-
Notifications
You must be signed in to change notification settings - Fork 0
/
Main.py
126 lines (97 loc) · 3.53 KB
/
Main.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 InitialStartPytorch as IS
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import torch
from torch import nn
from torch.utils.data import Dataset
from torch.utils.data import DataLoader
#-----[Settings]-----
learning_rate = 1e-3
batch_size = 400
epochs = 200
#--------------------
testIMG = open('/Users/MNIST Data/train-images-idx3-ubyte', 'rb')
testLB = open('/Users/MNIST Data/train-labels.idx1-ubyte', 'rb')
trainingSize = 5000
testingSize = 5000
dataSetSize = trainingSize + testingSize
dataSet = np.zeros((dataSetSize, 784))
labelValueSetSize = dataSetSize
labelValueSet = np.zeros((labelValueSetSize))
labelSetSize = dataSetSize
labelSet = np.zeros((labelSetSize, 10))
IS.STEP1(testIMG, testLB)
IS.STEP2(labelValueSetSize, labelValueSet, testLB)
IS.STEP3(dataSetSize, dataSet, testIMG)
IS.STEP4(labelSetSize, labelSet, labelValueSet)
testIMG.close()
testLB.close()
trainingPixels = dataSet[0:trainingSize]
trainingValueLabels = labelValueSet[0:trainingSize]
trainingLabels = labelSet[0:trainingSize]
testingPixels = dataSet[trainingSize:dataSetSize]
testingValuesLabels = labelValueSet[trainingSize:dataSetSize]
testingLabels = labelSet[trainingSize:dataSetSize]
T_training = torch.from_numpy(trainingPixels)
T_testing = torch.from_numpy(testingPixels)
T_trainingLabel = torch.from_numpy(trainingLabels)
T_testingLabel = torch.from_numpy(testingValuesLabels)
#trainingPixels.tofile("selectedTraining.csv", sep=",")
#testingPixels.tofile("selectedTesting.csv", sep=",")
class CustomDataSet(Dataset):
def __init__(self, dataSet, labelSet) -> None:
self.data = dataSet .to(torch.float)
self.labels = labelSet .to(torch.float)
def __len__(self):
return np.shape(self.data)[0]
def __getitem__(self, idx):
return self.data[idx], self.labels[idx]
#1e-3, 400, 200
trainingOBJ = CustomDataSet(T_training, T_trainingLabel)
testingOBJ = CustomDataSet(T_testing, T_testingLabel)
trainDataLoad = DataLoader(trainingOBJ, batch_size, shuffle=True)
testDataLoad = DataLoader(testingOBJ, batch_size, shuffle=True)
class NeuralNetwork(nn.Module):
def __init__(self, *args, **kwargs) -> None:
super().__init__(*args, **kwargs)
self.nerualNetwork = nn.Sequential(
nn.Linear(28*28,16),
nn.Sigmoid(),
nn.Linear(16,16),
nn.Sigmoid(),
nn.Linear(16,10),
nn.Sigmoid()
)
def forward(self, x):
return self.nerualNetwork(x.to(torch.float32))
loss_fn = nn.MSELoss(reduction="sum")
nerualNetwork1 = NeuralNetwork()
optimizer = torch.optim.SGD(nerualNetwork1.parameters(), learning_rate)
def trainingFunc(dataLoad, model, lossFunc, optimizerFunc):
model.train()
for batch, (x,y) in enumerate(dataLoad):
pred = model(x)
loss = lossFunc(pred, y)
loss.backward()
optimizerFunc.step()
optimizerFunc.zero_grad()
loss = loss.item()
print(loss)
def testingFunc(dataLoad, model):
correct = 0
model.eval()
with torch.no_grad():
for X, y in dataLoad:
pred = model(X)
#test = (pred.argmax(1) == y).type(torch.float).sum().item()
#print(pred)
test = pred.argmax(1)
correct += (pred.argmax(1) == y).type(torch.float).sum().item()
correct /= len(dataLoad.dataset)
print(f"Accuracy: {(100*correct)}%\n")
for t in range(epochs):
print(f"epoch: {t}\n")
trainingFunc(trainDataLoad, nerualNetwork1, loss_fn, optimizer)
testingFunc(testDataLoad, nerualNetwork1)
#---------