评估xgboost

1.划分训练测试集，看看测试集上好坏

之前就有这里就不赘述了

2.k-fold

# k-fold cross validation evaluation of xgboost model
from numpy import loadtxt
import xgboost
from sklearn.model_selection import KFold
from sklearn.model_selection import cross_val_score
# load data
dataset = loadtxt('pima-indians-diabetes.csv', delimiter=",")
# split data into X and y
X = dataset[:,0:8]
Y = dataset[:,8]
# CV model
model = xgboost.XGBClassifier()
kfold = KFold(n_splits=10, random_state=7)
results = cross_val_score(model, X, Y, cv=kfold)
print("Accuracy: %.2f%% (%.2f%%)" % (results.mean()*100, results.std()*100))

但是呢，如果我们数据集不均衡，可以采用分层k-fold，保证每个fold的分布和data分布一样

这个时候用

StratifiedKFold

# stratified k-fold cross validation evaluation of xgboost model
from numpy import loadtxt
import xgboost
from sklearn.model_selection import StratifiedKFold
from sklearn.model_selection import cross_val_score
# load data
dataset = loadtxt('pima-indians-diabetes.csv', delimiter=",")
# split data into X and y
X = dataset[:,0:8]
Y = dataset[:,8]
# CV model
model = xgboost.XGBClassifier()
kfold = StratifiedKFold(n_splits=10, random_state=7)
results = cross_val_score(model, X, Y, cv=kfold)
print("Accuracy: %.2f%% (%.2f%%)" % (results.mean()*100, results.std()*100))

https://machinelearningmastery.com/evaluate-gradient-boosting-models-xgboost-python/