深度学习(python)-Keras-1 Mnist手写数字识别

最新推荐文章于 2024-09-09 23:24:11 发布
最新推荐文章于 2024-09-09 23:24:11 发布
阅读量1.6k 收藏 1
点赞数 1
分类专栏： 深度学习 文章标签： 深度学习 keras
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/qq_36771850/article/details/86935464
版权

使用Python的Keras库来学习深度学习中的 “Hello World” ------ Mnist。
Mnist是机器学习领域里的一个经典数据集，这个数据集包含60000张训练图像和10000张测试图像。
mnist

mnist手写数字识别

用以下代码来实现mnist的网络并训练测试
注：通过**(train_images,train_labels),(test_images,test_labels) = mnist.load_data()**来导入mnist数据集，是通过网址的方式来下载数据集，可是这个网址是外网，需要翻墙。

from keras.datasets import mnist
from keras import models
from keras import layers
from keras.utils import to_categorical

(train_images,train_labels),(test_images,test_labels) = mnist.load_data()

network = models.Sequential()
network.add(layers.Dense(512,activation='relu',input_shape=(28*28,)))
network.add(layers.Dense(10,activation='softmax'))

network.compile(optimizer='rmsprop',
                loss='categorical_crossentropy',
                metrics=['accuracy'])

train_images = train_images.reshape((60000,28 * 28))
train_images = train_images.astype('float32') / 255

test_images = test_images.reshape((10000,28 * 28))
test_images = test_images.astype('float32') / 255

train_labels = to_categorical(train_labels)
test_labels = to_categorical(test_labels)

#开始训练网络
network.fit(train_images,train_labels,epochs=5,batch_size=128)

#检查模型在测试集上的性能
test_loss,test_acc = network.evaluate(test_images,test_labels)
print('test_acc:',test_acc)

下面显示运行结果，训练过程中显示了两个数字：loss(损失)和acc(准确率)，一个epoch表示所有训练样本运算学习一遍。可以看到最终的准确率达到了0.9891。最下面一组显示了用测试集测试的结果，10000张图片的准确率为0.9808。

Using TensorFlow backend.
Epoch 1/5
2018-10-12 21:06:38.599707: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2

  128/60000 [..............................] - ETA: 1:23 - loss: 2.3651 - acc: 0.0703
 1024/60000 [..............................] - ETA: 13s - loss: 1.3802 - acc: 0.5850 
 1920/60000 [..............................] - ETA: 8s - loss: 1.0771 - acc: 0.6906 
 2816/60000 [>.............................] - ETA: 7s - loss: 0.9057 - acc: 0.7436
 3712/60000 [>.............................] - ETA: 6s - loss: 0.8109 - acc: 0.7689
 4608/60000 [=>............................] - ETA: 5s - loss: 0.7271 - acc: 0.7925
 5504/60000 [=>............................] - ETA: 5s - loss: 0.6819 - acc: 0.8056
 6400/60000 [==>...........................] - ETA: 4s - loss: 0.6355 - acc: 0.8187
 7296/60000 [==>...........................] - ETA: 4s - loss: 0.6053 - acc: 0.8273
 8192/60000 [===>..........................] - ETA: 4s - loss: 0.5708 - acc: 0.8378
 9088/60000 [===>..........................] - ETA: 4s - loss: 0.5488 - acc: 0.8446
 9984/60000 [===>..........................] - ETA: 3s - loss: 0.5305 - acc: 0.8498
10880/60000 [====>.........................] - ETA: 3s - loss: 0.5148 - acc: 0.8537
11776/60000 [====>.........................] - ETA: 3s - loss: 0.4999 - acc: 0.8578
12672/60000 [=====>........................] - ETA: 3s - loss: 0.4871 - acc: 0.8612
13440/60000 [=====>........................] - ETA: 3s - loss: 0.4742 - acc: 0.8649
14208/60000 [======>.......................] - ETA: 3s - loss: 0.4654 - acc: 0.8670
14976/60000 [======>.......................] - ETA: 3s - loss: 0.4554 - acc: 0.8698
15872/60000 [======>.......................] - ETA: 3s - loss: 0.4439 - acc: 0.8732
16768/60000 [=======>......................] - ETA: 3s - loss: 0.4360 - acc: 0.8752
17664/60000 [=======>......................] - ETA: 3s - loss: 0.4291 - acc: 0.8774
18560/60000 [========>.....................] - ETA: 2s - loss: 0.4209 - acc: 0.8794
19456/60000 [========>.....................] - ETA: 2s - loss: 0.4116 - acc: 0.8822
20352/60000 [=========>....................] - ETA: 2s - loss: 0.4041 - acc: 0.8842
21248/60000 [=========>....................] - ETA: 2s - loss: 0.3980 - acc: 0.8864
