-
Notifications
You must be signed in to change notification settings - Fork 11
/
Copy pathmain_onoszko_2021.py
126 lines (107 loc) · 4.18 KB
/
main_onoszko_2021.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
import math
import torch
import torch.nn as nn
import torch.nn.functional as F
from torchvision.transforms import Compose, Normalize, RandomVerticalFlip
from gossipy.core import AntiEntropyProtocol, CreateModelMode, StaticP2PNetwork
from gossipy.data import DataDispatcher
from gossipy.model import TorchModel
from gossipy.data.handler import ClassificationDataHandler
from gossipy.model.handler import TorchModelHandler
from gossipy.node import PENSNode
from gossipy.simul import GossipSimulator, SimulationReport
from gossipy.data import get_CIFAR10
from gossipy.utils import plot_evaluation
# AUTHORSHIP
__version__ = "0.0.1"
__author__ = "Mirko Polato"
__copyright__ = "Copyright 2022, gossipy"
__license__ = "MIT"
__maintainer__ = "Mirko Polato, PhD"
__email__ = "[email protected]"
__status__ = "Development"
#
class CIFAR10Net(TorchModel):
def __init__(self):
super().__init__()
self.conv1 = nn.Conv2d(3, 32, 3)
self.pool = nn.MaxPool2d(2, 2)
self.conv2 = nn.Conv2d(32, 64, 3)
self.conv3 = nn.Conv2d(64, 64, 3)
self.fc1 = nn.Linear(64 * 2 * 2, 64)
self.fc2 = nn.Linear(64, 10)
def init_weights(self, *args, **kwargs) -> None:
def _init_weights(m: nn.Module):
if isinstance(m, nn.Linear) or isinstance(m, nn.Conv2d):
nn.init.xavier_uniform_(m.weight)
nn.init.zeros_(m.bias)
#self.apply(_init_weights)
pass
def forward(self, x):
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = self.pool(F.relu(self.conv3(x)))
x = x.view(-1, 64 * 2 * 2)
x = F.relu(self.fc1(x))
x = self.fc2(x)
return x
def __repr__(self) -> str:
return "CIFAR10Net(size=%d)" %self.get_size()
class CustomDataDispatcher(DataDispatcher):
def assign(self, seed: int = 42) -> None:
self.tr_assignments = [[] for _ in range(self.n)]
self.te_assignments = [[] for _ in range(self.n)]
n_ex = self.data_handler.size()
ex_x_user = math.ceil(n_ex / self.n)
for idx, i in enumerate(range(0, n_ex, ex_x_user)):
self.tr_assignments[idx] = list(range(i, min(i + ex_x_user, n_ex)))
if self.eval_on_user:
n_eval_ex = self.data_handler.eval_size()
eval_ex_x_user = math.ceil(n_eval_ex / self.n)
for idx, i in enumerate(range(0, n_eval_ex, eval_ex_x_user)):
self.te_assignments[idx] = list(range(i, min(i + eval_ex_x_user, n_eval_ex)))
# Dataset loading
transform = Compose([Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])])
rotate180 = RandomVerticalFlip(p=1.0)
train_set, test_set = get_CIFAR10()
Xtr, ytr = (transform(train_set[0]), train_set[1])
Xte, yte = (transform(test_set[0]), test_set[1])
rotated_Xtr, rotated_Xte = rotate180(Xtr), rotate180(Xte)
half = train_set[0].size(0) // 2
train_set = torch.cat((Xtr[:half], rotated_Xtr[half:])), torch.cat((ytr[:half], ytr[half:]))
half_te = test_set[0].size(0) // 2
test_set = torch.cat((Xte[:half_te], rotated_Xte[half_te:])), torch.cat((yte[:half_te], yte[half_te:]))
data_handler = ClassificationDataHandler(train_set[0], train_set[1],
test_set[0], test_set[1])
data_dispatcher = CustomDataDispatcher(data_handler, n=5, eval_on_user=False, auto_assign=True)
nodes = PENSNode.generate(
data_dispatcher=data_dispatcher,
p2p_net=StaticP2PNetwork(5),
model_proto=TorchModelHandler(
net=CIFAR10Net(),
optimizer= torch.optim.SGD,
optimizer_params = {
"lr": 0.01,
"weight_decay": 0.001
},
criterion = F.cross_entropy,
create_model_mode= CreateModelMode.MERGE_UPDATE,
batch_size= 8,
local_epochs= 3),
round_len=100,
sync=False,
n_sampled= 10,
m_top= 2,
step1_rounds= 100)
simulator = GossipSimulator(
nodes = nodes,
data_dispatcher=data_dispatcher,
delta=100,
protocol=AntiEntropyProtocol.PUSH,
sampling_eval=0.1
)
report = SimulationReport()
simulator.add_receiver(report)
simulator.init_nodes(seed=42)
simulator.start(n_rounds=500)
plot_evaluation([[ev for _, ev in report.get_evaluation(False)]], "Overall test results")