[分類算法]常用功能實作

前言：分類是機器學習中的重要的一種功能，在機器學習的研究曆史中，誕生了大量的分類算法，而每種算法都有其優勢和不足。

本文彙總了常用的分類算法及其實作方式，友善快速查詢使用。（本文使用鸢尾花資料集，是三類别分類）

以下9種分類算法，使用相同的資料進行訓練和測試，在測試集上的準确率(accuracy)分别為：

1.随機森林：100%

2.決策樹：100%

3.K近鄰：100%

4.支援向量機：100%

5.邏輯回歸：96.67%

6.線性支援向量機：100%

1 import numpy as np
  2 import pandas as pd
  3 import matplotlib as mpl
  4 import matplotlib.pyplot as plt
  5 import sklearn
  6 from sklearn import datasets
  7 from sklearn.metrics import accuracy_score
  8 
  9 from sklearn.ensemble import RandomForestClassifier
 10 from sklearn.tree import DecisionTreeClassifier
 11 
 12 from sklearn.neighbors import KNeighborsClassifier
 13 from sklearn.svm import SVC, LinearSVC
 14 from sklearn.linear_model import LogisticRegression
 15 
 16 from sklearn.linear_model import SGDClassifier
 17 from sklearn.linear_model import Perceptron
 18 from sklearn.naive_bayes import GaussianNB
 19 
 20 from sklearn.model_selection import train_test_split
 21 from sklearn.model_selection import cross_val_score
 22 
 23 from sklearn.model_selection import GridSearchCV
 24 
 25 iris = datasets.load_iris()
 26 x,y = iris.data,iris.target
 27 
 28 x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=0.2,random_state=0)
 29 
 30 res = []
 31 
 32 #1. 随機森林分類
 33 print('随機森林分類')
 34 clf = RandomForestClassifier(n_estimators=100)
 35 clf.fit(x_train, y_train)
 36 cross_score = cross_val_score(clf, x_train, y_train, cv=3, scoring="accuracy")
 37 print(cross_score)
 38 y_predict = clf.predict(x_test)
 39 score = accuracy_score(y_test,y_predict)
 40 res.append(score)
 41 print()
 42 
 43 #2. 決策樹分類
 44 print('決策樹分類')
 45 clf = DecisionTreeClassifier()
 46 clf.fit(x_train, y_train)
 47 cross_score = cross_val_score(clf, x_train, y_train, cv=3, scoring="accuracy")
 48 print(cross_score)
 49 y_predict = clf.predict(x_test)
 50 score = accuracy_score(y_test,y_predict)
 51 res.append(score)
 52 print()
 53 
 54 #3. KNN
 55 print('KNN')
 56 clf = KNeighborsClassifier(n_neighbors = 13)
 57 clf.fit(x_train, y_train)
 58 cross_score = cross_val_score(clf, x_train, y_train, cv=3, scoring="accuracy")
 59 print(cross_score)
 60 y_predict = clf.predict(x_test)
 61 score = accuracy_score(y_test,y_predict)
 62 res.append(score)
 63 print()
 64 
 65 #4. SVM分類
 66 print('SVM')
 67 clf = SVC(gamma='scale')
 68 clf.fit(x_train, y_train)
 69 cross_score = cross_val_score(clf, x_train, y_train, cv=3, scoring="accuracy")
 70 print(cross_score)
 71 y_predict = clf.predict(x_test)
 72 score = accuracy_score(y_test,y_predict)
 73 res.append(score)
 74 print()
 75 
 76 #5. 邏輯回歸分類
 77 print('LogisticRegression')
 78 clf = LogisticRegression(solver='lbfgs',multi_class='ovr')
 79 clf.fit(x_train, y_train)
 80 cross_score = cross_val_score(clf, x_train, y_train, cv=3, scoring="accuracy")
 81 print(cross_score)
 82 y_predict = clf.predict(x_test)
 83 score = accuracy_score(y_test,y_predict)
 84 res.append(score)
 85 print()
 86 
 87 #6. linear svm分類
 88 print('linear SVM')
 89 clf = LinearSVC(max_iter=10000)
 90 clf.fit(x_train, y_train)
 91 cross_score = cross_val_score(clf, x_train, y_train, cv=3, scoring="accuracy")
 92 print(cross_score)
 93 y_predict = clf.predict(x_test)
 94 score = accuracy_score(y_test,y_predict)
 95 res.append(score)
 96 print()
 97 
 98 #7. 随機梯度下降分類
 99 print('SGD')
100 clf = SGDClassifier(max_iter=1000,tol=1e-3)
101 clf.fit(x_train, y_train)
102 cross_score = cross_val_score(clf, x_train, y_train, cv=3, scoring="accuracy")
103 print(cross_score)
104 y_predict = clf.predict(x_test)
105 score = accuracy_score(y_test,y_predict)
106 res.append(score)
107 print()
108 
109 #8. 感覺機分類
110 print('Perceptron')
111 clf = Perceptron(max_iter=1000,tol=1e-3)
112 clf.fit(x_train, y_train)
113 cross_score = cross_val_score(clf, x_train, y_train, cv=3, scoring="accuracy")
114 print(cross_score)
115 y_predict = clf.predict(x_test)
116 score = accuracy_score(y_test,y_predict)
117 res.append(score)
118 print()
119 
120 #9. 樸素貝葉斯分類
121 print('Naive Bayes')
122 clf = GaussianNB()
123 clf.fit(x_train, y_train)
124 cross_score = cross_val_score(clf, x_train, y_train, cv=3, scoring="accuracy")
125 print(cross_score)
126 y_predict = clf.predict(x_test)
127 score = accuracy_score(y_test,y_predict)
128 res.append(score)
129 print()
130 
131 #10. 得分比較
132 print(res)