File size: 6,982 Bytes
2b5b9ef
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import torch
import torch.nn as nn
import torch.nn.functional as F
import sys

sys.path.insert(0, '.')  # nopep8
from ldm.modules.losses_audio.vqperceptual import *

def sequence_mask(length, max_length=None):# length shape (B,)
    if max_length is None:
        max_length = length.max()
    x = torch.arange(max_length, dtype=length.dtype, device=length.device)# (max_length)
    return x.unsqueeze(0) < length.unsqueeze(1)# (B,max_length)

class LPAPSWithDiscriminator(nn.Module):
    def __init__(self, disc_start, logvar_init=0.0, kl_weight=1.0, pixelloss_weight=1.0,

                 disc_num_layers=3, disc_in_channels=3, disc_factor=1.0, disc_weight=1.0,

                 perceptual_weight=1.0, use_actnorm=False, disc_conditional=False,

                 disc_loss="hinge",pad_value=-1):
        super().__init__()
        assert disc_loss in ["hinge", "vanilla"]
        self.pad_val = pad_value
        self.kl_weight = kl_weight
        self.pixel_weight = pixelloss_weight
        self.perceptual_weight = perceptual_weight
        if self.perceptual_weight > 0:
            self.perceptual_loss = LPAPS().eval()# LPIPS用于日常图像,而LPAPS用于梅尔谱图
        
        # output log variance
        self.logvar = nn.Parameter(torch.ones(size=()) * logvar_init)

        self.discriminator = NLayerDiscriminator(input_nc=disc_in_channels,
                                                 n_layers=disc_num_layers,
                                                 use_actnorm=use_actnorm,
                                                 ).apply(weights_init)
        self.discriminator_iter_start = disc_start
        if disc_loss == "hinge":
            self.disc_loss = hinge_d_loss
        elif disc_loss == "vanilla":
            self.disc_loss = vanilla_d_loss
        else:
            raise ValueError(f"Unknown GAN loss '{disc_loss}'.")
        print(f"LPAPSWithDiscriminator running with {disc_loss} loss.")
        self.disc_factor = disc_factor
        self.discriminator_weight = disc_weight
        self.disc_conditional = disc_conditional


    def calculate_adaptive_weight(self, nll_loss, g_loss, last_layer=None):
        if last_layer is not None:
            nll_grads = torch.autograd.grad(nll_loss, last_layer, retain_graph=True)[0]
            g_grads = torch.autograd.grad(g_loss, last_layer, retain_graph=True)[0]
        else:
            nll_grads = torch.autograd.grad(nll_loss, self.last_layer[0], retain_graph=True)[0]
            g_grads = torch.autograd.grad(g_loss, self.last_layer[0], retain_graph=True)[0]

        d_weight = torch.norm(nll_grads) / (torch.norm(g_grads) + 1e-4)
        d_weight = torch.clamp(d_weight, 0.0, 1e4).detach()
        d_weight = d_weight * self.discriminator_weight
        return d_weight

    def forward(self, inputs, reconstructions, posteriors, optimizer_idx,

                global_step, last_layer=None, cond=None, split="train", weights=None):
        if len(inputs.shape) == 3:
            inputs,reconstructions = inputs.unsqueeze(1),reconstructions.unsqueeze(1)            
        
        b,c,h,w = inputs.shape
        x_lengths = (inputs.mean(dim=(1,2)) > self.pad_val).long().sum(-1)
        x_mask = sequence_mask(x_lengths, max_length = w)[:,None,None,:].to(inputs.dtype)# (B,1,1,max_length), 0 is the padded place
        
        
        rec_loss = torch.abs(inputs.contiguous() - reconstructions.contiguous())
        if self.perceptual_weight > 0:
            p_loss = self.perceptual_loss(inputs.contiguous(), reconstructions.contiguous())
            # print(f"p_loss {p_loss}")
            rec_loss = rec_loss + self.perceptual_weight * p_loss
        else:
            p_loss = torch.tensor([0.0])

        nll_loss = rec_loss / torch.exp(self.logvar) + self.logvar
        weighted_nll_loss = nll_loss
        if weights is not None:
            weighted_nll_loss = weights*nll_loss
        weighted_nll_loss = torch.sum(weighted_nll_loss) / weighted_nll_loss.shape[0]
        nll_loss = torch.sum(nll_loss) / nll_loss.shape[0]
        kl_loss = posteriors.kl()
        kl_loss = torch.sum(kl_loss) / kl_loss.shape[0]

        # !!!!!!!!!!!!   use the following line to avoid discriminator fail  !!!!!!!!!!!!!
        reconstructions = reconstructions*x_mask + (1-x_mask)*self.pad_val
        # now the GAN part
        if optimizer_idx == 0:
            # generator update
            if cond is None:
                assert not self.disc_conditional
                logits_fake = self.discriminator(reconstructions.contiguous())
            else:
                assert self.disc_conditional
                logits_fake = self.discriminator(torch.cat((reconstructions.contiguous(), cond), dim=1))
            g_loss = -torch.mean(logits_fake)

            try:
                d_weight = self.calculate_adaptive_weight(nll_loss, g_loss, last_layer=last_layer)
            except RuntimeError:
                assert not self.training
                d_weight = torch.tensor(0.0)

            disc_factor = adopt_weight(self.disc_factor, global_step, threshold=self.discriminator_iter_start)
            loss = weighted_nll_loss + self.kl_weight * kl_loss + d_weight * disc_factor * g_loss

            log = {"{}/total_loss".format(split): loss.clone().detach().mean(),
                   "{}/logvar".format(split): self.logvar.detach(),
                   "{}/kl_loss".format(split): kl_loss.detach().mean(),
                   "{}/nll_loss".format(split): nll_loss.detach().mean(),
                   "{}/rec_loss".format(split): rec_loss.detach().mean(),
                   "{}/d_weight".format(split): d_weight.detach(),
                   "{}/disc_factor".format(split): torch.tensor(disc_factor),
                   "{}/g_loss".format(split): g_loss.detach().mean(),
                   }
            return loss, log

        if optimizer_idx == 1:
            # second pass for discriminator update
            if cond is None:
                logits_real = self.discriminator(inputs.contiguous().detach())
                logits_fake = self.discriminator(reconstructions.contiguous().detach())
            else:
                logits_real = self.discriminator(torch.cat((inputs.contiguous().detach(), cond), dim=1))
                logits_fake = self.discriminator(torch.cat((reconstructions.contiguous().detach(), cond), dim=1))

            disc_factor = adopt_weight(self.disc_factor, global_step, threshold=self.discriminator_iter_start)
            d_loss = disc_factor * self.disc_loss(logits_real, logits_fake)

            log = {"{}/disc_loss".format(split): d_loss.clone().detach().mean(),
                   "{}/logits_real".format(split): logits_real.detach().mean(),
                   "{}/logits_fake".format(split): logits_fake.detach().mean()
                   }
            return d_loss, log