现在的许多手写字体识别代码都是基于已有的mnist手写字体数据集进行的，而kaggle需要用到网站上给出的数据集并生成测试集的输出用于提交。这里选择keras搭建卷积网络进行识别，可以直接生成测试集的结果,最终结果识别率大概97%左右的样子。
|
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
|
# -*- coding: utf-8 -*-"""Created on Tue Jun 6 19:07:10 2017@author: Administrator"""from keras.models import Sequentialfrom keras.layers import Dense, Dropout, Activation, Flatten from keras.layers import Convolution2D, MaxPooling2D from keras.utils import np_utilsimport osimport pandas as pdimport numpy as npfrom tensorflow.examples.tutorials.mnist import input_datafrom keras import backend as Kimport tensorflow as tf# 全局变量 batch_size = 100nb_classes = 10epochs = 20# input image dimensions img_rows, img_cols = 28, 28# number of convolutional filters to use nb_filters = 32# size of pooling area for max pooling pool_size = (2, 2) # convolution kernel size kernel_size = (3, 3) inputfile='F:/data/kaggle/mnist/train.csv'inputfile2= 'F:/data/kaggle/mnist/test.csv'outputfile= 'F:/data/kaggle/mnist/test_label.csv'pwd = os.getcwd()os.chdir(os.path.dirname(inputfile)) train= pd.read_csv(os.path.basename(inputfile)) #从训练数据文件读取数据os.chdir(pwd)pwd = os.getcwd()os.chdir(os.path.dirname(inputfile)) test= pd.read_csv(os.path.basename(inputfile2)) #从测试数据文件读取数据os.chdir(pwd)x_train=train.iloc[:,1:785] #得到特征数据y_train=train['label']y_train = np_utils.to_categorical(y_train, 10)mnist=input_data.read_data_sets("MNIST_data/",one_hot=True) #导入数据x_test=mnist.test.imagesy_test=mnist.test.labels# 根据不同的backend定下不同的格式 if K.image_dim_ordering() == 'th': x_train=np.array(x_train) test=np.array(test) x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols) x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols) input_shape = (1, img_rows, img_cols) test = test.reshape(test.shape[0], 1, img_rows, img_cols) else: x_train=np.array(x_train) test=np.array(test) x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1) X_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1) test = test.reshape(test.shape[0], img_rows, img_cols, 1) input_shape = (img_rows, img_cols, 1) x_train = x_train.astype('float32') x_test = X_test.astype('float32') test = test.astype('float32') x_train /= 255X_test /= 255test/=255print('X_train shape:', x_train.shape) print(x_train.shape[0], 'train samples') print(x_test.shape[0], 'test samples') print(test.shape[0], 'testOuput samples') model=Sequential()#model initialmodel.add(Convolution2D(nb_filters, (kernel_size[0], kernel_size[1]), padding='same', input_shape=input_shape)) # 卷积层1 model.add(Activation('relu')) #激活层 model.add(Convolution2D(nb_filters, (kernel_size[0], kernel_size[1]))) #卷积层2 model.add(Activation('relu')) #激活层 model.add(MaxPooling2D(pool_size=pool_size)) #池化层 model.add(Dropout(0.25)) #神经元随机失活 model.add(Flatten()) #拉成一维数据 model.add(Dense(128)) #全连接层1 model.add(Activation('relu')) #激活层 model.add(Dropout(0.5)) #随机失活 model.add(Dense(nb_classes)) #全连接层2 model.add(Activation('softmax')) #Softmax评分 #编译模型 model.compile(loss='categorical_crossentropy', optimizer='adadelta', metrics=['accuracy']) #训练模型 model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs,verbose=1) model.predict(x_test)#评估模型 score = model.evaluate(x_test, y_test, verbose=0) print('Test score:', score[0]) print('Test accuracy:', score[1]) y_test=model.predict(test)sess=tf.InteractiveSession()y_test=sess.run(tf.arg_max(y_test,1))y_test=pd.DataFrame(y_test)y_test.to_csv(outputfile) |
以上就是本文的全部内容，希望对大家的学习有所帮助，也希望大家多多支持服务器之家。
原文链接：https://blog.csdn.net/u013948010/article/details/78318827