22016/60000 [==========>...................] - ETA: 2s - loss: 0.3921 - acc: 0.8879
22784/60000 [==========>...................] - ETA: 2s - loss: 0.3862 - acc: 0.8896
23552/60000 [==========>...................] - ETA: 2s - loss: 0.3813 - acc: 0.8909
24448/60000 [===========>..................] - ETA: 2s - loss: 0.3761 - acc: 0.8923
25344/60000 [===========>..................] - ETA: 2s - loss: 0.3746 - acc: 0.8928
26240/60000 [============>.................] - ETA: 2s - loss: 0.3688 - acc: 0.8943
27136/60000 [============>.................] - ETA: 2s - loss: 0.3627 - acc: 0.8962
28032/60000 [=============>................] - ETA: 2s - loss: 0.3578 - acc: 0.8975
28928/60000 [=============>................] - ETA: 2s - loss: 0.3534 - acc: 0.8986
29824/60000 [=============>................] - ETA: 2s - loss: 0.3499 - acc: 0.8997
30720/60000 [==============>...............] - ETA: 2s - loss: 0.3456 - acc: 0.9009
31616/60000 [==============>...............] - ETA: 1s - loss: 0.3416 - acc: 0.9021
32512/60000 [===============>..............] - ETA: 1s - loss: 0.3368 - acc: 0.9034
33408/60000 [===============>..............] - ETA: 1s - loss: 0.3333 - acc: 0.9042
34304/60000 [================>.............] - ETA: 1s - loss: 0.3307 - acc: 0.9051
35200/60000 [================>.............] - ETA: 1s - loss: 0.3262 - acc: 0.9062
36096/60000 [=================>............] - ETA: 1s - loss: 0.3235 - acc: 0.9070
36992/60000 [=================>............] - ETA: 1s - loss: 0.3203 - acc: 0.9081
37888/60000 [=================>............] - ETA: 1s - loss: 0.3167 - acc: 0.9090
38784/60000 [==================>...........] - ETA: 1s - loss: 0.3127 - acc: 0.9102
39680/60000 [==================>...........] - ETA: 1s - loss: 0.3089 - acc: 0.9112
40576/60000 [===================>..........] - ETA: 1s - loss: 0.3068 - acc: 0.9117
41472/60000 [===================>..........] - ETA: 1s - loss: 0.3033 - acc: 0.9128
42368/60000 [====================>.........] - ETA: 1s - loss: 0.3000 - acc: 0.9138
43264/60000 [====================>.........] - ETA: 1s - loss: 0.2972 - acc: 0.9146
44032/60000 [=====================>........] - ETA: 1s - loss: 0.2949 - acc: 0.9154
44800/60000 [=====================>........] - ETA: 1s - loss: 0.2928 - acc: 0.9159
45696/60000 [=====================>........] - ETA: 0s - loss: 0.2904 - acc: 0.9164
46592/60000 [======================>.......] - ETA: 0s - loss: 0.2878 - acc: 0.9171
47488/60000 [======================>.......] - ETA: 0s - loss: 0.2858 - acc: 0.9175
48256/60000 [=======================>......] - ETA: 0s - loss: 0.2836 - acc: 0.9180
49152/60000 [=======================>......] - ETA: 0s - loss: 0.2812 - acc: 0.9188
50048/60000 [========================>.....] - ETA: 0s - loss: 0.2788 - acc: 0.9194
50944/60000 [========================>.....] - ETA: 0s - loss: 0.2770 - acc: 0.9200
51840/60000 [========================>.....] - ETA: 0s - loss: 0.2752 - acc: 0.9205
52736/60000 [=========================>....] - ETA: 0s - loss: 0.2741 - acc: 0.9208
53632/60000 [=========================>....] - ETA: 0s - loss: 0.2721 - acc: 0.9215
54528/60000 [==========================>...] - ETA: 0s - loss: 0.2695 - acc: 0.9222
55424/60000 [==========================>...] - ETA: 0s - loss: 0.2675 - acc: 0.9227
56320/60000 [===========================>..] - ETA: 0s - loss: 0.2651 - acc: 0.9234
57216/60000 [===========================>..] - ETA: 0s - loss: 0.2633 - acc: 0.9240
58112/60000 [============================>.] - ETA: 0s - loss: 0.2615 - acc: 0.9245
59008/60000 [============================>.] - ETA: 0s - loss: 0.2598 - acc: 0.9249
59904/60000 [============================>.] - ETA: 0s - loss: 0.2581 - acc: 0.9254
60000/60000 [==============================] - 4s 65us/step - loss: 0.2579 - acc: 0.9255
Epoch 2/5

