如何用卷积神经网络CNN识别手写数字集资料.docx

1、如何用卷积神经网络CNN识别手写数字集资料如何用卷积神经网络CNN识别手写数字集？前几天用CNN识别手写数字集，后来看到kaggle上有一个比赛是识别手写数字集的，已经进行了一年多了，目前有1179个有效提交，最高的是100%，我做了一下，用keras做的，一开始用最简单的MLP，准确率只有98.19%，然后不断改进，现在是99.78%，然而我看到排名第一是100%，心碎 = =，于是又改进了一版，现在把最好的结果记录一下，如果提升了再来更新。手写数字集相信大家应该很熟悉了，这个程序相当于学一门新语言的“Hello World”，或者mapreduce的“WordCount”：）这里就不多做介

2、绍了，简单给大家看一下：复制代码 1 # Author：Charlotte 2 # Plot mnist dataset 3 from keras.datasets import mnist 4 import matplotlib.pyplot as plt 5 # load the MNIST dataset 6 (X_train, y_train), (X_test, y_test) = mnist.load_data() 7 # plot 4 images as gray scale 8 plt.subplot(221) 9 plt.imshow(X_train0, cmap=plt.g

3、et_cmap(PuBuGn_r)10 plt.subplot(222)11 plt.imshow(X_train1, cmap=plt.get_cmap(PuBuGn_r)12 plt.subplot(223)13 plt.imshow(X_train2, cmap=plt.get_cmap(PuBuGn_r)14 plt.subplot(224)15 plt.imshow(X_train3, cmap=plt.get_cmap(PuBuGn_r)16 # show the plot17 plt.show()复制代码图： 1.BaseLine版本一开始我没有想过用CNN做，因为比较耗时，所以

4、想看看直接用比较简单的算法看能不能得到很好的效果。之前用过机器学习算法跑过一遍，最好的效果是SVM，96.8%（默认参数，未调优），所以这次准备用神经网络做。BaseLine版本用的是MultiLayer Percepton（多层感知机）。这个网络结构比较简单，输入-隐含-输出。隐含层采用的rectifier linear unit，输出直接选取的softmax进行多分类。网络结构：代码：复制代码 1 # coding:utf-8 2 # Baseline MLP for MNIST dataset 3 import numpy 4 from keras.datasets import mni

5、st 5 from keras.models import Sequential 6 from keras.layers import Dense 7 from keras.layers import Dropout 8 from keras.utils import np_utils 9 10 seed = 711 numpy.random.seed(seed)12 #加载数据13 (X_train, y_train), (X_test, y_test) = mnist.load_data()14 15 num_pixels = X_train.shape1 * X_train.shape2

6、16 X_train = X_train.reshape(X_train.shape0, num_pixels).astype(float32)17 X_test = X_test.reshape(X_test.shape0, num_pixels).astype(float32)18 19 X_train = X_train / 25520 X_test = X_test / 25521 22 # 对输出进行one hot编码23 y_train = np_utils.to_categorical(y_train)24 y_test = np_utils.to_categorical(y_t

7、est)25 num_classes = y_test.shape126 27 # MLP模型28 def baseline_model():29 model = Sequential()30 model.add(Dense(num_pixels, input_dim=num_pixels, init=normal, activation=relu)31 model.add(Dense(num_classes, init=normal, activation=softmax)32 model.summary()33 pile(loss=categorical_crossentropy, opt

8、imizer=adam, metrics=accuracy)34 return model35 36 # 建立模型37 model = baseline_model()38 39 # Fit40 model.fit(X_train, y_train, validation_data=(X_test, y_test), nb_epoch=10, batch_size=200, verbose=2)41 42 #Evaluation43 scores = model.evaluate(X_test, y_test, verbose=0)44 print(Baseline Error: %.2f%

9、% (100-scores1*100)#输出错误率复制代码结果：复制代码 1 Layer (type) Output Shape Param # Connected to 2 = 3 dense_1 (Dense) (None, 784) 615440 dense_input_100 4 _ 5 dense_2 (Dense) (None, 10) 7850 dense_100 6 = 7 Total params: 623290 8 _ 9 Train on 60000 samples, validate on 10000 samples10 Epoch 1/1011 3s - loss:

10、0.2791 - acc: 0.9203 - val_loss: 0.1420 - val_acc: 0.957912 Epoch 2/1013 3s - loss: 0.1122 - acc: 0.9679 - val_loss: 0.0992 - val_acc: 0.969914 Epoch 3/1015 3s - loss: 0.0724 - acc: 0.9790 - val_loss: 0.0784 - val_acc: 0.974516 Epoch 4/1017 3s - loss: 0.0509 - acc: 0.9853 - val_loss: 0.0774 - val_ac

11、c: 0.977318 Epoch 5/1019 3s - loss: 0.0366 - acc: 0.9898 - val_loss: 0.0626 - val_acc: 0.979420 Epoch 6/1021 3s - loss: 0.0265 - acc: 0.9930 - val_loss: 0.0639 - val_acc: 0.979722 Epoch 7/1023 3s - loss: 0.0185 - acc: 0.9956 - val_loss: 0.0611 - val_acc: 0.981124 Epoch 8/1025 3s - loss: 0.0150 - acc

12、: 0.9967 - val_loss: 0.0616 - val_acc: 0.981626 Epoch 9/1027 4s - loss: 0.0107 - acc: 0.9980 - val_loss: 0.0604 - val_acc: 0.982128 Epoch 10/1029 4s - loss: 0.0073 - acc: 0.9988 - val_loss: 0.0611 - val_acc: 0.981930 Baseline Error: 1.81%复制代码可以看到结果还是不错的，正确率98.19%，错误率只有1.81%，而且只迭代十次效果也不错。这个时候我还是没想到去用

13、CNN，而是想如果迭代100次，会不会效果好一点？于是我迭代了100次，结果如下：Epoch 100/1008s - loss: 4.6181e-07 - acc: 1.0000 - val_loss: 0.0982 - val_acc: 0.9854Baseline Error: 1.46%从结果中可以看出，迭代100次也只提高了0.35%，没有突破99%，所以就考虑用CNN来做。 2.简单的CNN网络keras的CNN模块还是很全的，由于这里着重讲CNN的结果，对于CNN的基本知识就不展开讲了。网络结构：代码：复制代码 1 #coding: utf-8 2 #Simple CNN 3 im

14、port numpy 4 from keras.datasets import mnist 5 from keras.models import Sequential 6 from keras.layers import Dense 7 from keras.layers import Dropout 8 from keras.layers import Flatten 9 from keras.layers.convolutional import Convolution2D10 from keras.layers.convolutional import MaxPooling2D11 fr

15、om keras.utils import np_utils12 13 seed = 714 numpy.random.seed(seed)15 16 #加载数据17 (X_train, y_train), (X_test, y_test) = mnist.load_data()18 # reshape to be sampleschannelswidthheight19 X_train = X_train.reshape(X_train.shape0, 1, 28, 28).astype(float32)20 X_test = X_test.reshape(X_test.shape0, 1,

16、 28, 28).astype(float32)21 22 # normalize inputs from 0-255 to 0-123 X_train = X_train / 25524 X_test = X_test / 25525 26 # one hot encode outputs27 y_train = np_utils.to_categorical(y_train)28 y_test = np_utils.to_categorical(y_test)29 num_classes = y_test.shape130 31 # define a simple CNN model32

17、def baseline_model():33 # create model34 model = Sequential()35 model.add(Convolution2D(32, 5, 5, border_mode=valid, input_shape=(1, 28, 28), activation=relu)36 model.add(MaxPooling2D(pool_size=(2, 2)37 model.add(Dropout(0.2)38 model.add(Flatten()39 model.add(Dense(128, activation=relu)40 model.add(

18、Dense(num_classes, activation=softmax)41 # Compile model42 pile(loss=categorical_crossentropy, optimizer=adam, metrics=accuracy)43 return model44 45 # build the model46 model = baseline_model()47 48 # Fit the model49 model.fit(X_train, y_train, validation_data=(X_test, y_test), nb_epoch=10, batch_si

