-
Notifications
You must be signed in to change notification settings - Fork 0
/
parameterFinder.py
285 lines (211 loc) · 10.8 KB
/
parameterFinder.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
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
import torch.cuda
import torch
import numpy as np
import sys
import torch.nn as nn
import time
from sklearn.preprocessing import FunctionTransformer
import wandb
import torch.nn.functional as F
#-------------------------------------------------------------------------
torch.cuda.set_device(0)
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
np.set_printoptions(threshold=sys.maxsize)
torch.set_printoptions(threshold=10_000)
#-------------------------------------------------------------------------
ga = np.load("TestingGammaNew.npy", allow_pickle=True)
no = np.load("TestingnurtronNew.npy", allow_pickle=True)
transformer = FunctionTransformer(np.log1p, validate=True)
ga = transformer.transform(ga)
no = transformer.transform(no)
ga = torch.Tensor(ga).cuda()
no = torch.Tensor(no).cuda()
input_data = torch.Tensor(np.load("Input_new.npy", allow_pickle=True)).cuda()
predict_data = torch.Tensor(np.load("label_new.npy", allow_pickle=True)).cuda()
input_data = input_data.type(torch.FloatTensor)
predict_data = predict_data.type(torch.LongTensor)
top20= np.float64([99.9])
gacount = 0
nocount = 0
correctCount =0
wrongCount =0
input_size = 248
wrongCount2 =0
counter1 = 0
counter2 = 0
best = np.float64([99]) #antioverfit
a = np.float64([99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99]) #antioverfit
BATCH_SIZE = 100
num_epochs = 5000000
print_interval = 3000
testing_loss = 0.0
counter = 0
DROPOUT1 = .3
DROPOUT2 = .1
DROPOUT3 = .3
num_layers1 = 1
num_layers2 = 1
num_layers3 = 2
hidden_size = 100
class RNN(nn.Module):
def __init__(self, input_size, hidden_size, num_layers1, num_layers2, num_layers3, num_classes):
super(RNN, self).__init__()
self.num_layers1 = num_layers1
self.num_layers2 = num_layers2
self.num_layers3 = num_layers3
self.hidden_size = hidden_size
self.drop1 = torch.nn.Dropout(DROPOUT1)
self.lstm1 = nn.LSTM(input_size, hidden_size, num_layers1, batch_first=True , )
self.drop2 = torch.nn.Dropout(DROPOUT2)
self.lstm2 = nn.LSTM(hidden_size, hidden_size, num_layers2, batch_first=True , )
self.drop3 = torch.nn.Dropout(DROPOUT3)
self.lstm3 = nn.LSTM(hidden_size, hidden_size, num_layers3, batch_first=True , )
self.fc = nn.Linear(hidden_size, num_classes)
def forward(self, x):
h01 = torch.zeros(self.num_layers1, x.size(0), self.hidden_size).to(device)
c01 = torch.zeros(self.num_layers1, x.size(0), self.hidden_size).to(device)
h02 = torch.zeros(self.num_layers2, x.size(0), self.hidden_size).to(device)
c02 = torch.zeros(self.num_layers2, x.size(0), self.hidden_size).to(device)
h03 = torch.zeros(self.num_layers3, x.size(0), self.hidden_size).to(device)
c03 = torch.zeros(self.num_layers3, x.size(0), self.hidden_size).to(device)
out, _ = self.lstm1(x, (h01,c01))
out = self.drop1(out)
out, _ = self.lstm2(out, (h02,c02))
out = self.drop2(out)
out, _ = self.lstm3(out, (h03,c03))
out = self.drop3(out)
out = out[:, -1, :]
out = self.fc(out)
return out
input_size = 248
num_classes = 2
lr = [ 0.0005, 0.001, .009]
betasright = [ 0.998, 0.997, .996]
betasleft = [0.6, 0.5, 0.4]
wd = [1e-10, 1e-11, 1e-12]
eps = [1e-06, 1e-07, 1e-08]
config_dict = {
"lr" : [ 0.0005, 0.001, .009],
"betasright" : [ 0.998, 0.997, .996],
"betasleft" : [0.6, 0.5, 0.4],
"wd" : [1e-10, 1e-11, 1e-12],
"eps" : [1e-06, 1e-07, 1e-08]
}
model = RNN(input_size, hidden_size, num_layers1, num_layers2, num_layers3 , num_classes).to(device)
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.RAdam(model.parameters(), lr=0.001, betas=(0.6, 0.997), eps=1e-06, weight_decay=1e-11 )
model.train()
model.to(device)
input_data.to(device)
predict_data.to(device)
PATH = "model.pt"
torch.save({'model_state_dict1': model.state_dict(),}, PATH)
start_time = time.time()
overallacc = 0
for WD in wd:
for EPS in eps:
for LR in lr:
for BETASRIGHT in betasright:
for BETASLEFT in betasleft:
counter = counter + 1
checkpoint = torch.load(PATH)
model.load_state_dict(checkpoint['model_state_dict1'])
optimizer = torch.optim.RAdam(model.parameters(), lr=LR, betas=(BETASLEFT, BETASRIGHT), eps=EPS, weight_decay=WD )
a = np.float64([99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99]) #antioverfit
testing_loss = 0.0
model.train().cuda()
#wandb.init(project="my-test-project", entity="samuelfipps" , config=optimizer)
for epoch in range(num_epochs):
start_time = time.time()
if(overallacc < a[15]): # part of anti overfit
train_loss = 0.0
testing_loss = 0.0
model.train()
for i in (range(0, len(input_data), BATCH_SIZE)):
batch_X = input_data[i:i+BATCH_SIZE]
batch_y = predict_data[i:i+BATCH_SIZE]
batch_X = batch_X.to(device) #gpu # input data here!!!!!!!!!!!!!!!!!!!!!!!!!!