  128/60000 [..............................] - ETA: 4s - loss: 0.1088 - acc: 0.9688
 1024/60000 [..............................] - ETA: 3s - loss: 0.1040 - acc: 0.9736
 1792/60000 [..............................] - ETA: 3s - loss: 0.1255 - acc: 0.9699
 2688/60000 [>.............................] - ETA: 3s - loss: 0.1264 - acc: 0.9665
 3584/60000 [>.............................] - ETA: 3s - loss: 0.1265 - acc: 0.9660
 4480/60000 [=>............................] - ETA: 3s - loss: 0.1217 - acc: 0.9665
 5376/60000 [=>............................] - ETA: 3s - loss: 0.1199 - acc: 0.9648
 6272/60000 [==>...........................] - ETA: 3s - loss: 0.1181 - acc: 0.9659
 7168/60000 [==>...........................] - ETA: 3s - loss: 0.1211 - acc: 0.9653
 8064/60000 [===>..........................] - ETA: 3s - loss: 0.1220 - acc: 0.9652
 8960/60000 [===>..........................] - ETA: 3s - loss: 0.1207 - acc: 0.9655
 9600/60000 [===>..........................] - ETA: 3s - loss: 0.1203 - acc: 0.9654
10368/60000 [====>.........................] - ETA: 3s - loss: 0.1226 - acc: 0.9643
11264/60000 [====>.........................] - ETA: 3s - loss: 0.1213 - acc: 0.9645
12160/60000 [=====>........................] - ETA: 3s - loss: 0.1220 - acc: 0.9643
13056/60000 [=====>........................] - ETA: 3s - loss: 0.1212 - acc: 0.9647
13952/60000 [=====>........................] - ETA: 2s - loss: 0.1206 - acc: 0.9650
14848/60000 [======>.......................] - ETA: 2s - loss: 0.1199 - acc: 0.9646
15616/60000 [======>.......................] - ETA: 2s - loss: 0.1193 - acc: 0.9648
16384/60000 [=======>......................] - ETA: 2s - loss: 0.1190 - acc: 0.9648
17152/60000 [=======>......................] - ETA: 2s - loss: 0.1175 - acc: 0.9653
17920/60000 [=======>......................] - ETA: 2s - loss: 0.1179 - acc: 0.9652
18816/60000 [========>.....................] - ETA: 2s - loss: 0.1174 - acc: 0.9652
19712/60000 [========>.....................] - ETA: 2s - loss: 0.1164 - acc: 0.9655
20608/60000 [=========>....................] - ETA: 2s - loss: 0.1168 - acc: 0.9653
21376/60000 [=========>....................] - ETA: 2s - loss: 0.1161 - acc: 0.9655
22144/60000 [==========>...................] - ETA: 2s - loss: 0.1155 - acc: 0.9655
22912/60000 [==========>...................] - ETA: 2s - loss: 0.1144 - acc: 0.9659
23808/60000 [==========>...................] - ETA: 2s - loss: 0.1136 - acc: 0.9662
24704/60000 [===========>..................] - ETA: 2s - loss: 0.1136 - acc: 0.9661
25600/60000 [===========>..................] - ETA: 2s - loss: 0.1121 - acc: 0.9665
26496/60000 [============>.................] - ETA: 2s - loss: 0.1121 - acc: 0.9663
27392/60000 [============>.................] - ETA: 2s - loss: 0.1116 - acc: 0.9667
28288/60000 [=============>................] - ETA: 2s - loss: 0.1111 - acc: 0.9668
29184/60000 [=============>................] - ETA: 2s - loss: 0.1107 - acc: 0.9669
30080/60000 [==============>...............] - ETA: 1s - loss: 0.1108 - acc: 0.9669
30848/60000 [==============>...............] - ETA: 1s - loss: 0.1109 - acc: 0.9669
31616/60000 [==============>...............] - ETA: 1s - loss: 0.1108 - acc: 0.9670
32384/60000 [===============>..............] - ETA: 1s - loss: 0.1107 - acc: 0.9671
33152/60000 [===============>..............] - ETA: 1s - loss: 0.1099 - acc: 0.9672
33920/60000 [===============>..............] - ETA: 1s - loss: 0.1104 - acc: 0.9669
34688/60000 [================>.............] - ETA: 1s - loss: 0.1101 - acc: 0.9669
35456/60000 [================>.............] - ETA: 1s - loss: 0.1094 - acc: 0.9672
36224/60000 [=================>............] - ETA: 1s - loss: 0.1091 - acc: 0.9673
36992/60000 [=================>............] - ETA: 1s - loss: 0.1091 - acc: 0.9674
37760/60000 [=================>............] - ETA: 1s - loss: 0.1087 - acc: 0.9675
38400/60000 [==================>...........] - ETA: 1s - loss: 0.1083 - acc: 0.9676
39040/60000 [==================>...........] - ETA: 1s - loss: 0.1084 - acc: 0.9675
39808/60000 [==================>...........] - ETA: 1s - loss: 0.1084 - acc: 0.9675
40704/60000 [===================>..........] - ETA: 1s - loss: 0.1079 - acc: 0.9676
41600/60000 [===================>..........] - ETA: 1s - loss: 0.1068 - acc: 0.9680
42496/60000 [====================>.........] - ETA: 1s - loss: 0.1066 - acc: 0.9680
43392/60000 [====================>.........] - ETA: 1s - loss: 0.1069 - acc: 0.9680
44288/60000 [=====================>........] - ETA: 1s - loss: 0.1067 - acc: 0.9680
45056/60000 [=====================>........] - ETA: 0s - loss: 0.1060 - acc: 0.9682
45824/60000 [=====================>........] - ETA: 0s - loss: 0.1058 - acc: 0.9682
46592/60000 [======================>.......] - ETA: 0s - loss: 0.1055 - acc: 0.9683
47360/60000 [======================>.......] - ETA: 0s - loss: 0.1051 - acc: 0.9684
48128/60000 [=======================>......] - ETA: 0s - loss: 0.1048 - acc: 0.9685
48896/60000 [=======================>......] - ETA: 0s - loss: 0.1045 - acc: 0.9686
49664/60000 [=======================>......] - ETA: 0s - loss: 0.1041 - acc: 0.9687
50432/60000 [========================>.....] - ETA: 0s - loss: 0.1043 - acc: 0.9687
51200/60000 [========================>.....] - ETA: 0s - loss: 0.1042 - acc: 0.9688
51968/60000 [========================>.....] - ETA: 0s - loss: 0.1038 - acc: 0.9688
52736/60000 [=========================>....] - ETA: 0s - loss: 0.1038 - acc: 0.9689
53504/60000 [=========================>....] - ETA: 0s - loss: 0.1039 - acc: 0.9689
54272/60000 [==========================>...] - ETA: 0s - loss: 0.1036 - acc: 0.9691
55040/60000 [==========================>...] - ETA: 0s - loss: 0.1031 - acc: 0.9692
55808/60000 [==========================>...] - ETA: 0s - loss: 0.1034 - acc: 0.9690
56576/60000 [===========================>..] - ETA: 0s - loss: 0.1030 - acc: 0.9692
57472/60000 [===========================>..] - ETA: 0s - loss: 0.1026 - acc: 0.9693
58368/60000 [============================>.] - ETA: 0s - loss: 0.1025 - acc: 0.9694
59264/60000 [============================>.] - ETA: 0s - loss: 0.1021 - acc: 0.9696
60000/60000 [==============================] - 4s 68us/step - loss: 0.1021 - acc: 0.9696
Epoch 3/5

