1.3 机器学习方法之分类问题
1. 分类问题
分类问题是监督学习的一个核心问题,它从数据中心学习一个分类决策函数或分类模型,对新的输入进行输出预测,输出变量取有限个离散值。
决策树、贝叶斯、SVM支持向量机、逻辑回归、集成学习。
1.1.决策树
给定训练数据,如何构建决策树:
以下三步完成。
(1) 特征选择
决定用哪个特征来划分特征空间,选出对训练数据集具有分类能力的特征。
算法
(2) 决策树生成
在决策树各个点上按照一定的方法选择特征,递归构建决策树
(3) 决策树剪枝
1.2. 贝叶斯
参见博文《机器学习之贝叶斯样本分类》
1.3. 支持向量机SVM
一种有监督学习的方法,尝试寻找一个最优决策边界,使距离两个类别最近的样本最远。
from sklearn import svm
# from beyes_classify import load_data
import numpy as np
import matplotlib.pyplot as plt
def load_data():
'''
load file data
'''
with open('F:/study/AI/src/ml/sklearn/bayes.csv') as csv_file:
data = csv.reader(csv_file)
first_row = next(data)
n_sample = int(first_row[0])
n_feature = int(first_row[1])
train_sets = np.empty((n_sample, 2), dtype=np.float64)
target_sets = np.empty((n_sample), dtype=np.int)
for i, ir in enumerate(data):
train_sets[i] = np.asarray(ir[:-1], dtype=np.float64)
target_sets[i] = np.asarray(ir[-1], dtype=np.int)
return n_sample, n_feature, train_sets, target_sets
if __name__ == '__main__':
n_sample, n_feature, train_sets, target_sets = load_data()
clf = svm.SVC(kernel='linear')
clf.fit(train_sets, target_sets)
# clf.fit(train_sets[:,0], train_sets[:,1])
plt.scatter(train_sets[:5,0], train_sets[:5,1], color = 'black')
plt.scatter(train_sets[5:,0], train_sets[5:,1], color = 'blue')
w=clf.coef_[0]
a=-w[0]/w[1]
xx=np.linspace(0,5)#产生-5到5的线性连续值,间隔为1
yy=a*xx-(clf.intercept_[0])/w[1] #clf.intercept_[0]是w3.即为公式a1*x1+a2*x2+w3中的w3。(clf.intercept_[0])/w[1]即为直线的截距
plt.plot(xx, yy)
plt.show()
10个训练样本,二分类0/1。
1.4. 逻辑回归
参见博文《Tensorflow之逻辑回归二分类以及交叉熵》
import numpy as np
import matplotlib.pyplot as plt
def loadDataSet():
X_train = np.empty((100, 3))
y_train = np.empty((100, 1), dtype=np.int)
fr = open('D:\\02-Work\\01-AI\\csdn_code\\machine-learning\\classification\\logistic_data.txt')
i = 0
for line in fr.readlines():
lineArr = line.strip().split()
temp = [1.0, float(lineArr[0]), float(lineArr[1])]
X_train[i] = np.asarray(temp)
y_train[i] = np.asarray(int(lineArr[2]), dtype=np.int)
i += 1
return X_train,y_train
class log(object):
def __init__(self):
self.W=None
def sigmoid(self,X):
return 1.0 / (1.0 + np.exp(-X))
def loss(self,X_train,y_train):
m,n=X_train.shape
h=self.sigmoid(X_train.dot(self.W))
# 代价函数 交叉熵
loss=(y_train.T.dot(np.log(h))+(1-y_train).T.dot(np.log(1-h)))/-m
loss=loss[0,0]
# 代价函数求偏导
dW=X_train.T.dot((h - y_train)) / m
return loss,dW
def train(self,X_train,y_train,learn_rate=0.01,iters=5000):
m,n=X_train.shape
print(m,n)
self.W=np.random.rand(n,1)
loss_list = []
#梯度下降法迭代,迭代结束得出的最后W,即决策边界权重
for i in range(iters):
loss,dW=self.loss(X_train,y_train)
self.W-=learn_rate*dW
loss_list.append(loss)
if i % 1000 == 0:
print('iters = %d,loss = %f' % (i, loss))
