File size: 6,398 Bytes
ca4133a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
import argparse, os, sys, glob
import torch
import numpy as np
from omegaconf import OmegaConf
from PIL import Image
from tqdm import tqdm, trange
from einops import rearrange
from torchvision.utils import make_grid
from pytorch_lightning import seed_everything

from ldm.util import instantiate_from_config
from ldm.models.diffusion.ddim import DDIMSampler
from ldm.models.diffusion.plms import PLMSSampler
import pdb 
import random 


def load_model_from_config(config, ckpt, verbose=False):
    print(f"Loading model from {ckpt}")
    pl_sd = torch.load(ckpt, map_location="cpu")
    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


if __name__ == "__main__":
    parser = argparse.ArgumentParser()

    parser.add_argument(
        "--prompt",
        type=str,
        nargs="?",
        default="a painting of a virus monster playing guitar",
        help="the prompt to render"
    )

    parser.add_argument(
        "--outdir",
        type=str,
        nargs="?",
        help="dir to write results to",
        default="outputs/txt2img-samples"
    )
    parser.add_argument(
        "--ddim_steps",
        type=int,
        default=200,
        help="number of ddim sampling steps",
    )

    parser.add_argument(
        "--plms",
        action='store_true',
        help="use plms sampling",
    )

    parser.add_argument(
        "--ddim_eta",
        type=float,
        default=0.0,
        help="ddim eta (eta=0.0 corresponds to deterministic sampling",
    )
    parser.add_argument(
        "--n_iter",
        type=int,
        default=1,
        help="sample this often",
    )

    parser.add_argument(
        "--H",
        type=int,
        default=256,
        help="image height, in pixel space",
    )

    parser.add_argument(
        "--W",
        type=int,
        default=256,
        help="image width, in pixel space",
    )

    parser.add_argument(
        "--n_samples",
        type=int,
        default=4,
        help="how many samples to produce for the given prompt",
    )

    parser.add_argument(
        "--scale",
        type=float,
        default=5.0,
        help="unconditional guidance scale: eps = eps(x, empty) + scale * (eps(x, cond) - eps(x, empty))",
    )
    parser.add_argument(
        "--seed",
        type=int,
        default=42,
        help="the seed (for reproducible sampling)",
    )

    parser.add_argument(
        "--ema",
        action='store_true',
    )

    parser.add_argument(
        "--ckpt_path", 
        type=str, 
        default="/data/pretrained_models/ldm/text2img-large/model.ckpt", 
        help="Path to pretrained ldm text2img model")

    parser.add_argument(
        "--embedding_path", 
        type=str, 
        help="Path to a pre-trained embedding manager checkpoint")

    opt = parser.parse_args()
    seed = random.randint(1,500)
    seed_everything(seed)

    # config = OmegaConf.load("configs/latent-diffusion/txt2img-1p4B-eval_with_tokens.yaml")  # TODO: Optionally download from same location as ckpt and chnage this logic
    if opt.H == 512:
        config = OmegaConf.load("configs/finetune/finetune_clip.yaml")
    else:
        config = OmegaConf.load("configs/finetune/finetune_bert.yaml")

    config = OmegaConf.load("configs/finetune/finetune_generic.yaml")
    # config = OmegaConf.load("configs/latent-diffusion/txt2img-1p4B-finetune2.yaml")
    model = load_model_from_config(config, opt.ckpt_path)  # TODO: check path

    # model = load_model_from_config(config, "ckpt/model.ckpt")
    # model.cond_stage_model.load_state_dict(checkpoint['model_state_dict'])

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

    if opt.plms:
        sampler = PLMSSampler(model)
    else:
        sampler = DDIMSampler(model, device=device)

    os.makedirs(opt.outdir, exist_ok=True)
    outpath = opt.outdir
    
    prompt = opt.prompt


    sample_path = os.path.join(outpath, "samples")
    os.makedirs(sample_path, exist_ok=True)
    base_count = len(os.listdir(sample_path))
    start_code = None
    # start_code = torch.randn([opt.n_samples, 4, 32, 32], device=device)

    all_samples=list()
    with torch.no_grad():
        with model.ema_scope():
            uc = None
            if opt.scale != 1.0:
                uc = model.get_learned_conditioning(opt.n_samples * [""])
            for n in trange(opt.n_iter, desc="Sampling"):
                c = model.get_learned_conditioning(opt.n_samples * [prompt])
                shape = [4, opt.H//8, opt.W//8]
                samples_ddim, _ = sampler.sample(S=opt.ddim_steps,
                                                conditioning=c,
                                                batch_size=opt.n_samples,
                                                shape=shape,
                                                verbose=False,
                                                unconditional_guidance_scale=opt.scale,
                                                unconditional_conditioning=uc,
                                                eta=opt.ddim_eta)


                x_samples_ddim = model.decode_first_stage(samples_ddim)
                x_samples_ddim = torch.clamp((x_samples_ddim+1.0)/2.0, min=0.0, max=1.0)

                for x_sample in x_samples_ddim:
                    x_sample = 255. * rearrange(x_sample.cpu().numpy(), 'c h w -> h w c')
                    Image.fromarray(x_sample.astype(np.uint8)).save(os.path.join(sample_path, f"{base_count:04}.png"))
                    base_count += 1
                all_samples.append(x_samples_ddim)
                prompt = prompt

    # additionally, save as grid
    grid = torch.stack(all_samples, 0)
    grid = rearrange(grid, 'n b c h w -> (n b) c h w')
    grid = make_grid(grid, nrow=opt.n_samples)

    # to image
    grid = 255. * rearrange(grid, 'c h w -> h w c').cpu().numpy()
    Image.fromarray(grid.astype(np.uint8)).save(os.path.join(outpath, f'{prompt.replace(" ", "-")}.jpg'))

    print(f"Your samples are ready and waiting four you here: \n{outpath} \nEnjoy.")