下面将介绍XGBoost的Python模块,内容如下:
* 编译及导入Python模块
* 数据接口
* 参数设置
* 训练模型l
* 提前终止程序
* 预测
A walk through python example for UCI Mushroom dataset is provided.
首先安装XGBoost的C++版本,然后进入源文件的根目录下的 wrappers
文件夹执行如下脚本安装Python模块
python setup.py install
安装完成后按照如下方式导入XGBoost的Python模块
import xgboost as xgb
=
XGBoost可以加载libsvm格式的文本数据,加载的数据格式可以为Numpy的二维数组和XGBoost的二进制的缓存文件。加载的数据存储在对象DMatrix
中。
dtrain = xgb.DMatrix('train.svm.txt')
dtest = xgb.DMatrix('test.svm.buffer')
DMatrix
对象时,可以用如下方式data = np.random.rand(5,10) # 5 entities, each contains 10 features
label = np.random.randint(2, size=5) # binary target
dtrain = xgb.DMatrix( data, label=label)
scipy.sparse
格式的数据转化为 DMatrix
格式时,可以使用如下方式csr = scipy.sparse.csr_matrix( (dat, (row,col)) )
dtrain = xgb.DMatrix( csr )
DMatrix
格式的数据保存成XGBoost的二进制格式,在下次加载时可以提高加载速度,使用方式如下dtrain = xgb.DMatrix('train.svm.txt')
dtrain.save_binary("train.buffer")
DMatrix
中的缺失值:dtrain = xgb.DMatrix( data, label=label, missing = -999.0)
w = np.random.rand(5,1)
dtrain = xgb.DMatrix( data, label=label, missing = -999.0, weight=w)
XGBoost使用key-value格式保存参数. Eg
* Booster(基本学习器)参数
param = {'bst:max_depth':2, 'bst:eta':1, 'silent':1, 'objective':'binary:logistic' }
param['nthread'] = 4
plst = param.items()
plst += [('eval_metric', 'auc')] # Multiple evals can be handled in this way
plst += [('eval_metric', 'ams@0')]
evallist = [(dtest,'eval'), (dtrain,'train')]
=
有了参数列表和数据就可以训练模型了
* 训练
num_round = 10
bst = xgb.train( plst, dtrain, num_round, evallist )
bst.save_model('0001.model')
# dump model
bst.dump_model('dump.raw.txt')
# dump model with feature map
bst.dump_model('dump.raw.txt','featmap.txt')
bst = xgb.Booster({'nthread':4}) #init model
bst.load_model("model.bin") # load data
=
如果有评价数据,可以提前终止程序,这样可以找到最优的迭代次数。如果要提前终止程序必须至少有一个评价数据在参数evals
中。 If there’s more than one, it will use the last.
train(..., evals=evals, early_stopping_rounds=10)
The model will train until the validation score stops improving. Validation error needs to decrease at least every early_stopping_rounds
to continue training.
If early stopping occurs, the model will have two additional fields: bst.best_score
and bst.best_iteration
. Note that train()
will return a model from the last iteration, not the best one.
This works with both metrics to minimize (RMSE, log loss, etc.) and to maximize (MAP, NDCG, AUC).
=
Prediction
After you training/loading a model and preparing the data, you can start to do prediction.
data = np.random.rand(7,10) # 7 entities, each contains 10 features
dtest = xgb.DMatrix( data, missing = -999.0 )
ypred = bst.predict( xgmat )
If early stopping is enabled during training, you can predict with the best iteration.
ypred = bst.predict(xgmat,ntree_limit=bst.best_iteration)