常見的機器學習算法（十）決策樹、隨機森林、極限樹

直接調用sklearn的API：

from sklearn import tree                              #決策樹分類器#
module = tree.DecisionTreeClassifier(criterion='gini')
module.fit(x, y)
module.score(x, y)
module.predict(test)

from sklearn.ensemble import RandomForestClassifier  #隨機森林#
from sklearn.ensemble import RandomForestRegressor
module = RandomForestClassifier()
module.fit(x, y)
module.predict(test)

整體代碼1：

from sklearn.datasets import load_iris
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import cross_val_score  # 交叉驗證函數

iris = load_iris()
dtc = DecisionTreeClassifier()
dtc.fit(iris.data, iris.target)

# 使用決策樹分類器作爲評估模型，iris.data鳶尾花數據作爲特徵，iris.target鳶尾花分類標籤作爲目標結果，設定cv爲10，使用10折交叉驗證，得到最終的交叉驗證得分。
print(cross_val_score(dtc, iris.data, iris.target, cv=10))
print('-----------------------------')
print(dtc.predict(iris.data))#預測出對應標籤

整體代碼2：

from sklearn.model_selection import cross_val_score #交叉驗證
from sklearn.datasets import make_blobs
from sklearn.tree import DecisionTreeClassifier #決策樹
from sklearn.ensemble import RandomForestClassifier #隨機森林
from sklearn.ensemble import ExtraTreesClassifier #極限樹

#創建100個類,共10000個樣本，每個樣本10個特徵
x ,y = make_blobs(n_samples=10000, n_features=10, centers=100, random_state=0)

dtc = DecisionTreeClassifier(random_state=0)
score_dtc = cross_val_score(dtc, x, y)
print('score_dtc均值: ', score_dtc.mean())

rfc = RandomForestClassifier(n_estimators=10, random_state=0)
score_rfc = cross_val_score(rfc, x, y)
print('score_rfc均值: ', score_rfc.mean())

etc = ExtraTreesClassifier(n_estimators=10, random_state=0)
score_etc = cross_val_score(etc, x, y)
print('score_etc均值: ', score_etc.mean())

性能對比： 極限樹 > 隨機森林 > 決策樹