File size: 3,366 Bytes
c9e4251
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6e970c6
c9e4251
 
 
6e970c6
c9e4251
 
6e970c6
 
c9e4251
 
 
6e970c6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c9e4251
 
 
 
6e970c6
c9e4251
6e970c6
c9e4251
 
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
# Copyright 2022 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and

# limitations under the License.


import torch

import tqdm
from diffusers import DiffusionPipeline


class DDIM(DiffusionPipeline):
    def __init__(self, unet, noise_scheduler):
        super().__init__()
        self.register_modules(unet=unet, noise_scheduler=noise_scheduler)

    def __call__(self, batch_size=1, generator=None, torch_device=None, eta=0.0, num_inference_steps=50):
        # eta corresponds to η in paper and should be between [0, 1]
        if torch_device is None:
            torch_device = "cuda" if torch.cuda.is_available() else "cpu"

        num_trained_timesteps = self.noise_scheduler.num_timesteps
        inference_step_times = range(0, num_trained_timesteps, num_trained_timesteps // num_inference_steps)

        self.unet.to(torch_device)
        image = self.noise_scheduler.sample_noise(
            (batch_size, self.unet.in_channels, self.unet.resolution, self.unet.resolution),
            device=torch_device,
            generator=generator,
        )

        for t in tqdm.tqdm(reversed(range(num_inference_steps)), total=num_inference_steps):
            # get actual t and t-1
            train_step = inference_step_times[t]
            prev_train_step = inference_step_times[t - 1] if t > 0 else -1

            # compute alphas
            alpha_prod_t = self.noise_scheduler.get_alpha_prod(train_step)
            alpha_prod_t_prev = self.noise_scheduler.get_alpha_prod(prev_train_step)
            alpha_prod_t_rsqrt = 1 / alpha_prod_t.sqrt()
            alpha_prod_t_prev_rsqrt = 1 / alpha_prod_t_prev.sqrt()
            beta_prod_t_sqrt = (1 - alpha_prod_t).sqrt()
            beta_prod_t_prev_sqrt = (1 - alpha_prod_t_prev).sqrt()

            # compute relevant coefficients
            coeff_1 = (
                (alpha_prod_t_prev - alpha_prod_t).sqrt()
                * alpha_prod_t_prev_rsqrt
                * beta_prod_t_prev_sqrt
                / beta_prod_t_sqrt
                * eta
            )
            coeff_2 = ((1 - alpha_prod_t_prev) - coeff_1**2).sqrt()

            # model forward
            with torch.no_grad():
                noise_residual = self.unet(image, train_step)

            # predict mean of prev image
            pred_mean = alpha_prod_t_rsqrt * (image - beta_prod_t_sqrt * noise_residual)
            pred_mean = torch.clamp(pred_mean, -1, 1)
            pred_mean = (1 / alpha_prod_t_prev_rsqrt) * pred_mean + coeff_2 * noise_residual

            # if eta > 0.0 add noise. Note eta = 1.0 essentially corresponds to DDPM
            if eta > 0.0:
                noise = self.noise_scheduler.sample_noise(image.shape, device=image.device, generator=generator)
                image = pred_mean + coeff_1 * noise
            else:
                image = pred_mean

        return image