batch_y = batch_y.to(device) #gpu # larget data here!!!!!!!!!!!!!!!!!!!!!!!!!!
batch_X = batch_X.reshape(-1, 1, input_size).to(device)
output = model(batch_X)
loss = criterion(output, batch_y).to(device)
#wandb.log({"loss": loss})
optimizer.zero_grad()
loss.backward()
optimizer.step()
print(f"Epoch [{epoch + 1}/{num_epochs}], " f"Step [{i + 1}/{len(input_data)}], " f"Loss: {loss.item():.4f}")
secondTime = time. time()
print("total time for 1 epoch: ", secondTime-start_time)
model.eval()
wrongCount = 0
counter1 = 0
rightcounter = 0
with torch.no_grad():
for i in (range(0, len(ga), BATCH_SIZE)):
batch_X = ga[i:i+BATCH_SIZE]
batch_X = batch_X.reshape(-1, 1, input_size).to(device)
output = model(batch_X)
_, pred = torch.max(output, dim=1)
for i in pred:
counter1 = counter1 + 1
if i !=0:
wrongCount = wrongCount + 1
else:
rightcounter = rightcounter + 1
#neutron pulse tester
model.eval()
wrongCount2 = 0
counter2 = 0
rightcounter2 =0
with torch.no_grad():
for i in (range(0, len(no), BATCH_SIZE)):
batch_X = no[i:i+BATCH_SIZE]
batch_X = batch_X.reshape(-1, 1, input_size).to(device)
output = model(batch_X)
_, pred = torch.max(output, dim=1)
for i in pred:
counter2 = counter2 + 1
if i ==0:
wrongCount2 = wrongCount2 + 1
else:
rightcounter2 = rightcounter2 + 1
print()
print(f"Accuracy for No pluse: {wrongCount2 / (counter2) * 100:.4f}%")
print(f"Accuracy for ga pluse: {wrongCount / (counter1) * 100:.4f}%")
print("Wrong count for No pluse: ", wrongCount2)
print("Wrong count for ga pluse: ", wrongCount)
print("Right count for No pluse: ", rightcounter)
print("Right count for ga pluse: ", rightcounter2)
accuracy2 = wrongCount2
accuracy = wrongCount
wrongcountoverall = wrongCount2+ wrongCount
overallacc = (accuracy2+accuracy) /(counter1 + counter2)
#wandb.log({"Testing acc": overallacc*100})
#wandb.log({"Wrong count Overall": wrongcountoverall})
print(f"Accuracy overall : {overallacc *100:.4f}%")
print("Wrong count overall: ", wrongcountoverall)
trainttime = time. time()
print("total time for testing: ", trainttime-start_time)
print()
overallacc = overallacc*100
a = np.insert(a,0,overallacc) # part of anti overfit
a = np.delete(a,22)
#print("top20 list = " ,top20)
if epoch == 0 and counter == 1:
top20[0] = overallacc
if epoch == 0 and counter != 1:
top20 = np.append(top20, overallacc)
elif overallacc < top20[counter-1]:
top20[counter-1] = overallacc
#wandb.finish(exit_code=0)
torch.save(model, "models/GRUModel.pth")
print(optimizer)
print("lr= ", LR, "betaright= ", BETASRIGHT, "betaleft= ", BETASLEFT, " wd= ", WD, "eps= ", EPS)
print("round: ", counter, " out of 243")
best = np.append(best, overallacc)
print(best)
print("the top ones are: ")
print(top20)
#average = np.average(top20)
#print("Average = ")
#print(average*100)