机器学习部分核心代码XGBClassfier版

FBP_ML_XGBClassifier.py

@@ -0,0 +1,72 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+# @Author: Daijingbo
+# @Date  : 2019/6/16
+# @Desc  :FBP ML XGBClassifier
+# http://www.captainbed.net/blog-acredjb
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import xgboost as xgb
+from xgboost import plot_importance
+from xgboost import XGBClassifier
+# from sklearn import preprocessing
+# from sklearn.preprocessing import Imputer
+from sklearn.feature_extraction import DictVectorizer
+# from sklearn.cross_validation import train_test_split
+from sklearn.model_selection import KFold
+from sklearn.model_selection import train_test_split
+def trainandTest(X, y,X_t):
+    X_train, X_test, y_train, y_test=train_test_split(X, y, test_size=0.25, random_state=33)
+    ### feature_extraction
+    vec=DictVectorizer(sparse=False)
+    X_train=vec.fit_transform(X_train.to_dict(orient='record'))
+    X_test=vec.transform(X_t.to_dict(orient='record'))
+############第三处调参：选择全参数和无参数（默认）################################
+    # model=xgb.XGBClassifier(learning_rate =0.1,n_estimators=1000,max_depth=4,min_child_weight=6,gamma=0,subsample=0.8,colsample_bytree=0.8,reg_alpha=0.005,objective='binary:logistic',nthread=4,scale_pos_weight=1,seed=27)
+    model=XGBClassifier()#ok无参数
+
+    model.fit(X_train,y_train)
+    # 对测试集进行预测
+    ans = model.predict(X_test)
+    ans_len = len(ans)
+    id_list = np.arange(5953, 5975)
+    data_arr = []
+    for row in range(0, ans_len):
+        data_arr.append([int(id_list[row]), ans[row]])
+        print(ans[row])
+    np_data = np.array(data_arr)
+    # 写入文件
+    pd_data = pd.DataFrame(np_data, columns=['id', 'y'])
+    pd_data.to_csv('FBP_submit.csv', index=None)
+    # # 显示重要特征
+    # plot_importance(model)
+    # plt.show()
+
+if __name__ == '__main__':
+
+    f0='ysb';  # bet365
+    f1='li';  # 8jbb
+    f2='bet365';  # wl
+    f3='hg';  # 10ysb
+    f4='wl';
+    f5='ms';
+    f6='ao';  # li
+    f7='interw';
+    f8='w';
+    f9='10bet';  # interw
+    f10='SNAI';  # 9ms
+
+################第一处调参：选择训练集数据的行数1000-4000-all##################
+    trainFilePath = 'E:/PythonLearn/pc_ex/AdaBoost_PeiLv/FBP_train.csv'
+    # trainFilePath='E:/PythonLearn/pc_ex/AdaBoost_PeiLv/FBP_train-3000.csv'
+    # trainFilePath = 'E:/PythonLearn/pc_ex/AdaBoost_PeiLv/FBP_train-all.csv'
+    testFilePath = 'E:/PythonLearn/pc_ex/AdaBoost_PeiLv/FBP_predict.csv'
+    data = pd.read_csv(trainFilePath)
+    X_test= pd.read_csv(testFilePath)
+###############第二处调参：选择全部特征还是部分特征###########################
+    X_train=data[[f0,f1,f2,f3,f4,f5,f6,f7,f8,f9,f10]]#全特征
+    # X_train=data[[f10, f7, f5,f6]]
+
+    y_train=data['y']
+    trainandTest(X_train, y_train,X_test)