python包xgboost安装和简单使用

1、xgboost安装

首先，xgboost是python的一个包，用于数据分析以及boost的实现。

cp36代表python3.6。

numpy-1.13.1+mkl-cp36-cp36m-win_amd64.whl scipy-0.19.1-cp36-cp36m-win_amd64.whl xgboost-0.6-cp36-cp36m-win_amd64.whl 

在这里-xgboost下载，然后在cmd，在下载文件的路径，运行：

F:\下载软件>pip3 install xgboost-0.6+20171121-cp36-cp36m-win_amd64.whl

注：一开始使用pip安装的时候，出错 xgboost-0.6+20171121-cp36-cp36m-win_amd64.whl is not a supported wheel on this platform.

一开始按照知乎专栏的文章安装了xgboost，但是加载不了模型，然后，再按照简书那篇文章，重新安装了xgboost。感觉，应该随意按照其中一篇文章，都可以安装成功。

2、xgboost调试

xgboost可以分成两种实现：1、原生接口；2、scikitlearn实现；xgboost处理的问题分成两种：1、分类问题；2、回归问题

2.1 Demo1 基于XGBoost原生接口的分类

#xgboost#分类from sklearn.datasets import load_irisimport xgboost as xgbfrom xgboost import plot_importancefrom matplotlib import pyplot as pltfrom sklearn.model_selection import train_test_split# read in the iris datairis = load_iris()X = iris.datay = iris.targetX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1234565)params = {    'booster': 'gbtree',    'objective': 'multi:softmax',    'num_class': 3,    'gamma': 0.1,    'max_depth': 6,    'lambda': 2,    'subsample': 0.7,    'colsample_bytree': 0.7,    'min_child_weight': 3,    'silent': 1,    'eta': 0.1,    'seed': 1000,    'nthread': 4,}plst = params.items()dtrain = xgb.DMatrix(X_train, y_train)num_rounds = 500model = xgb.train(plst, dtrain, num_rounds)# 对测试集进行预测dtest = xgb.DMatrix(X_test)ans = model.predict(dtest)# 计算准确率cnt1 = 0cnt2 = 0for i in range(len(y_test)):    if ans[i] == y_test[i]:        cnt1 += 1    else:        cnt2 += 1print("Accuracy: %.2f %% " % (100 * cnt1 / (cnt1 + cnt2)))# 显示重要特征plot_importance(model)plt.show()

运行结果：

2.2 Demo2-基于Scikit-learn接口的分类

#5.3 基于Scikit-learn接口的分类from sklearn.datasets import load_irisimport xgboost as xgbfrom xgboost import plot_importancefrom matplotlib import pyplot as pltfrom sklearn.model_selection import train_test_split# read in the iris datairis = load_iris()X = iris.datay = iris.targetX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)# 训练模型model = xgb.XGBClassifier(max_depth=5, learning_rate=0.1, n_estimators=160, silent=True, objective='multi:softmax')model.fit(X_train, y_train)# 对测试集进行预测ans = model.predict(X_test)# 计算准确率cnt1 = 0cnt2 = 0for i in range(len(y_test)):    if ans[i] == y_test[i]:        cnt1 += 1    else:        cnt2 += 1print("Accuracy: %.2f %% " % (100 * cnt1 / (cnt1 + cnt2)))# 显示重要特征plot_importance(model)plt.show()

以下两个demo是回归问题的实现：

2.3 demo3-基于Scikit-learn接口的回归

#5.4 基于Scikit-learn接口的回归import xgboost as xgbfrom xgboost import plot_importancefrom matplotlib import pyplot as pltfrom sklearn.model_selection import train_test_splitimport csv# 读取文件原始数据#读取数据方法1# data = []# labels = []# labels2 = []# with open(r"G:\0研究生\tianchiCompetition\训练小样本3_label.csv", encoding='UTF-8') as fileObject:    # for line in fileObject:        # line_split = line.split(',')        # data.append(line_split[10:])        # labels.append(line_split[8])# X = []# for row in data:    # row = [float(x) for x in row]    # X.append(row)# y = [float(x) for x in labels]#读取数据方法2#####X=[]y=[]            #定义空列表csvFile = open(r"G:\0研究生\tianchiCompetition\训练小样本3_label.csv", "r")reader = csv.reader(csvFile)for item in reader:    # if reader.line_num==1:        # continue    item=[float(ii) for ii in item]    X.append(item)###把读取的数据转化成float格式for i in range(len(X)):    y.append(X[i].pop())# print('x',X)# print('Y',Y)# XGBoost训练过程X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)model = xgb.XGBRegressor(max_depth=5, learning_rate=0.1, n_estimators=160, silent=True, objective='reg:gamma')model.fit(X_train, y_train)# 对测试集进行预测ans = model.predict(X_test)# 显示重要特征plot_importance(model)plt.show() 

2.4基于XGBoost原生接口的回归

# 5.2 基于XGBoost原生接口的回归import xgboost as xgbfrom xgboost import plot_importancefrom matplotlib import pyplot as pltfrom sklearn.model_selection import train_test_splitimport csvfrom xgboost import XGBClassifier# # 读取文件原始数据# data = []# labels = []# labels2 = []# with open("lppz5.csv", encoding='UTF-8') as fileObject:    # for line in fileObject:        # line_split = line.split(',')        # data.append(line_split[10:])        # labels.append(line_split[8])# X = []# for row in data:    # row = [float(x) for x in row]    # X.append(row)# y = [float(x) for x in labels]#读取数据方法2#####X=[]y=[]            #定义空列表csvFile = open(r"G:\0研究生\tianchiCompetition\训练小样本4large_label.csv", "r")reader = csv.reader(csvFile)for item in reader:    # if reader.line_num==1:        # continue    item=[float(ii) for ii in item]    X.append(item)###把读取的数据转化成float格式for i in range(len(X)):    y.append(X[i].pop())# XGBoost训练过程X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)params = {    'booster': 'gbtree',    'objective': 'reg:gamma',    'gamma': 0.1,    'max_depth': 5,    'lambda': 3,    'subsample': 0.7,    'colsample_bytree': 0.7,    'min_child_weight': 3,    'silent': 1,    'eta': 0.1,    'seed': 1000,    'nthread': 4,}dtrain = xgb.DMatrix(X_train, y_train)# num_rounds = 300num_rounds = 126plst = params.items()model = xgb.train(plst, dtrain, num_rounds)# 对测试集进行预测dtest = xgb.DMatrix(X_test)ans = model.predict(dtest)# 显示重要特征plot_importance(model)plt.show()

运行程序出错：

raise ValueError('Booster.get_score() results in empty')ValueError: Booster.get_score() results in empty

现在，还不是很懂这是什么问题。

把读取的代码改了，

for i in range(len(X)):    y.append(X[i][-1])

就是X保留label值，运行没有问题，但是，运行结果却是：

这很奇怪，为什么只有最后一列（label）的feature，这样对我没有什么意义。

---

参考：

1. xgboost: 速度快效果好的 boosting 模型；
2. xgboost基本入门_知乎专栏；
3. xgboost安装指南；
4. xgboost在Windows安装_简书