-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathMnist.py
212 lines (191 loc) · 7.85 KB
/
Mnist.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
# import torch
# import torch.nn as nn
# from torch.utils.data import DataLoader
# from torchvision import datasets, transforms
# import matplotlib.pyplot as plt
# import torch.nn.functional as F
#
# batch_size = 100
# lr = 0.01
# num_epoch = 1
#
# # 数据归一化、标准化
# data_form = transforms.Compose([transforms.ToTensor(), transforms.Normalize([0], [1])])
#
# # 从torchvision获取数据集
# train_dataset = datasets.MNIST(root="./MNIST_data", train=True, transform=data_form, download=True)
# test_dataset = datasets.MNIST(root="./MNIST_data", train=True, transform=data_form, download=False)
#
# train_loader = DataLoader(train_dataset, batch_size, shuffle=True)
# test_loader = DataLoader(test_dataset, batch_size, shuffle=False)
#
# # print(train_dataset.train_data.size())
# # print(train_dataset.train_labels.size())
#
# class Net(nn.Module):
# def __init__(self, in_dim, n_hidden1, n_hidden2, out_dim):
# super(Net, self).__init__()
# # 初始输入、隐藏、输出层
# # Sequential 按顺序执行，先liner然后relue; true liner的输出不保留，直接被relu计算覆盖
# self.layer1 = nn.Sequential(nn.Linear(in_dim, n_hidden1), nn.ReLU(inplace=True))
# self.layer2 = nn.Sequential(nn.Linear(n_hidden1, n_hidden2), nn.ReLU(inplace=True))
# self.layer3 = nn.Sequential(nn.Linear(n_hidden2, out_dim))
#
# def forward(self, x):
# y1 = self.layer1(x)
# y2 = self.layer2(y1)
# y3 = self.layer3(y2)
#
# return y3
#
# net = Net(784, 256, 128, 10)
#
# if torch.cuda.is_available():
# net = net.cuda()
#
# loss_fn = nn.CrossEntropyLoss()
# optimizer = torch.optim.SGD(net.parameters(), lr=lr)
#
# ##标记训练数据
# net.train()
# for epoch in range(num_epoch):
# for i, (img, label) in enumerate(train_loader):
# # print(img.size())
# img = img.reshape(img.size(0), -1) # 形状转换
#
# if torch.cuda.is_available():
# img = img.cuda()
# label = label.cuda()
#
# out = net(img)
#
# loss_ = loss_fn(out, label) # [100,10]
# optimizer.zero_grad()
# loss_.backward()
# optimizer.step()
#
# if i % 10 == 0:
# print("epoch:{},i:{},loss{:,3}".format(epoch, i, loss_.item())) # loss{:,3}三位精度
#
# #评估模型
# net.eval()
# eval_loss = 0.0#所有数据损失
# evl_acc = 0
# for date in test_loader:
# img,label = date
# img = img.reshape(img.size(0),-1)
# if torch.cuda.is_available():
# img = img.cuda()
# label = label.cuda()# [100,1]
# out = net(img)
# loss_ = loss_fn(out, label) # [100,10]
#
# eval_loss += loss_.item() * label.size()
# #计算精度
# max_out = torch.argmax(out,1)#取最大值的索引 1代表轴
#
# acc = (label == max_out).sum()
# evl_acc += acc.item()
#
# print(label)
# print(torch.argmax(out,1))
# print("test_loss:{:.3},test_acc:{:.3}".format(
# eval_loss / (len(test_dataset)),
# evl_acc / (len(test_dataset))
# ))
#
# '''老师的代码'''
import torch
from torch import nn, optim
from torch.utils.data import DataLoader
from torchvision import datasets, transforms
import torch.nn.functional as F
batch_size = 100
learning_rate = 0.01
num_epoches = 5
