首页 机器学习

# 手写数字识别

# 一、前置库

import numpy as np
import matplotlib.pyplot as plt
import os
from array import array as pyarray
import struct
from sklearn.metrics import accuracy_score,classification_report
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import MultinomialNB
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC

# 二、数据集

数据集采用的从官网上下载的 mnist 数据集，下载地址为 http://yann.lecun.com/exdb/mnist/

一共包含了 4 个部分

1、训练数据集：train−images−idx3−ubyte.gztrain-images-idx3-ubyte.gz （9.45 MB，包含 60,000 个样本）

2、训练数据集标签:train−labels−idx1−ubyte.gztrain-labels-idx1-ubyte.gz（28.2 KB，包含 60,000 个标签）

3、测试数据集：t10k−images−idx3−ubyte.gzt10k-images-idx3-ubyte.gz（1.57 MB ，包含 10,000 个样本）

4、测试数据集标签：t10k−labels−idx1−ubyte.gzt10k-labels-idx1-ubyte.gz（4.43 KB，包含 10,000 个样本的标签）

可以看到这些数据集和以往的数据集不太一样，需要一些特殊的方法才可以将它们读出来，处理代码如下：

def load_mnist(image_file,label_file,path="F:/Mechine Learning/Handy Writing/mnist"):
    digits = np.arange(10)
    fname_image = os.path.join(path,image_file)
    fname_label = os.path.join(path,label_file)
    
    flbl = open(fname_label,'rb')
    magic_nr,size = struct.unpack(">II",flbl.read(8))
    lbl = pyarray("b",flbl.read())
    flbl.close()
    
    fimg = open(fname_image,'rb')
    magic_nr,size,rows,cols = struct.unpack(">IIII",fimg.read(16))
    img = pyarray("B",fimg.read())
    fimg.close()
    
    ind = [k for k in range(size) if lbl[k] in digits ]
    N = len(ind)
    
    images = np.zeros((N,rows*cols),dtype=np.uint8)
    labels = np.zeros((N,1),dtype=np.uint8)
    for i in range(len(ind)):
        images[i] = np.array(img[ ind[i]*rows*cols : (ind[i]+1)*rows*cols]).reshape((1,rows*cols))
        labels[i] = lbl[ind[i]]
    
    return images,labels
train_image,train_label = load_mnist("train-images.idx3-ubyte","train-labels.idx1-ubyte")
test_image,test_label = load_mnist("t10k-images.idx3-ubyte","t10k-labels.idx1-ubyte")
这样所有数据被读进来了，下面看看部分图片
def show_image(imgdata,imgtarget,show_column,show_row):
    for index,(im,it) in enumerate(list(zip(imgdata,imgtarget))):
        xx = im.reshape(28,28)
        plt.subplots_adjust(left=1,bottom=None,right=3,top=2,wspace=None,hspace=None)
        plt.subplot(show_row,show_column,index+1)
        plt.axis('off')
        plt.imshow(xx,cmap='gray',interpolation='nearest')
        plt.title('label:%i'%it)
show_image(train_image[:50],train_label[:50],10,5)
显示如下：
部分图片显示
对数据再进行一些处理，进行归一化
train_image = [im/255.0 for im in train_image]
test_image = [im/255.0 for im in test_image]
接下来用不同的方法进行分类，大部分模型都是用的sklearnsklearn 里的
# 一、KNeighbors
from sklearn.metrics import accuracy_score,classification_report
from sklearn.neighbors import KNeighborsClassifier
knc = KNeighborsClassifier(n_neighbors=10)
knc.fit(train_image,train_label.ravel())
predict = knc.predict(test_image)\
print("accuracy_score:%.4lf" % accuracy_score(predict,test_label))
print("Classification report for classifier %s:\n%s\n" % (knc,classification_report(test_label,predict)))
结果如下：
KNeighbors结果
准确率为：accuracy_score:0.9665accuracy\_score:0.9665
# 二、MultinomialNB
from sklearn.naive_bayes import MultinomialNB
mnb = MultinomialNB()
mnb.fit(train_image,train_label.ravel())
predict1 = mnb.predict(test_image)
print("accuracy_score:%.4lf" % accuracy_score(predict1,test_label))
print("Classification report for classifier %s:\n%s\n" % (mnb, classification_report(test_label, predict1)))
结果如下：
MultinomialNB结果
# 三、决策树
from sklearn.tree import DecisionTreeClassifier
dtc = DecisionTreeClassifier()
dtc.fit(train_image,train_label.ravel())
predict2 = dtc.predict(test_image)
print("accuracy_score:%.4lf" % accuracy_score(predict2,test_label))
print("Classification report for classifier %s:\n%s\n" % (dtc, classification_report(test_label, predict2)))
结果如下：
决策树结果
# 四、随机森林
from sklearn.ensemble import RandomForestClassifier

rfc = RandomForestClassifier()
rfc.fit(train_image,train_label.ravel())
predict3 = rfc.predict(test_image)
print("accuracy_score:%.4lf" % accuracy_score(predict3,test_label))
print("Classification report for classifier %s:\n%s\n" % (rfc, classification_report(test_label, predict3)))

结果如下：
随机森林结果
# 五、逻辑回归
from sklearn.linear_model import LogisticRegression
lr = LogisticRegression()
lr.fit(train_image,train_label.ravel())
predict4 = lr.predict(test_image)
print("accuracy_score: %.4lf" % accuracy_score(predict4,test_label))
print("Classification report for classifier %s:\n%s\n" % (lr, classification_report(test_label, predict4)))
结果如下：
逻辑回归结果
# 六、SVM
from sklearn.svm import SVC
svc = SVC()
svc.fit(train_image,train_label.ravel())
predict5 = svc.predict(test_image)
print("accuracy_score: %.4lf" % accuracy_score(predict5,test_label))
print("Classification report for classifier %s:\n%s\n" % (svc, classification_report(test_label, predict5)))
结果如下：
SVM结果