  128/60000 [..............................] - ETA: 4s - loss: 0.0370 - acc: 0.9766
 1024/60000 [..............................] - ETA: 3s - loss: 0.0760 - acc: 0.9756
 1920/60000 [..............................] - ETA: 3s - loss: 0.0629 - acc: 0.9807
 2816/60000 [>.............................] - ETA: 3s - loss: 0.0644 - acc: 0.9819
 3712/60000 [>.............................] - ETA: 3s - loss: 0.0643 - acc: 0.9822
 4608/60000 [=>............................] - ETA: 3s - loss: 0.0651 - acc: 0.9811
 5504/60000 [=>............................] - ETA: 3s - loss: 0.0657 - acc: 0.9816
 6400/60000 [==>...........................] - ETA: 3s - loss: 0.0694 - acc: 0.9798
 7296/60000 [==>...........................] - ETA: 3s - loss: 0.0671 - acc: 0.9807
 8064/60000 [===>..........................] - ETA: 3s - loss: 0.0670 - acc: 0.9807
 8832/60000 [===>..........................] - ETA: 3s - loss: 0.0677 - acc: 0.9797
 9728/60000 [===>..........................] - ETA: 3s - loss: 0.0657 - acc: 0.9801
10624/60000 [====>.........................] - ETA: 3s - loss: 0.0655 - acc: 0.9800
11392/60000 [====>.........................] - ETA: 3s - loss: 0.0662 - acc: 0.9802
12288/60000 [=====>........................] - ETA: 3s - loss: 0.0664 - acc: 0.9800
13184/60000 [=====>........................] - ETA: 2s - loss: 0.0665 - acc: 0.9795
14080/60000 [======>.......................] - ETA: 2s - loss: 0.0671 - acc: 0.9797
14976/60000 [======>.......................] - ETA: 2s - loss: 0.0678 - acc: 0.9791
15872/60000 [======>.......................] - ETA: 2s - loss: 0.0688 - acc: 0.9792
16640/60000 [=======>......................] - ETA: 2s - loss: 0.0683 - acc: 0.9793
17408/60000 [=======>......................] - ETA: 2s - loss: 0.0691 - acc: 0.9791
18176/60000 [========>.....................] - ETA: 2s - loss: 0.0703 - acc: 0.9789
18944/60000 [========>.....................] - ETA: 2s - loss: 0.0700 - acc: 0.9790
19840/60000 [========>.....................] - ETA: 2s - loss: 0.0691 - acc: 0.9793
20736/60000 [=========>....................] - ETA: 2s - loss: 0.0691 - acc: 0.9793
21632/60000 [=========>....................] - ETA: 2s - loss: 0.0682 - acc: 0.9796
22528/60000 [==========>...................] - ETA: 2s - loss: 0.0682 - acc: 0.9796
23296/60000 [==========>...................] - ETA: 2s - loss: 0.0686 - acc: 0.9793
24064/60000 [===========>..................] - ETA: 2s - loss: 0.0682 - acc: 0.9794
24960/60000 [===========>..................] - ETA: 2s - loss: 0.0683 - acc: 0.9793
25856/60000 [===========>..................] - ETA: 2s - loss: 0.0685 - acc: 0.9792
26752/60000 [============>.................] - ETA: 2s - loss: 0.0684 - acc: 0.9791
27648/60000 [============>.................] - ETA: 2s - loss: 0.0679 - acc: 0.9793
28416/60000 [=============>................] - ETA: 2s - loss: 0.0680 - acc: 0.9793
29312/60000 [=============>................] - ETA: 1s - loss: 0.0685 - acc: 0.9795
30080/60000 [==============>...............] - ETA: 1s - loss: 0.0685 - acc: 0.9794
30848/60000 [==============>...............] - ETA: 1s - loss: 0.0680 - acc: 0.9795
31616/60000 [==============>...............] - ETA: 1s - loss: 0.0677 - acc: 0.9797
32512/60000 [===============>..............] - ETA: 1s - loss: 0.0676 - acc: 0.9796
33408/60000 [===============>..............] - ETA: 1s - loss: 0.0685 - acc: 0.9794
34304/60000 [================>.............] - ETA: 1s - loss: 0.0683 - acc: 0.9793
35200/60000 [================>.............] - ETA: 1s - loss: 0.0680 - acc: 0.9794
36096/60000 [=================>............] - ETA: 1s - loss: 0.0678 - acc: 0.9794
36992/60000 [=================>............] - ETA: 1s - loss: 0.0678 - acc: 0.9794
37888/60000 [=================>............] - ETA: 1s - loss: 0.0679 - acc: 0.9794
38784/60000 [==================>...........] - ETA: 1s - loss: 0.0675 - acc: 0.9795
39552/60000 [==================>...........] - ETA: 1s - loss: 0.0679 - acc: 0.9795
40320/60000 [===================>..........] - ETA: 1s - loss: 0.0677 - acc: 0.9795
41088/60000 [===================>..........] - ETA: 1s - loss: 0.0676 - acc: 0.9796
41984/60000 [===================>..........] - ETA: 1s - loss: 0.0674 - acc: 0.9796
42880/60000 [====================>.........] - ETA: 1s - loss: 0.0676 - acc: 0.9795
43776/60000 [====================>.........] - ETA: 1s - loss: 0.0675 - acc: 0.9796
44672/60000 [=====================>........] - ETA: 0s - loss: 0.0672 - acc: 0.9797
45440/60000 [=====================>........] - ETA: 0s - loss: 0.0672 - acc: 0.9798
46208/60000 [======================>.......] - ETA: 0s - loss: 0.0680 - acc: 0.9797
47104/60000 [======================>.......] - ETA: 0s - loss: 0.0679 - acc: 0.9798
47872/60000 [======================>.......] - ETA: 0s - loss: 0.0679 - acc: 0.9798
48640/60000 [=======================>......] - ETA: 0s - loss: 0.0681 - acc: 0.9797
49408/60000 [=======================>......] - ETA: 0s - loss: 0.0680 - acc: 0.9797
50176/60000 [========================>.....] - ETA: 0s - loss: 0.0680 - acc: 0.9798
50944/60000 [========================>.....] - ETA: 0s - loss: 0.0678 - acc: 0.9799
51712/60000 [========================>.....] - ETA: 0s - loss: 0.0673 - acc: 0.9800
52480/60000 [=========================>....] - ETA: 0s - loss: 0.0674 - acc: 0.9801
53248/60000 [=========================>....] - ETA: 0s - loss: 0.0676 - acc: 0.9800
54016/60000 [==========================>...] - ETA: 0s - loss: 0.0674 - acc: 0.9800
54912/60000 [==========================>...] - ETA: 0s - loss: 0.0673 - acc: 0.9801
55680/60000 [==========================>...] - ETA: 0s - loss: 0.0672 - acc: 0.9802
56576/60000 [===========================>..] - ETA: 0s - loss: 0.0676 - acc: 0.9801
57472/60000 [===========================>..] - ETA: 0s - loss: 0.0676 - acc: 0.9801
58368/60000 [============================>.] - ETA: 0s - loss: 0.0678 - acc: 0.9800
59264/60000 [============================>.] - ETA: 0s - loss: 0.0677 - acc: 0.9800
60000/60000 [==============================] - 4s 66us/step - loss: 0.0673 - acc: 0.9801
Epoch 4/5

