LGB XGB CBT参数

LGB：

1、lgb.Dataset()

train_data = lgb.Dataset(data, label=label, feature_name=['c1', 'c2', 'c3'], categorical_feature=['c3'], weight=w )LightGBM 可以直接使用 categorical features（分类特征）作为 input（输入）. 它不需要被转换成 one-hot coding（独热编码）, 并且它比 one-hot coding（独热编码）更快（约快上 8 倍）注意: 在你构造 Dataset 之前, 你应该将分类特征转换为 int 类型的值.需要时可以设置权重weight，也可用函数train_data.set_weight(w)

2、交叉验证cv

num_round = 10

lgb.cv(param, train_data, num_round, nfold=5)

3、

#创建成lgb特征的数据集格式

lgb_train = lgb.Dataset(X_train, y_train) # 将数据保存到LightGBM二进制文件将使加载更快

lgb_eval = lgb.Dataset(X_test, y_test, reference=lgb_train) # 创建验证数据

#将参数写成字典下形式

params = {

'task': 'train','boosting_type': 'gbdt', # 设置提升类型'objective': 'regression', # 目标函数'metric': {'l2', 'auc'}, # 评估函数'num_leaves': 31, # 叶子节点数'learning_rate': 0.05, # 学习速率'feature_fraction': 0.9, # 建树的特征选择比例'bagging_fraction': 0.8, # 建树的样本采样比例'bagging_freq': 5, # k 意味着每 k 次迭代执行bagging'verbose': 1 # <0 显示致命的, =0 显示错误 (警告), >0 显示信息

}

print(‘Start training…’)

#训练 cv and train

gbm = lgb.train(params,lgb_train,num_boost_round=20,valid_sets=lgb_eval,early_stopping_rounds=5) # 训练数据需要参数列表和数据集

print(‘Save model…’)

gbm.save_model(‘model.txt’) # 训练后保存模型到文件

print(‘Start predicting…’)

#预测数据集

y_pred = gbm.predict(X_test, num_iteration=gbm.best_iteration) #如果在训练期间启用了早期停止，可以通过best_iteration方式从最佳迭代中获得预测

#评估模型

print(‘The rmse of prediction is:’, mean_squared_error(y_test, y_pred) ** 0.5) # 计算真实值和预测值之间的均方根误差

#读模型数据import pandas as pddf = pd.read('data.csv')#剔除模型无关变量和完全线性相关的变量#写入Train_X,test_X,train_y,test_yfrom sklearn.cross_validation import train_test_splittrain_X, test_X, train_y, test_y = train_test_split(X, Y, train_size = 0.8, random_state = 123)import lightgbm as lgblgb_train = lgb.Dataset(train_X, train_y, free_raw_data=False)lgb_eval = lgb.Dataset(test_X, test_y, reference=lgb_train,free_raw_data=False)param = {'task': 'train','boosting_type': 'gbdt','objective': 'binary','metric': {'l2', 'auc'},'num_leaves': 40,'learning_rate': 0.01,'feature_fraction': 0.8,'bagging_fraction': 0.8,'bagging_freq': 5,'verbose': 0}param['is_unbalance']='true'param['metric'] = 'auc'bst=lgb.cv(param,lgb_train, num_boost_round=1000, nfold=6, early_stopping_rounds=100)estimators = lgb.train(param,lgb_train,num_boost_round=len(bst['auc-mean']))print('Start training...')y_pred = estimators.predict(test_X, num_iteration=estimators.best_iteration)from sklearn import metricsprint('The roc of prediction is:', metrics.roc_auc_score(test_y, y_pred) )print('Feature names:', estimators.feature_name())print('Feature importances:', list(estimators.feature_importance()))

XGB

def XGB_predict(train_x,train_y,val_X,val_Y,test_x,res):print("XGB test")# create dataset for lightgbmxgb_val = xgb.DMatrix(val_X, label=val_Y)xgb_train = xgb.DMatrix(X_train, label=y_train)xgb_test = xgb.DMatrix(test_x)# specify your configurations as a dictparams = {'booster': 'gbtree',# 'objective': 'multi:softmax', # 多分类的问题、# 'objective': 'multi:softprob', # 多分类概率'objective': 'binary:logistic','eval_metric': 'auc',# 'num_class': 9, # 类别数，与 multisoftmax 并用'gamma': 0.1, # 用于控制是否后剪枝的参数,越大越保守，一般0.1、0.2这样子。'max_depth': 8, # 构建树的深度，越大越容易过拟合'alpha': 0, # L1正则化系数'lambda': 10, # 控制模型复杂度的权重值的L2正则化项参数，参数越大，模型越不容易过拟合。'subsample': 0.7, # 随机采样训练样本'colsample_bytree': 0.5, # 生成树时进行的列采样'min_child_weight': 3,# 这个参数默认是 1，是每个叶子里面 h 的和至少是多少，对正负样本不均衡时的 0-1 分类而言# ，假设 h 在 0.01 附近，min_child_weight 为 1 意味着叶子节点中最少需要包含 100 个样本。# 这个参数非常影响结果，控制叶子节点中二阶导的和的最小值，该参数值越小，越容易 overfitting。'silent': 0, # 设置成1则没有运行信息输出，最好是设置为0.'eta': 0.03, # 如同学习率'seed': 1000,'nthread': -1, # cpu 线程数'missing': 1,'scale_pos_weight': (np.sum(y==0)/np.sum(y==1)) # 用来处理正负样本不均衡的问题,通常取：sum(negative cases) / sum(positive cases)# 'eval_metric': 'auc'}plst = list(params.items())num_rounds = 5000 # 迭代次数watchlist = [(xgb_train, 'train'), (xgb_val, 'val')]# 交叉验证# result = xgb.cv(plst, xgb_train, num_boost_round=200, nfold=4, early_stopping_rounds=200, verbose_eval=True, folds=StratifiedKFold(n_splits=4).split(X, y))# 训练模型并保存# early_stopping_rounds 当设置的迭代次数较大时，early_stopping_rounds 可在一定的迭代次数内准确率没有提升就停止训练model = xgb.train(plst, xgb_train, num_rounds, watchlist, early_stopping_rounds=200)res['score'] = model.predict(xgb_test)res['score'] = res['score'].apply(lambda x: float('%.6f' % x))return res

CBT

如果觉得《LGB XGB CBT参数》对你有帮助，请点赞、收藏，并留下你的观点哦！