_imgs:
            break
        noise_level = torch.Tensor( opt.num_samples * [opt.noise_level]).to(sampler.model.device).long()
        sampler.make_schedule(opt.steps, ddim_eta=opt.eta, verbose=True)
        result = paint(
            sampler=sampler,
            image=image,
            prompt="",
            seed=opt.seed,
            scale=opt.scale,
            h=opt.height, w=opt.width, steps=opt.steps,
            num_samples=opt.num_samples,
            callback=None,
            noise_level=noise_level
        )
        outpath = os.path.join(opt.output_dir, os.path.split(image)[1]).replace('.jpg', '.png').replace('.jpeg', '.png').replace('.JPG', '.png').replace('.JPEG', '.png')
        result.save(outpath)
        counter += 1