  128/60000 [..............................] - ETA: 4s - loss: 0.0537 - acc: 0.9766
 1024/60000 [..............................] - ETA: 3s - loss: 0.0386 - acc: 0.9912
 1792/60000 [..............................] - ETA: 3s - loss: 0.0440 - acc: 0.9900
 2560/60000 [>.............................] - ETA: 3s - loss: 0.0523 - acc: 0.9863
 3456/60000 [>.............................] - ETA: 3s - loss: 0.0512 - acc: 0.9864
 4352/60000 [=>............................] - ETA: 3s - loss: 0.0566 - acc: 0.9830
 5120/60000 [=>............................] - ETA: 3s - loss: 0.0545 - acc: 0.9834
 5888/60000 [=>............................] - ETA: 3s - loss: 0.0534 - acc: 0.9835
 6784/60000 [==>...........................] - ETA: 3s - loss: 0.0533 - acc: 0.9833
 7680/60000 [==>...........................] - ETA: 3s - loss: 0.0554 - acc: 0.9824
 8576/60000 [===>..........................] - ETA: 3s - loss: 0.0534 - acc: 0.9832
 9472/60000 [===>..........................] - ETA: 3s - loss: 0.0529 - acc: 0.9830
10368/60000 [====>.........................] - ETA: 3s - loss: 0.0522 - acc: 0.9836
11264/60000 [====>.........................] - ETA: 3s - loss: 0.0521 - acc: 0.9838
12160/60000 [=====>........................] - ETA: 3s - loss: 0.0535 - acc: 0.9836
13056/60000 [=====>........................] - ETA: 3s - loss: 0.0518 - acc: 0.9843
13952/60000 [=====>........................] - ETA: 2s - loss: 0.0523 - acc: 0.9843
14848/60000 [======>.......................] - ETA: 2s - loss: 0.0518 - acc: 0.9844
15488/60000 [======>.......................] - ETA: 2s - loss: 0.0518 - acc: 0.9845
16256/60000 [=======>......................] - ETA: 2s - loss: 0.0519 - acc: 0.9843
17152/60000 [=======>......................] - ETA: 2s - loss: 0.0508 - acc: 0.9845
17920/60000 [=======>......................] - ETA: 2s - loss: 0.0512 - acc: 0.9843
18816/60000 [========>.....................] - ETA: 2s - loss: 0.0509 - acc: 0.9845
19584/60000 [========>.....................] - ETA: 2s - loss: 0.0518 - acc: 0.9843
20480/60000 [=========>....................] - ETA: 2s - loss: 0.0515 - acc: 0.9843
21376/60000 [=========>....................] - ETA: 2s - loss: 0.0518 - acc: 0.9840
22272/60000 [==========>...................] - ETA: 2s - loss: 0.0511 - acc: 0.9843
23168/60000 [==========>...................] - ETA: 2s - loss: 0.0508 - acc: 0.9844
24064/60000 [===========>..................] - ETA: 2s - loss: 0.0505 - acc: 0.9845
24960/60000 [===========>..................] - ETA: 2s - loss: 0.0502 - acc: 0.9845
25856/60000 [===========>..................] - ETA: 2s - loss: 0.0503 - acc: 0.9845
26752/60000 [============>.................] - ETA: 2s - loss: 0.0502 - acc: 0.9844
27648/60000 [============>.................] - ETA: 2s - loss: 0.0509 - acc: 0.9843
28544/60000 [=============>................] - ETA: 2s - loss: 0.0515 - acc: 0.9841
29440/60000 [=============>................] - ETA: 1s - loss: 0.0514 - acc: 0.9840
30336/60000 [==============>...............] - ETA: 1s - loss: 0.0514 - acc: 0.9839
31232/60000 [==============>...............] - ETA: 1s - loss: 0.0510 - acc: 0.9841
32000/60000 [===============>..............] - ETA: 1s - loss: 0.0507 - acc: 0.9842
32896/60000 [===============>..............] - ETA: 1s - loss: 0.0500 - acc: 0.9844
33792/60000 [===============>..............] - ETA: 1s - loss: 0.0506 - acc: 0.9842
34688/60000 [================>.............] - ETA: 1s - loss: 0.0501 - acc: 0.9843
35584/60000 [================>.............] - ETA: 1s - loss: 0.0499 - acc: 0.9843
36352/60000 [=================>............] - ETA: 1s - loss: 0.0495 - acc: 0.9845
37248/60000 [=================>............] - ETA: 1s - loss: 0.0494 - acc: 0.9845
38144/60000 [==================>...........] - ETA: 1s - loss: 0.0498 - acc: 0.9845
39040/60000 [==================>...........] - ETA: 1s - loss: 0.0495 - acc: 0.9846
39936/60000 [==================>...........] - ETA: 1s - loss: 0.0493 - acc: 0.9846
40832/60000 [===================>..........] - ETA: 1s - loss: 0.0495 - acc: 0.9845
41728/60000 [===================>..........] - ETA: 1s - loss: 0.0496 - acc: 0.9845
42624/60000 [====================>.........] - ETA: 1s - loss: 0.0499 - acc: 0.9843
43520/60000 [====================>.........] - ETA: 1s - loss: 0.0500 - acc: 0.9844
44416/60000 [=====================>........] - ETA: 0s - loss: 0.0500 - acc: 0.9845
45312/60000 [=====================>........] - ETA: 0s - loss: 0.0497 - acc: 0.9846
46208/60000 [======================>.......] - ETA: 0s - loss: 0.0502 - acc: 0.9844
47104/60000 [======================>.......] - ETA: 0s - loss: 0.0498 - acc: 0.9846
48000/60000 [=======================>......] - ETA: 0s - loss: 0.0498 - acc: 0.9846
48768/60000 [=======================>......] - ETA: 0s - loss: 0.0497 - acc: 0.9846
49664/60000 [=======================>......] - ETA: 0s - loss: 0.0495 - acc: 0.9848
50560/60000 [========================>.....] - ETA: 0s - loss: 0.0498 - acc: 0.9848
51456/60000 [========================>.....] - ETA: 0s - loss: 0.0498 - acc: 0.9847
52352/60000 [=========================>....] - ETA: 0s - loss: 0.0497 - acc: 0.9847
53248/60000 [=========================>....] - ETA: 0s - loss: 0.0498 - acc: 0.9847
54144/60000 [==========================>...] - ETA: 0s - loss: 0.0496 - acc: 0.9847
55040/60000 [==========================>...] - ETA: 0s - loss: 0.0498 - acc: 0.9847
55936/60000 [==========================>...] - ETA: 0s - loss: 0.0495 - acc: 0.9848
56832/60000 [===========================>..] - ETA: 0s - loss: 0.0493 - acc: 0.9849
57728/60000 [===========================>..] - ETA: 0s - loss: 0.0494 - acc: 0.9847
58624/60000 [============================>.] - ETA: 0s - loss: 0.0494 - acc: 0.9847
59520/60000 [============================>.] - ETA: 0s - loss: 0.0491 - acc: 0.9848
60000/60000 [==============================] - 4s 63us/step - loss: 0.0491 - acc: 0.9849
Epoch 5/5

