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
| import torch
from torch.utils.data import Dataset, DataLoader
import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
import time
class Iris:
def __init__(self, sepal_length, sepal_width, petal_length, petal_width, species):
self.sepal_length = sepal_length
self.sepal_width = sepal_width
self.petal_length = petal_length
self.petal_width = petal_width
self.species = species
def feature(self):
return [self.sepal_length, self.sepal_width, self.petal_length, self.petal_width]
def label(self):
if self.species == 'Iris-setosa':
return 0
elif self.species == 'Iris-versicolor':
return 1
elif self.species == 'Iris-virginica':
return 2
class IrisDataset(Dataset):
def __init__(self, features, labels):
self.features = features
self.labels = labels
def __len__(self):
return len(self.features)
def __getitem__(self, index):
item_feature = self.features[index]
item_label = self.labels[index]
return item_feature, item_label
class Model(torch.nn.Module):
def __init__(self, n_feature, n_hidden, n_label):
super(Model, self).__init__()
self.hidden = torch.nn.Linear(n_feature, n_hidden)
self.output = torch.nn.Linear(n_hidden, n_label)
def forward(self, x):
x = torch.nn.functional.relu(self.hidden(x))
x = torch.nn.functional.softmax(self.output(x), dim=1)
return x
def get_data():
tic = time.time()
print('读取cvs文件')
csv_path = '/Users/bakako/Downloads/archive/Iris.csv'
datasource = pd.read_csv(csv_path)
datasource_len = len(datasource)
iris_features = np.zeros((datasource_len, 4), dtype=float)
iris_labels = []
print('构造数据结构')
for i in range(datasource_len):
iris = Iris(datasource.SepalLengthCm[i], datasource.SepalWidthCm[i],
datasource.PetalLengthCm[i], datasource.PetalWidthCm[i],
datasource.Species[i])
iris_features[[i], :] = iris.feature()
iris_labels.append(iris.label())
train_features, \
test_features, \
train_labels, \
test_labels = train_test_split(iris_features, iris_labels, test_size=0.2)
print('构造训练集')
train_dataset = IrisDataset(torch.FloatTensor(train_features),
torch.LongTensor(train_labels))
test_features = torch.FloatTensor(test_features)
test_labels = torch.LongTensor(test_labels)
toc = time.time()
print('Loading Time: ' + str(1000 * (toc - tic)) + 'ms')
print('')
return train_dataset, test_features, test_labels
def train_model(dataset, epochs, gpu):
tic = time.time()
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model = Model(n_feature=4, n_hidden=40, n_label=3)
if gpu:
model.to(device)
optimizer = torch.optim.SGD(model.parameters(), lr=0.02)
loss_func = torch.nn.CrossEntropyLoss()
batch_size = 5
dataloader = DataLoader(dataset, batch_size=batch_size, shuffle=True, num_workers=0, pin_memory=True)
for epoch in range(epochs):
for step, (features, labels) in enumerate(dataloader):
if gpu:
features = features.to(device)
labels = labels.to(device)
output = model(features)
loss = loss_func(output, labels)
optimizer.zero_grad()
loss.backward()
optimizer.step()
print('Epoch: ', epoch + 1, '| Step: ', step + 1, '| Loss: ', loss.item())
toc = time.time()
print('训练结束')
print('Train time:' + str(1000 * (toc - tic)) + 'ms')
print('')
return model
if __name__ == '__main__':
gpu = False
train_dataset, test_features, test_labels = get_data()
model = train_model(train_dataset, 100, gpu)
result = model(test_features)
prediction = torch.max(result, 1)[1]
correct = (prediction == test_labels).sum().item()
accuracy = float(correct) / float(len(test_features))
print("莺尾花预测准确率:", accuracy)
|