File size: 6,183 Bytes
c0eac48
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b0132ea
c0eac48
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from einops import rearrange, repeat, reduce, pack, unpack

# from vector_quantize_pytorch import ResidualVQ

#Borrow from vector_quantize_pytorch

def log(t, eps = 1e-20):
    return torch.log(t.clamp(min = eps))

def gumbel_noise(t):
    noise = torch.zeros_like(t).uniform_(0, 1)
    return -log(-log(noise))

def gumbel_sample(
    logits,
    temperature = 1.,
    stochastic = False,
    dim = -1,
    training = True
):

    if training and stochastic and temperature > 0:
        sampling_logits = (logits / temperature) + gumbel_noise(logits)
    else:
        sampling_logits = logits

    ind = sampling_logits.argmax(dim = dim)

    return ind

class QuantizeEMAReset(nn.Module):
    def __init__(self, nb_code, code_dim, args):
        super(QuantizeEMAReset, self).__init__()
        self.nb_code = nb_code
        self.code_dim = code_dim
        self.mu = args.mu  ##TO_DO
        self.reset_codebook()

    def reset_codebook(self):
        self.init = False
        self.code_sum = None
        self.code_count = None
        self.register_buffer('codebook', torch.zeros(self.nb_code, self.code_dim, requires_grad=False))

    def _tile(self, x):
        nb_code_x, code_dim = x.shape
        if nb_code_x < self.nb_code:
            n_repeats = (self.nb_code + nb_code_x - 1) // nb_code_x
            std = 0.01 / np.sqrt(code_dim)
            out = x.repeat(n_repeats, 1)
            out = out + torch.randn_like(out) * std
        else:
            out = x
        return out

    def init_codebook(self, x):
        out = self._tile(x)
        self.codebook = out[:self.nb_code]
        self.code_sum = self.codebook.clone()
        self.code_count = torch.ones(self.nb_code, device=self.codebook.device)
        self.init = True

    def quantize(self, x, sample_codebook_temp=0.):
        # N X C -> C X N
        k_w = self.codebook.t()
        # x: NT X C
        # NT X N
        distance = torch.sum(x ** 2, dim=-1, keepdim=True) - \
                   2 * torch.matmul(x, k_w) + \
                   torch.sum(k_w ** 2, dim=0, keepdim=True)  # (N * L, b)

        # code_idx = torch.argmin(distance, dim=-1)

        code_idx = gumbel_sample(-distance, dim = -1, temperature = sample_codebook_temp, stochastic=True, training = self.training)

        return code_idx

    def dequantize(self, code_idx):
        x = F.embedding(code_idx, self.codebook)
        return x
    
    def get_codebook_entry(self, indices):
        return self.dequantize(indices).permute(0, 2, 1)

    @torch.no_grad()
    def compute_perplexity(self, code_idx):
        # Calculate new centres
        code_onehot = torch.zeros(self.nb_code, code_idx.shape[0], device=code_idx.device)  # nb_code, N * L
        code_onehot.scatter_(0, code_idx.view(1, code_idx.shape[0]), 1)

        code_count = code_onehot.sum(dim=-1)  # nb_code
        prob = code_count / torch.sum(code_count)
        perplexity = torch.exp(-torch.sum(prob * torch.log(prob + 1e-7)))
        return perplexity

    @torch.no_grad()
    def update_codebook(self, x, code_idx):
        code_onehot = torch.zeros(self.nb_code, x.shape[0], device=x.device) # nb_code, N * L
        code_onehot.scatter_(0, code_idx.view(1, x.shape[0]), 1)

        code_sum = torch.matmul(code_onehot, x) # nb_code, c
        code_count = code_onehot.sum(dim=-1) # nb_code

        out = self._tile(x)
        code_rand = out[:self.nb_code]

        # Update centres
        self.code_sum = self.mu * self.code_sum + (1. - self.mu) * code_sum
        self.code_count = self.mu * self.code_count + (1. - self.mu) * code_count

        usage = (self.code_count.view(self.nb_code, 1) >= 1.0).float()
        code_update = self.code_sum.view(self.nb_code, self.code_dim) / self.code_count.view(self.nb_code, 1)
        self.codebook = usage * code_update + (1-usage) * code_rand


        prob = code_count / torch.sum(code_count)
        perplexity = torch.exp(-torch.sum(prob * torch.log(prob + 1e-7)))

        return perplexity

    def preprocess(self, x):
        # NCT -> NTC -> [NT, C]
        # x = x.permute(0, 2, 1).contiguous()
        # x = x.view(-1, x.shape[-1])
        x = rearrange(x, 'n c t -> (n t) c')
        return x

    def forward(self, x, return_idx=False, temperature=0.):
        N, width, T = x.shape

        x = self.preprocess(x)
        if self.training and not self.init:
            self.init_codebook(x)

        code_idx = self.quantize(x, temperature)
        x_d = self.dequantize(code_idx)

        if self.training:
            perplexity = self.update_codebook(x, code_idx)
        else:
            perplexity = self.compute_perplexity(code_idx)

        commit_loss = F.mse_loss(x, x_d.detach()) # It's right. the t2m-gpt paper is wrong on embed loss and commitment loss.

        # Passthrough
        x_d = x + (x_d - x).detach()

        # Postprocess
        x_d = x_d.view(N, T, -1).permute(0, 2, 1).contiguous()
        code_idx = code_idx.view(N, T).contiguous()
        # print(code_idx[0])
        if return_idx:
            return x_d, code_idx, commit_loss, perplexity
        return x_d, commit_loss, perplexity
    
class QuantizeEMA(QuantizeEMAReset):
    @torch.no_grad()
    def update_codebook(self, x, code_idx):
        code_onehot = torch.zeros(self.nb_code, x.shape[0], device=x.device) # nb_code, N * L
        code_onehot.scatter_(0, code_idx.view(1, x.shape[0]), 1)

        code_sum = torch.matmul(code_onehot, x) # nb_code, c
        code_count = code_onehot.sum(dim=-1) # nb_code

        # Update centres
        self.code_sum = self.mu * self.code_sum + (1. - self.mu) * code_sum
        self.code_count = self.mu * self.code_count + (1. - self.mu) * code_count

        usage = (self.code_count.view(self.nb_code, 1) >= 1.0).float()
        code_update = self.code_sum.view(self.nb_code, self.code_dim) / self.code_count.view(self.nb_code, 1)
        self.codebook = usage * code_update + (1-usage) * self.codebook

        prob = code_count / torch.sum(code_count)
        perplexity = torch.exp(-torch.sum(prob * torch.log(prob + 1e-7)))

        return perplexity