  128/60000 [..............................] - ETA: 4s - loss: 0.0221 - acc: 1.0000
 1024/60000 [..............................] - ETA: 3s - loss: 0.0466 - acc: 0.9893
 1920/60000 [..............................] - ETA: 3s - loss: 0.0402 - acc: 0.9891
 2816/60000 [>.............................] - ETA: 3s - loss: 0.0361 - acc: 0.9890
 3712/60000 [>.............................] - ETA: 3s - loss: 0.0356 - acc: 0.9892
 4608/60000 [=>............................] - ETA: 3s - loss: 0.0357 - acc: 0.9896
 5504/60000 [=>............................] - ETA: 3s - loss: 0.0342 - acc: 0.9898
 6400/60000 [==>...........................] - ETA: 3s - loss: 0.0355 - acc: 0.9895
 7296/60000 [==>...........................] - ETA: 3s - loss: 0.0338 - acc: 0.9904
 8192/60000 [===>..........................] - ETA: 3s - loss: 0.0338 - acc: 0.9901
 9088/60000 [===>..........................] - ETA: 3s - loss: 0.0332 - acc: 0.9904
 9984/60000 [===>..........................] - ETA: 3s - loss: 0.0341 - acc: 0.9899
10880/60000 [====>.........................] - ETA: 3s - loss: 0.0331 - acc: 0.9903
11776/60000 [====>.........................] - ETA: 3s - loss: 0.0342 - acc: 0.9898
12544/60000 [=====>........................] - ETA: 2s - loss: 0.0335 - acc: 0.9902
13312/60000 [=====>........................] - ETA: 2s - loss: 0.0332 - acc: 0.9904
14080/60000 [======>.......................] - ETA: 2s - loss: 0.0332 - acc: 0.9901
14848/60000 [======>.......................] - ETA: 2s - loss: 0.0328 - acc: 0.9902
15744/60000 [======>.......................] - ETA: 2s - loss: 0.0325 - acc: 0.9905
16640/60000 [=======>......................] - ETA: 2s - loss: 0.0323 - acc: 0.9906
17536/60000 [=======>......................] - ETA: 2s - loss: 0.0317 - acc: 0.9909
18432/60000 [========>.....................] - ETA: 2s - loss: 0.0314 - acc: 0.9908
19328/60000 [========>.....................] - ETA: 2s - loss: 0.0319 - acc: 0.9906
20224/60000 [=========>....................] - ETA: 2s - loss: 0.0320 - acc: 0.9904
21120/60000 [=========>....................] - ETA: 2s - loss: 0.0322 - acc: 0.9903
22016/60000 [==========>...................] - ETA: 2s - loss: 0.0326 - acc: 0.9902
22912/60000 [==========>...................] - ETA: 2s - loss: 0.0323 - acc: 0.9903
23808/60000 [==========>...................] - ETA: 2s - loss: 0.0331 - acc: 0.9900
24704/60000 [===========>..................] - ETA: 2s - loss: 0.0336 - acc: 0.9898
25600/60000 [===========>..................] - ETA: 2s - loss: 0.0338 - acc: 0.9897
26496/60000 [============>.................] - ETA: 2s - loss: 0.0340 - acc: 0.9897
27392/60000 [============>.................] - ETA: 2s - loss: 0.0337 - acc: 0.9898
28288/60000 [=============>................] - ETA: 1s - loss: 0.0337 - acc: 0.9897
29056/60000 [=============>................] - ETA: 1s - loss: 0.0338 - acc: 0.9897
29824/60000 [=============>................] - ETA: 1s - loss: 0.0338 - acc: 0.9897
30720/60000 [==============>...............] - ETA: 1s - loss: 0.0347 - acc: 0.9894
31616/60000 [==============>...............] - ETA: 1s - loss: 0.0344 - acc: 0.9895
32512/60000 [===============>..............] - ETA: 1s - loss: 0.0342 - acc: 0.9894
33408/60000 [===============>..............] - ETA: 1s - loss: 0.0346 - acc: 0.9892
34304/60000 [================>.............] - ETA: 1s - loss: 0.0345 - acc: 0.9892
35200/60000 [================>.............] - ETA: 1s - loss: 0.0348 - acc: 0.9892
36096/60000 [=================>............] - ETA: 1s - loss: 0.0347 - acc: 0.9893
36992/60000 [=================>............] - ETA: 1s - loss: 0.0344 - acc: 0.9892
37888/60000 [=================>............] - ETA: 1s - loss: 0.0344 - acc: 0.9892
38784/60000 [==================>...........] - ETA: 1s - loss: 0.0344 - acc: 0.9893
39680/60000 [==================>...........] - ETA: 1s - loss: 0.0344 - acc: 0.9893
40576/60000 [===================>..........] - ETA: 1s - loss: 0.0349 - acc: 0.9892
41472/60000 [===================>..........] - ETA: 1s - loss: 0.0351 - acc: 0.9891
42368/60000 [====================>.........] - ETA: 1s - loss: 0.0348 - acc: 0.9892
43264/60000 [====================>.........] - ETA: 1s - loss: 0.0347 - acc: 0.9893
44160/60000 [=====================>........] - ETA: 0s - loss: 0.0348 - acc: 0.9893
45056/60000 [=====================>........] - ETA: 0s - loss: 0.0348 - acc: 0.9893
45952/60000 [=====================>........] - ETA: 0s - loss: 0.0349 - acc: 0.9892
46720/60000 [======================>.......] - ETA: 0s - loss: 0.0347 - acc: 0.9893
47616/60000 [======================>.......] - ETA: 0s - loss: 0.0347 - acc: 0.9893
48512/60000 [=======================>......] - ETA: 0s - loss: 0.0350 - acc: 0.9893
49408/60000 [=======================>......] - ETA: 0s - loss: 0.0348 - acc: 0.9893
50304/60000 [========================>.....] - ETA: 0s - loss: 0.0350 - acc: 0.9892
51200/60000 [========================>.....] - ETA: 0s - loss: 0.0353 - acc: 0.9891
52096/60000 [=========================>....] - ETA: 0s - loss: 0.0350 - acc: 0.9893
52992/60000 [=========================>....] - ETA: 0s - loss: 0.0352 - acc: 0.9892
53888/60000 [=========================>....] - ETA: 0s - loss: 0.0355 - acc: 0.9892
54784/60000 [==========================>...] - ETA: 0s - loss: 0.0354 - acc: 0.9892
55680/60000 [==========================>...] - ETA: 0s - loss: 0.0353 - acc: 0.9893
56576/60000 [===========================>..] - ETA: 0s - loss: 0.0357 - acc: 0.9892
57472/60000 [===========================>..] - ETA: 0s - loss: 0.0361 - acc: 0.9892
58240/60000 [============================>.] - ETA: 0s - loss: 0.0363 - acc: 0.9891
59136/60000 [============================>.] - ETA: 0s - loss: 0.0364 - acc: 0.9891
60000/60000 [==============================] - 4s 62us/step - loss: 0.0363 - acc: 0.9891