print(self.W)
return loss_list, self.W
def decision_boundary(weight):
'''决策分界线'''
X = np.arange(-5,5,0.5)
y = (-weight[0]-weight[1]*X)/weight[2]
plt.plot(X,y)
def dataLabelPlt(X_train,y_train):
'''样本数据散点图'''
m,n=X_train.shape
for i in range(m):
color = ''
if y_train[i] == 1:
color = 'b'
else:
color = 'r'
plt.scatter(X_train[i][1],X_train[i][2],color=color)
if __name__ == '__main__':
X_train,y_train = loadDataSet()
class_object = log()
loss_list, weight = class_object.train(X_train, y_train)
dataLabelPlt(X_train,y_train)
decision_boundary(weight)
plt.show()
样本数据logistic_data.txt:
-0.017612 14.053064 0
-1.395634 4.662541 1
-0.752157 6.538620 0
-1.322371 7.152853 0
0.423363 11.054677 0
0.406704 7.067335 1
0.667394 12.741452 0
-2.460150 6.866805 1
0.569411 9.548755 0
-0.026632 10.427743 0
0.850433 6.920334 1
1.347183 13.175500 0
1.176813 3.167020 1
-1.781871 9.097953 0
-0.566606 5.749003 1
0.931635 1.589505 1
-0.024205 6.151823 1
-0.036453 2.690988 1
-0.196949 0.444165 1
1.014459 5.754399 1
1.985298 3.230619 1
-1.693453 -0.557540 1
-0.576525 11.778922 0
-0.346811 -1.678730 1
-2.124484 2.672471 1
1.217916 9.597015 0
-0.733928 9.098687 0
-3.642001 -1.618087 1
0.315985 3.523953 1
1.416614 9.619232 0
-0.386323 3.989286 1
0.556921 8.294984 1
1.224863 11.587360 0
-1.347803 -2.406051 1
1.196604 4.951851 1
0.275221 9.543647 0
0.470575 9.332488 0
-1.889567 9.542662 0
-1.527893 12.150579 0
-1.185247 11.309318 0
-0.445678 3.297303 1
1.042222 6.105155 1
-0.618787 10.320986 0
1.152083 0.548467 1
0.828534 2.676045 1
-1.237728 10.549033 0
-0.683565 -2.166125 1
0.229456 5.921938 1
-0.959885 11.555336 0
0.492911 10.993324 0
0.184992 8.721488 0
-0.355715 10.325976 0
-0.397822 8.058397 0
0.824839 13.730343 0
1.507278 5.027866 1
0.099671 6.835839 1
-0.344008 10.717485 0
1.785928 7.718645 1
-0.918801 11.560217 0
-0.364009 4.747300 1
-0.841722 4.119083 1
0.490426 1.960539 1
-0.007194 9.075792 0
0.356107 12.447863 0
0.342578 12.281162 0
-0.810823 -1.466018 1
2.530777 6.476801 1
1.296683 11.607559 0
0.475487 12.040035 0
-0.783277 11.009725 0
0.074798 11.023650 0
-1.337472 0.468339 1
-0.102781 13.763651 0
-0.147324 2.874846 1
0.518389 9.887035 0
1.015399 7.571882 0
-1.658086 -0.027255 1
1.319944 2.171228 1
2.056216 5.019981 1
-0.851633 4.375691 1
-1.510047 6.061992 0
-1.076637 -3.181888 1
1.821096 10.283990 0
3.010150 8.401766 1
-1.099458 1.688274 1
-0.834872 -1.733869 1
-0.846637 3.849075 1
1.400102 12.628781 0
1.752842 5.468166 1
0.078557 0.059736 1
0.089392 -0.715300 1
1.825662 12.693808 0
0.197445 9.744638 0
0.126117 0.922311 1
-0.679797 1.220530 1
0.677983 2.556666 1
0.761349 10.693862 0
-2.168791 0.143632 1
1.388610 9.341997 0
0.317029 14.739025 0