-
Notifications
You must be signed in to change notification settings - Fork 7
/
Copy pathutils.py
177 lines (155 loc) · 6.31 KB
/
utils.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
import copy
import torch
from torchvision import datasets, transforms
from sampling import cifar_iid, svhn_iid, cifar_noniid_skew, svhn_noniid_skew, cifar100_noniid_skew, svhn_noniid_unequal, cifar10_noniid_unequal
def get_dataset(args):
""" Returns train and test datasets and a user group which is a dict where
the keys are the user index and the values are the corresponding data for
each of those users.
"""
if args.dataset == 'cifar-10':
data_dir = '../data/cifar-10/'
apply_transform = transforms.Compose(
[transforms.ToTensor(),
#transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])
train_dataset = datasets.CIFAR10(data_dir, train=True, download=True,
transform=apply_transform)
test_dataset = datasets.CIFAR10(data_dir, train=False, download=True,
transform=apply_transform)
# sample training data amongst users
if args.iid:
# Sample IID user data from Mnist
user_groups = cifar_iid(train_dataset, args.num_users)
else:
# Sample Non-IID user data from Mnist
if args.unequal:
# Chose uneuqal splits for every user
user_groups = cifar10_noniid_unequal(train_dataset, args.num_users)
else:
# Chose euqal splits for every user
user_groups = cifar_noniid_skew(train_dataset, args.num_users)
elif args.dataset == 'svhn':
data_dir = '../data/svhn/'
apply_transform = transforms.Compose(
[transforms.ToTensor(),
#transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])
train_dataset = datasets.SVHN(data_dir, split='train', download=True,
transform=apply_transform)
test_dataset = datasets.SVHN(data_dir, split='test', download=True,
transform=apply_transform)
# sample training data amongst users
if args.iid:
# Sample IID user data from Mnist
user_groups = svhn_iid(train_dataset, args.num_users)
else:
# Sample Non-IID user data from Mnist
if args.unequal:
# Chose uneuqal splits for every user
user_groups = svhn_noniid_unequal(train_dataset, args.num_users)
else:
# Chose euqal splits for every user
#print(train_dataset.labels)
user_groups = svhn_noniid_skew(train_dataset, args.num_users)
elif args.dataset == 'cifar-100':
data_dir = '../data/cifar-100/'
apply_transform = transforms.Compose(
[transforms.ToTensor(),
#transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])
train_dataset = datasets.CIFAR100(data_dir, train=True, download=True,
transform=apply_transform)
test_dataset = datasets.CIFAR100(data_dir, train=False, download=True,
transform=apply_transform)
# sample training data amongst users
if args.iid:
# Sample IID user data from Mnist
user_groups = cifar_iid(train_dataset, args.num_users)
else:
# Sample Non-IID user data from Mnist
if args.unequal:
# Chose uneuqal splits for every user
raise NotImplementedError()
else:
# Chose euqal splits for every user
user_groups = cifar100_noniid_skew(train_dataset, args.num_users)
return train_dataset, test_dataset, user_groups
# FedAvg
def average_weights(w):
"""
Returns the average of the weights.
"""
w_avg = copy.deepcopy(w[0])
for key in w_avg.keys():
print(key)
for i in range(1, len(w)):
w_avg[key] += w[i][key]
w_avg[key] = torch.div(w_avg[key], len(w))
return w_avg
# # FedAvg unequal
def average_weights_unequal(w, idx_num):
"""
Returns the average of the weights.
"""
w_avg = copy.deepcopy(w[0])
for key in w_avg.keys():
print(key)
w_avg[key] = w_avg[key] * float(idx_num[0]*len(w)/sum(idx_num))
for i in range(1, len(w)):
w_avg[key] += w[i][key] * float(idx_num[i]*len(w)/sum(idx_num))
w_avg[key] = torch.div(w_avg[key], len(w))
return w_avg
# SFAT
def average_weights_alpha(w, lw, idx, p):
"""
Returns the weighted average of the weights.
"""
w_avg = copy.deepcopy(w[0])
for key in w_avg.keys():
cou = 0
if (lw[0] >= idx):
w_avg[key] = w_avg[key] * p
for i in range(1, len(w)):
if (lw[i] >= idx) and (('bn' not in key)):
w_avg[key] = w_avg[key] + w[i][key] * p
else:
cou += 1
w_avg[key] = w_avg[key] + w[i][key]
w_avg[key] = torch.div(w_avg[key], cou+(len(w)-cou)*p)
return w_avg
# # SFAT unequal
def average_weights_alpha_unequal(w, lw, idx, p, idx_num):
"""
Returns the weighted average of the weights.
"""
w_avg = copy.deepcopy(w[0])
for key in w_avg.keys():
cou = 0
if (lw[0] >= idx):
w_avg[key] = w_avg[key] * p * float(idx_num[0]*len(w)/sum(idx_num))
else:
w_avg[key] = w_avg[key] * float(idx_num[0]*len(w)/sum(idx_num))
for i in range(1, len(w)):
if (lw[i] >= idx) and (('bn' not in key)):
w_avg[key] = w_avg[key] + w[i][key] * p * float(idx_num[i]*len(w)/sum(idx_num))
else:
cou += 1
w_avg[key] = w_avg[key] + w[i][key] * float(idx_num[i]*len(w)/sum(idx_num))
w_avg[key] = torch.div(w_avg[key], cou+(len(w)-cou)*p)
return w_avg
def exp_details(args):
print('\nExperimental details:')
print(f' Model : {args.model}')
print(f' Optimizer : {args.optimizer}')
print(f' Learning : {args.lr}')
print(f' Global Rounds : {args.epochs}\n')
print(' Federated parameters:')
if args.iid:
print(' IID')
else:
print(' Non-IID')
print(f' Fraction of users : {args.frac}')
print(f' Local Batch size : {args.local_bs}')
print(f' Local Epochs : {args.local_ep}\n')
return