   32/10000 [..............................] - ETA: 7s
 1344/10000 [===>..........................] - ETA: 0s
 2528/10000 [======>.......................] - ETA: 0s
 3840/10000 [==========>...................] - ETA: 0s
 5152/10000 [==============>...............] - ETA: 0s
 6400/10000 [==================>...........] - ETA: 0s
 7584/10000 [=====================>........] - ETA: 0s
 8864/10000 [=========================>....] - ETA: 0s
10000/10000 [==============================] - 0s 42us/step
test_acc: 0.9808
RYAN-LZ
关注
专栏目录
01-Keras基础应用（MNIST手写数字图像识别）
ML_GearYe的博客
09-10 774
tensorflow：2.6.0 python：3.7.11 一、版本检查 In[]: tf.__version__ Out[]: '2.6.0' In[]: import platform print(platform.python_version()) 3.7.11 二.MNIST的数据查看 1.相关包导入 In[]: import tensorflow as tf from tensorflow.keras im...
keras实现的mnist手写数字识别(带测试的手写图)
09-26
keras实现的mnist手写数字识别，模型已经训练好，也可以重新进行训练，带了自己手写的28x28的测试图片，有需要的可以下载
【Python】keras神经网络识别mnist
weixin_34221073的博客
01-08 121
上次用Matlab写过一个识别Mnist的神经网络，地址在：https://www.cnblogs.com/tiandsp/p/9042908.html 这次又用Keras做了一个差不多的，毕竟，现在最流行的项目都是Python做的，我也跟一下潮流：) 数据是从本地解析好的图像和标签载入的。 神经网络有两个隐含层，都有512个节点。 import numpy as np from k...
Python机器学习——利用Keras和基础神经网络进行手写数字识别（MNIST数据集）
最新发布
Eric005的博客
09-09 1264
其中：train_images保存用来训练的图像，train_labels是与之对应的标签。test_images和test_labels分别为用来验证的图像和标签，也就是验证集。对于输出 0-9 这10个标签而言，每个标签的地位应该是相等的，并不存在标签数字2大于数字1的情况。因此，在大部分情况下，都需要将标签转换为 one-hot 编码，也就独热编码，这样标签之间便没有任何大小而言。用 keras 中自带的mnist模块，加载数据集load_data进来，分别赋值给四个变量。每一行的向量代表一个标签。
手写体数字识别系统实现
07-11
UI 设计包括界面设计、菜单栏布局、进度条、结果展示和图片展示区域划分 几个主要任务。图像预处理包括图像的灰度化、二值化、反色处理、图形锐化、 数字分割、归一化等主要任务。特征提取为使用不同的特征提取方法，对于预处 理过的图像，进行数字特征提取。图像识别是对于数字图像，将特征提取出的数 字矩阵数组，采用不同的识别方法，一个个地识别出他们对应的数字，最终显示 在界面。
python keras MNIST CNN 数字识别 建立模型训练并评估
zheng_ruiguo的专栏
07-29 882
''' python keras MNIST CNN 数字识别 建立模型训练并评估 ''' #载入数据 from keras.datasets import mnist from keras.utils import to_categorical (train_images , train_labels),(test_images,test_labels) = mnist.load_data(...
基于python+Keras+Flask的Mnist手写数字集识别系统-毕业设计源码+全部数据+使用文档（高分优秀项目）
04-14
基于python + Keras + Flask的 Mnist 手写数字集识别系统-毕业设计源码+使用文档（高分优秀项目） 该项目是个人高分毕业设计项目源码，已获导师指导认可通过，答辩评审分达到97分，在window10/11测试环境严格调试，...
基于 Keras + Flask 的 Mnist 手写数字集识别系统.zip
04-03
