-
Notifications
You must be signed in to change notification settings - Fork 0
/
losses.py
192 lines (156 loc) · 7.13 KB
/
losses.py
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
import torch
import torch.nn as nn
import torch.nn.functional as F
def get_pesTriplet(embeddings,labels,centers,lambd):
C = centers.size()[0]
n = embeddings.size()[0]
labelSet = torch.arange(C)
pes_triplet = []
for i in range(n):
embedding = embeddings[i]
label = labels[i]
dis_to_centers = (embedding.repeat((C,1)) - centers).pow(2).sum(1)
ap_dis = dis_to_centers[label]
loss_val = lambd + ap_dis - dis_to_centers
#print(loss_val)
mask = loss_val.gt(0)
mask[label] = 0
if torch.sum(mask) > 0:
neg = labelSet[mask]
pes_triplet += [[i,label,ex_label] for ex_label in neg]
if len(pes_triplet)==0:
return None
return torch.LongTensor(pes_triplet).cuda() if embeddings.is_cuda else torch.LongTensor(pes_triplet)
def get_minTriplet(embeddings,labels,centers,lambd):
C = centers.size()[0]
n = embeddings.size()[0]
labelSet = torch.arange(C)
pes_triplet = []
for i in range(n):
embedding = embeddings[i]
label = labels[i]
dis_to_centers = (embedding.repeat((C,1)) - centers).pow(2).sum(1)
ap_dis = dis_to_centers[label]
loss_val = lambd + ap_dis - dis_to_centers
#print(loss_val)
argmin = torch.argmax(loss_val)
if argmin!=label and loss_val[argmin]>0:
pes_triplet += [[i,label,argmin]]
if len(pes_triplet)==0:
return None
return torch.LongTensor(pes_triplet).cuda() if embeddings.is_cuda else torch.LongTensor(pes_triplet)
class ContrastiveLoss(nn.Module):
"""
Contrastive loss
Takes embeddings of two samples and a target label == 1 if samples are from the same class and label == 0 otherwise
"""
def __init__(self, margin):
super(ContrastiveLoss, self).__init__()
self.margin = margin
def forward(self, output1, output2, target, size_average=True):
distances = (output2 - output1).pow(2).sum(1) # squared distances
losses = 0.5 * (target.float() * distances +
(1 + -1 * target).float() * F.relu(self.margin - distances.sqrt()).pow(2))
return losses.mean() if size_average else losses.sum()
class TripletLoss(nn.Module):
"""
Triplet loss
Takes embeddings of an anchor sample, a positive sample and a negative sample
"""
def __init__(self, margin):
super(TripletLoss, self).__init__()
self.margin = margin
def forward(self, anchor, positive, negative, size_average=True):
distance_positive = (anchor - positive).pow(2).sum(1) # .pow(.5)
distance_negative = (anchor - negative).pow(2).sum(1) # .pow(.5)
losses = F.relu(distance_positive - distance_negative + self.margin)
return losses.mean() if size_average else losses.sum()
class OnlineContrastiveLoss(nn.Module):
"""
Online Contrastive loss
Takes a batch of embeddings and corresponding labels.
Pairs are generated using pair_selector object that take embeddings and targets and return indices of positive
and negative pairs
"""
def __init__(self, margin, pair_selector):
super(OnlineContrastiveLoss, self).__init__()
self.margin = margin
self.pair_selector = pair_selector
def forward(self, embeddings, target):
positive_pairs, negative_pairs = self.pair_selector.get_pairs(embeddings, target)
if embeddings.is_cuda:
positive_pairs = positive_pairs.cuda()
negative_pairs = negative_pairs.cuda()
positive_loss = (embeddings[positive_pairs[:, 0]] - embeddings[positive_pairs[:, 1]]).pow(2).sum(1)
negative_loss = F.relu(
self.margin - (embeddings[negative_pairs[:, 0]] - embeddings[negative_pairs[:, 1]]).pow(2).sum(
1).sqrt()).pow(2)
loss = torch.cat([positive_loss, negative_loss], dim=0)
return loss.mean()
class OnlineCenterLoss(nn.Module):
def __init__(self,lambd):
super(OnlineCenterLoss,self).__init__()
self.lambd = lambd
def forward(self, embeddings, targets, centers):
triplets = get_pesTriplet(embeddings,targets,centers,self.lambd) # A:embedding P:center N:center(false)
if triplets is None:
zero = torch.Tensor([0.])
zero.requires_grad_()
return zero
#print(triplets)
ap_distances = (embeddings[triplets[:,0]] - centers[triplets[:,1]]).pow(2).sum(1)
an_distances = (embeddings[triplets[:,0]] - centers[triplets[:,2]]).pow(2).sum(1)
losses = F.relu(ap_distances - an_distances + self.lambd)
return losses.mean()
class OnlineCenterLossV2(nn.Module):
def __init__(self,lambd):
super(OnlineCenterLossV2,self).__init__()
self.lambd = lambd
def forward(self, embeddings, targets, centers):
triplets = get_minTriplet(embeddings,targets,centers,self.lambd) # A:embedding P:center N:center(false)
if triplets is None:
zero = torch.Tensor([0.])
zero.requires_grad_()
return zero
#print(triplets)
ap_distances = (embeddings[triplets[:,0]] - centers[triplets[:,1]]).pow(2).sum(1)
an_distances = (embeddings[triplets[:,0]] - centers[triplets[:,2]]).pow(2).sum(1)
losses = F.relu(ap_distances - an_distances + self.lambd)
return losses.mean()
class OnlineTripletLoss(nn.Module):
"""
Online Triplets loss
Takes a batch of embeddings and corresponding labels.
Triplets are generated using triplet_selector object that take embeddings and targets and return indices of
triplets
"""
def __init__(self, margin, triplet_selector):
super(OnlineTripletLoss, self).__init__()
self.margin = margin
self.triplet_selector = triplet_selector
def forward(self, embeddings, target):
triplets = self.triplet_selector.get_triplets(embeddings, target)
if embeddings.is_cuda:
triplets = triplets.cuda()
ap_distances = (embeddings[triplets[:, 0]] - embeddings[triplets[:, 1]]).pow(2).sum(1) # .pow(.5)
an_distances = (embeddings[triplets[:, 0]] - embeddings[triplets[:, 2]]).pow(2).sum(1) # .pow(.5)
losses = F.relu(ap_distances - an_distances + self.margin)
return losses.mean(), len(triplets)
class LiftedEmbeddingLoss(nn.Module):
def __init__(self, margin, triplet_selector):
super(LiftedEmbeddingLoss, self).__init__()
self.margin = margin
self.triplet_selector = triplet_selector
def forward(self, embeddings, target):
triplets = self.triplet_selector.get_triplets(embeddings, target)
losses = 0
for i in range(len(triplets)):
triplet = triplets[i]
ap_distances = (embeddings[triplet[0]] - embeddings[triplet[1]]).pow(2).sum(1) # .pow(.5)
an_distances = (embeddings[triplet[0]] - embeddings[triplet[2]]).pow(2).sum(1) # .pow(.5)
ap_exp_sum = torch.exp(ap_distances).sum()
an_exp_sum = torch.exp(self.margin - an_distances).sum()
ap = torch.log(ap_exp_sum)
an = torch.log(an_exp_sum)
losses += F.relu(ap+an)
return losses, len(triplets)