# 数据预处理：
# transforms.ToTensor()将图片转换成PyTorch中处理的对象Tensor,并且进行标准化（数据在0~1之间）
# transforms.Normalize()做归一化。它进行了减均值，再除以标准差。两个参数分别是均值和标准差
# transforms.Compose()函数则是将各种预处理的操作组合到了一起
data_tf = transforms.Compose(
[transforms.ToTensor(),
transforms.Normalize([0], [1])])
train_dataset = datasets.MNIST(root='./MNIST_data', train=True, transform=data_tf,
download=True) # 从torchvision包中下载数据集，并且保存，指明属于训练或测试数据,转换成Tensor()，一次性下载所有数据。
test_dataset = datasets.MNIST(root='./MNIST_data', train=False, transform=data_tf,
download=False) # 测试集的数据不需要再单独下载，已经在第一次统一下载了
train_loader = DataLoader(train_dataset, batch_size=batch_size,
shuffle=True) # 数据加载器，获取训练数据，从下载的数据中获取数据，并且选择获取批次、每次获取是否打乱
test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False) # 数据加载器，获取测试数据
# print(train_dataset.train_data.size())
# print(train_dataset.train_labels.size())
#
# print(test_dataset.test_data.size())
# print(test_dataset.test_labels.size())
class Net(nn.Module):
def __init__(self, in_dim, n_hidden_1, n_hidden_2, out_dim):
super(Net, self).__init__()
# Sequential()函数的功能是将网络的层组合到一起,按顺序执行
# 例如torch.nn.ReLU(inplace=True)
# inplace=True表示进行原地操作，对上一层传递下来的tensor直接进行修改，如x=x+3；
# inplace=False表示新建一个变量存储操作结果，如y=x+3，x=y；
# inplace=True可以节省运算内存，不用多存储变量。
self.layer1 = nn.Sequential(nn.Linear(in_dim, n_hidden_1, bias=True), nn.ReLU(True))
self.layer2 = nn.Sequential(nn.Linear(n_hidden_1, n_hidden_2, bias=True), nn.ReLU(True))
self.layer3 = nn.Sequential(nn.Linear(n_hidden_2, out_dim, bias=True))
def forward(self, x):
y1 = self.layer1(x)
y2 = self.layer2(y1)
y3 = self.layer3(y2)
return y3
net = Net(28 * 28, 256, 128, 10)
if torch.cuda.is_available():
net = net.cuda()
# 定义损失函数和优化器
loss_fn = nn.CrossEntropyLoss()#交叉熵
optimizer = optim.SGD(net.parameters(), lr=learning_rate)
net.train()
for epoch in range(num_epoches):
for i, (img, label) in enumerate(train_loader):
# print(img.size())
# img = img.view(img.size(0), -1)
# img = img.reshape(-1, 784)#转换形状
img = img.reshape(img.size(0), -1) # 转换形状
if torch.cuda.is_available():
img = img.cuda()
label = label.cuda()
out = net(img)
loss = loss_fn(out, label)
print_loss = loss.data.item()
optimizer.zero_grad()
loss.backward()
optimizer.step()
if i % 50 == 0:
print('epoch: {},i: {}, loss: {:.3}'.format(epoch, i, loss.data.item())) # 损失值显示3位精度
# print('epoch: {}, loss: {:.3}'.format(epoch, loss.data.item())) # 损失值显示3位精度
# 模型评估,此模式下，会固定模型中的BN层和Drpout层。
net.eval()
eval_loss = 0
eval_acc = 0
for data in test_loader:
img, label = data
# img = img.view(img.size(0), -1)
# img = img.reshape(-1,784)
img = img.reshape(img.size(0), -1)
if torch.cuda.is_available():
img = img.cuda()
label = label.cuda()
out = net(img)
loss = loss_fn(out, label)
eval_loss += loss.data.item() * label.size(0) # 平均损失*批次=每批数据的损失， 每批数据的损失*循环次数（+=叠加）=测试数据集的总损失
pred = torch.argmax(out, 1) # 返回每行中的最大值和最大值在每行中的索引
num_correct = (pred == label).sum() # 统计每批数据的精度
eval_acc += num_correct.item() # 每批的精度*循环次数（+=叠加）=测试数据集的总损失
'已经评估完所有测试集数据'
print(torch.argmax(out, 1))
print(label)
print(torch.max(out, 1))
print('Test Loss: {:.3}, Acc: {:.3}'.format(
eval_loss / (len(test_dataset)), # 计算所有测试数据集里的平均损失
eval_acc / (len(test_dataset)) # 计算所有测试数据集里的平均精度
))