标题 "基于 Keras + Flask 的 Mnist 手写数字集识别系统" 提示我们这个项目是关于使用深度学习技术，特别是 Keras 框架，来构建一个手写数字识别的应用程序。Mnist 数据集是机器学习领域的一个经典数据集，主要用于...
基于Keras+Flask的Mnist手写数字识别系统源码+项目说明(识别率95%+识别模型).zip
07-28
基于Keras+Flask的Mnist手写数字识别系统源码+项目说明(识别率95%+识别模型).zip 【资源介绍】 注释掉 app.py 中 使用 redis 记录访问次数的功能 主要是这个方法的使用与引用 get_visit_info() （如果你有 redis ...
Python-使用sklearnpytorch和keras进行mnist手写数字分类的教程
08-10
在本教程中，我们将深入探讨如何使用Python中的三个流行机器学习库——scikit-learn（简称sklearn）、PyTorch和Keras，对MNIST数据集进行手写数字分类。MNIST是机器学习领域的一个经典数据集，包含了60,000个训练...
python 数字手写体识别
08-17
python入门数字手写体识别，用python编程实现基于tensorflow的数字手写体识别的识别部分，
python手写体识别数据集
03-13
python手写体识别数据集，包括4个文件夹t10k-images-idx3-ubyte.gz，t10k-labels-idx1-ubyte.gz，train-images-idx3-ubyte.gz，train-labels-idx1-ubyte.gz
基于卷积神经网络的手写数字识别python代码实现
04-20
基于卷积神经网络的手写数字识别python代码实现 卷积神经网络（Convolutional Neural Network,CNN）中，卷积层的神经元只与前一层的部分神经元节点相连，即它的神经元间的连接是非全连接的，且同一层中某些神经元之间的连接的权重 w 和偏移 b 是共享的（即相同的），这样大量地减少了需要训练参数的数量。
Python实现将Keras内置的mnist数据集转化为图像进行存储
Together_CZ的博客
05-07 876
虽然很多时候一些模块内置的数据集使用起来非常的方便，搭建模型，做实例项目非常快，但是这种形式的数据建模本质上并不通用，我们使用模型大多是要解决现实生活中的很多实际问题，在不同的应用场景和业务问题中都会遇上各式各样的数据集，没有办法像那些使用起来很方便的内置数据集一样直接导入使用，这时候我们就要把问题的处理过程抽象成一个标准的数据流程，抽象出来标准流程的好处就是，之后任何问题我们都可以将其朝着标准流程的形式进行转化，从而讲一个复杂的问题标准化，就像很多时候搭建模型一样，明确需要定义的部分就是输入层的shape
Python深度学习库Keras学习之mnist手写体识别
qq_35546153的博客
04-02 1892
    本文为学习过程总结，此文参考了很多博客的精华，参考地址已贴出；目标：1，学习使用Keras库的函数2，完成mnist识别背景介绍：（1）Keras介绍    Keras是一个高层神经网络API，Keras由纯Python编写而成并基Tensorflow、Theano以及CNTK后端。Keras 为支持快速实验而生，能够把你的idea迅速转换为结果，如果你有如下需求，请选择Keras：简易和...
keras mnist python 机器学习 深度学习
Lifelong Learning Pan
07-04 254
from keras.datasets import mnist from keras.utils import np_utils import numpy as np np.random.seed(10); (x_Train,y_Train),(x_Test,y_Test)=mnist.load_data(); x_Train4D=x_Train.reshape(x_Train.shape...
keras实现mnist数据集手写数字识别
热门推荐
陈正经
07-16 2万+
经过几天的爬坑，“东搞西搞”终于把深度学习的“HELLO,WORLD”做出来了，以下是自己的实战过程： 关于keras识别手写数字的入门准备： 1.Keras是一个高层神经网络API，Keras由纯Python编写而成并基Tensorflow、Theano以及CNTK后端。所以在安装keras的前面，要先搭建tensorflow环境和安装https://www.tensorflow.org/i...
用python的numpy实现mnist手写数字识别
我不是坏人的博客
11-22 9249
用numpy实现mnist手写数字识别（不使用深度学习框架）；一万张测试集，以识别正确图片所占的比例作为准确率，准确率达到93%
Keras MNIST手写数字识别案例解析
该项目以MNIST数据集为基础，展现了如何使用深度学习技术进行图像识别任务。MNIST是一个包含了大量手写数字图片的大型数据库，常被用于训练各种图像处理系统。该示例项目使用Keras这个高级神经网络API，它运行在...
评论 1
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值