华为云技术分享python成长之路机器学习：10+分类算法汇总学习(代码片段)

  1 from sklearn import svm
  2 from sklearn.neural_network import MLPClassifier
  3 from sklearn.tree import DecisionTreeClassifier
  4 from sklearn.neighbors import KNeighborsClassifier
  5 from sklearn.naive_bayes import BernoulliNB
  6 from sklearn.naive_bayes import GaussianNB
  7 from sklearn.naive_bayes import MultinomialNB
  8 from sklearn.linear_model import LogisticRegression
  9 from sklearn.ensemble import RandomForestClassifier
 10 from sklearn.ensemble import AdaBoostClassifier
 11 import xgboost
 12 import lightgbm
 13 import time
 14 from sklearn import metrics
 15 from sklearn.model_selection import train_test_split
 16 import random
 17 import numpy
 18 from openpyxl import Workbook,load_workbook
 19 
 20 
 21 def train_test_algorithm(model, x_train, y_train, x_test, y_test):
 22     print(‘begin to train……‘)
 23     start_time = time.time()
 24     model.fit(x_train, y_train)
 25     print(‘begin to predict……‘)
 26     y_pred_model = model.predict(x_test)
 27     classifier = str(model).split(‘(‘)[0]
 28     print(‘begin to get the result of Classifier_Type:...‘.format(classifier))
 29     score = metrics.accuracy_score(y_test, y_pred_model)
 30     print(‘The accuracy score of 0 is: 1%‘.format(classifier, round(score * 100, 2)))
 31     end_time = time.time()
 32     use_time = end_time - start_time
 33     print(‘The time of using 0 is :1‘.format(classifier, round(use_time, 2)))
 34     print(‘ test end!

‘.format(classifier))
 35     return score, round(use_time, 2), classifier
 36 
 37 
 38 def get_redball():
 39     redlist = []
 40     while len(redlist) < 7:
 41         temp = random.randint(1, 34)
 42         if temp not in redlist:
 43             redlist.append(temp)
 44     return redlist
 45 
 46 
 47 def get_blueball():
 48     return random.randint(1, 17)
 49 
 50 
 51 def get_data():
 52     x_temp, y_temp = [], []
 53     for i in range(2000):
 54         x_temp.append(get_redball())
 55         y_temp.append(get_blueball())
 56     x = numpy.array(x_temp)
 57     y = numpy.array(y_temp)
 58     return x, y
 59 
 60 def get_data2():
 61     import csv
 62     data = csv.reader(open(‘winequality-red.csv‘,‘r‘))
 63     labels,xlist=[],[]
 64     firstline = True
 65     for i in data:
 66         if firstline:
 67             firstline = False
 68         else:
 69             r = i[0].split(‘;‘)
 70             labels.append(float(r[-1]))
 71             r.pop()
 72             floatrow = [float(num) for num in r]
 73             xlist.append(floatrow)
 74     # 转化为numpy格式
 75     x = numpy.array(xlist)
 76     y = numpy.array(labels)
 77     return x,y
 78 
 79 
 80 if __name__ == ‘__main__‘:
 81     # 定义各类分类算法
 82     svc = svm.SVC()  # 支持向量机，SVM
 83     mlp = MLPClassifier()  # 多层感知机，MLP
 84     dtc = DecisionTreeClassifier()  # 决策树,DT
 85     knc = KNeighborsClassifier()  # K最近邻,KNN
 86     bnb = BernoulliNB()  # 伯努利贝叶斯，BNB
 87     gnb = GaussianNB()  # 高斯贝叶斯,GNB
 88     mnb = MultinomialNB()  # 朴素贝叶斯，MNB
 89     lgr = LogisticRegression()  # 逻辑回归，LGR
 90     rfc = RandomForestClassifier()  # 随机森林，RFC
 91     abc = AdaBoostClassifier()  # AdaBoost
 92     xgb = xgboost.XGBClassifier()  # xgboost
 93     gbm = lightgbm.LGBMClassifier()  # lightgbm
 94 
 95     # 训练/测试数据生成
 96     # 真实红酒销售数据
 97     x, y = get_data2()
 98     # 随机生成的双色球红蓝球数据
 99     # x,y = get_data()
100     x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.20, random_state=531)
101     classify_type_list = [svc, mlp, dtc, knc, bnb, gnb, mnb, lgr, rfc, abc, xgb, gbm]
102     test_result = 
103     tablename = ‘wine_result‘
104     wb = load_workbook(‘result.xlsx‘)
105     try:
106         ws = wb[tablename]
107     except:
108         ws = wb.create_sheet(tablename)
109     ws.append([‘classifer_type‘,‘score‘,‘use_time‘])
110     for i, type in enumerate(classify_type_list):
111         score, use_time, classifier = train_test_algorithm(type, x_train, y_train, x_test, y_test)
112         ws.append([classifier,str(round(100*score,2))+‘%‘,use_time])
113     wb.save(‘result.xlsx‘)

1 base_score=0.5, booster=‘gbtree‘, colsample_bylevel=1,
2               colsample_bynode=1, colsample_bytree=1, gamma=0,
3               learning_rate=0.1, max_delta_step=0, max_depth=3,
4               min_child_weight=1, missing=None, n_estimators=100, n_jobs=1,
5               nthread=None, objective=‘binary:logistic‘, random_state=0,
6               reg_alpha=0, reg_lambda=1, scale_pos_weight=1, seed=None,
7               silent=None, subsample=1, verbosity=1