19、ze=128, verbose=2)50 51 # Final evaluation of the model52 scores = model.evaluate(X_test, y_test, verbose=0)53 print(CNN Error: %.2f% % (100-scores1*100)复制代码结果：复制代码 1 _ 2 Layer (type) Output Shape Param # Connected to 3 = 4 convolution2d_1 (Convolution2D) (None, 32, 24, 24) 832 convolution2d_input_1

20、00 5 _ 6 maxpooling2d_1 (MaxPooling2D) (None, 32, 12, 12) 0 convolution2d_100 7 _ 8 dropout_1 (Dropout) (None, 32, 12, 12) 0 maxpooling2d_100 9 _10 flatten_1 (Flatten) (None, 4608) 0 dropout_10011 _12 dense_1 (Dense) (None, 128) 589952 flatten_10013 _14 dense_2 (Dense) (None, 10) 1290 dense_10015 =1

21、6 Total params: 59207417 _18 Train on 60000 samples, validate on 10000 samples19 Epoch 1/1020 32s - loss: 0.2412 - acc: 0.9318 - val_loss: 0.0754 - val_acc: 0.976621 Epoch 2/1022 32s - loss: 0.0726 - acc: 0.9781 - val_loss: 0.0534 - val_acc: 0.982923 Epoch 3/1024 32s - loss: 0.0497 - acc: 0.9852 - v

22、al_loss: 0.0391 - val_acc: 0.985825 Epoch 4/1026 32s - loss: 0.0413 - acc: 0.9870 - val_loss: 0.0432 - val_acc: 0.985427 Epoch 5/1028 34s - loss: 0.0323 - acc: 0.9897 - val_loss: 0.0375 - val_acc: 0.986929 Epoch 6/1030 36s - loss: 0.0281 - acc: 0.9909 - val_loss: 0.0424 - val_acc: 0.986431 Epoch 7/1

23、032 36s - loss: 0.0223 - acc: 0.9930 - val_loss: 0.0328 - val_acc: 0.989333 Epoch 8/1034 36s - loss: 0.0198 - acc: 0.9939 - val_loss: 0.0381 - val_acc: 0.988035 Epoch 9/1036 36s - loss: 0.0156 - acc: 0.9954 - val_loss: 0.0347 - val_acc: 0.988437 Epoch 10/1038 36s - loss: 0.0141 - acc: 0.9955 - val_l

24、oss: 0.0318 - val_acc: 0.989339 CNN Error: 1.07%复制代码迭代的结果中，loss和acc为训练集的结果，val_loss和val_acc为验证机的结果。从结果上来看，效果不错，比100次迭代的MLP（1.46%）提升了0.39%，CNN的误差率为1.07%。这里的CNN的网络结构还是比较简单的，如果把CNN的结果再加几层，边复杂一代，结果是否还能提升？ 3.Larger CNN这一次我加了几层卷积层，代码：复制代码 1 # Larger CNN 2 import numpy 3 from keras.datasets import mnist 4

25、from keras.models import Sequential 5 from keras.layers import Dense 6 from keras.layers import Dropout 7 from keras.layers import Flatten 8 from keras.layers.convolutional import Convolution2D 9 from keras.layers.convolutional import MaxPooling2D10 from keras.utils import np_utils1112 seed = 713 nu

26、mpy.random.seed(seed)14 # load data15 (X_train, y_train), (X_test, y_test) = mnist.load_data()16 # reshape to be samplespixelswidthheight17 X_train = X_train.reshape(X_train.shape0, 1, 28, 28).astype(float32)18 X_test = X_test.reshape(X_test.shape0, 1, 28, 28).astype(float32)19 # normalize inputs from 0-255 to 0-120 X_train = X_train / 25521 X_test = X_test / 25522 # one hot encode outputs23 y_train = np_utils.to_categorical(y_train)24 y_test = np_utils.to_categorical(y_test)25 num_classes = y_t