随机森林

一、随机森林分类器 关键词：决策树、随机森林分类、print标准格式、交叉验证、plot、scipy-comb、确保随机性、常用接口。

1.1导入库和数据集

from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import load_wine

wine = load_wine()
print(wine.data)
print(wine.target)

输出：

 

 1.2决策树与RF对比

from sklearn.model_selection import train_test_split

Xtrain,Xtest,Ytrain,Ytest = train_test_split(wine.data,wine.target,test_size=0.3)

clf = DecisionTreeClassifier(random_state=0)
rfc = RandomForestClassifier(random_state=0)
clf.fit(Xtrain,Ytrain)
rfc.fit(Xtrain,Ytrain)
score_c = clf.score(Xtest,Ytest)
score_r = rfc.score(Xtest,Ytest)

print("Single Tree:{}".format(score_c)
     ,"Random Forest:{}".format(score_r))

输出：

# 二者在交叉验证下的效果对比
from sklearn.model_selection import cross_val_score
import matplotlib.pyplot as plt

rfc = RandomForestClassifier(n_estimators=25)
rfc_s = cross_val_score(rfc,wine.data,wine.target,cv=10)

clf = DecisionTreeClassifier()
clf_s = cross_val_score(clf,wine.data,wine.target,cv=10)

plt.plot(range(1,11),rfc_s,label='RandomForestClassifier')
plt.plot(range(1,11),clf_s,label='DecisionTreeClassifier')
plt.legend()
plt.show()

 

 

# 二者在10组交叉验证下的对比
rfc_l = []
clf_l = []

for i in range(1,11):
    rfc = RandomForestClassifier(n_estimators=25)
    rfc_s = cross_val_score(rfc,wine.data,wine.target,cv=10).mean()
    rfc_l.append(rfc_s)
    
    clf = DecisionTreeClassifier()
    clf_s = cross_val_score(clf,wine.data,wine.target,cv=10).mean()
    clf_l.append(clf_s)

plt.plot(range(1,11), rfc_l, label='RandomForestClassifier')
plt.plot(range(1,11), clf_l, label='DecisionTreeClassifier')
plt.legend()
plt.show()

 

 1.3n_estimators的学习曲线

# n_estimators的学习曲线
# n_jobs=-1 设定cpu里的所有core进行工作
nlr = []
for i in range(200):
    rfc = RandomForestClassifier(n_estimators=i+1,n_jobs=-1)
    rfc_s = cross_val_score(rfc,wine.data,wine.target,cv=10).mean()
    nlr.append(rfc_s)
print(max(nlr),nlr.index(max(nlr)))
plt.figure(figsize=[20,5])
plt.plot(range(1,201),nlr)
plt.show()

 

 1.4计算验证RF错误率

# 计算0.2的错误率下 n=25时rf判断错误的概率
# scipy中组合计算comb
import numpy as np
from scipy.special import comb

np.array([comb(25,i)*(0.2**i)*((1-0.2)**(25-i)) for i in range(13,26)]).sum()

 

 1.5RF的随机性

random_state,bootstrap,oob_score

rfc = RandomForestClassifier(n_estimators=20,random_state=2)
rfc = rfc.fit(Xtrain,Ytrain)
#随机森林的重要属性之一：estimators，查看森林中树的状况
rfc.estimators_[0].random_state #1872583848
for i in range(len(rfc.estimators_)):
    print(rfc.estimators_[i].random_state)

 

 

# 除了random_state之外的其他随机性
# 使用不同的训练集：bootstrap控制抽样技术的参数，有放回随机抽样，默认为True
# 不足之处：样本数n足够大时，会有约37%的训练数据被浪费掉，没有参与建模，这些数据
#被称为oob-out of bag data.可以直接使用oob来作为测试集。

# 实例化时，oob_score=True 对应属性：oob_score_
rfc = RandomForestClassifier(n_estimators=25, oob_score=True)
rfc = rfc.fit(wine.data,wine.target)

rfc.oob_score_

 

 1.6RF的常用接口

# 常用接口（与决策树一致）
rfc = RandomForestClassifier(n_estimators=25)
rfc = rfc.fit(Xtrain, Ytrain)
rfc.score(Xtest,Ytest)

print(rfc.feature_importances_) # 所有特征的重要性数值
print(rfc.apply(Xtest)) # 返回测试集中每个样本在所在树的叶子节点的索引
print(rfc.predict(Xtest)) # 预测标签
print(rfc.predict_proba(Xtest)) # 分配到每个标签的概率

1.7对比基分类器和rf的错误率

# 基分类器的准确率要大于0.5的原因
x = np.linspace(0,1,20)
y = []

for e in np.linspace(0,1,20):
    E = np.array([comb(25,i)*(e**i)*((1-e)**(25-i)) for i in range(13,26)]).sum()
    y.append(E)
plt.plot(x,y,'o-',label='when estimators are different')
plt.plot(x,x,'--',label='when estimators are same')
plt.xlabel("individual es's error")
plt.ylabel("randomforest's error")
plt.legend()
plt.show()

 

 

二、随机森林回归器

from sklearn.datasets import load_boston
from sklearn.model_selection import cross_val_score
from sklearn.ensemble import RandomForestRegressor
import sklearn

boston = load_boston()
reger = RandomForestRegressor(n_estimators=100,random_state=0)
cvs = cross_val_score(reger,boston.data,boston.target,cv=10
               ,scoring='neg_mean_squared_error')
print(cvs)
print(sorted(sklearn.metrics.SCORERS.keys()))
# 打印出可以作为scoring参数的选项