集成学习01_xgboost参数讲解与实战

本章分以下几块来讲解

一.xgboost 模型参数介绍

二.xgboost 两种方式实现

三. 网格搜索最优xgboost参数

一.XGBoost的参数

XGBoost的作者把所有的参数分成了三类,这里只介绍我们常用的一些参数,不常用的不做介绍

通用参数:宏观函数控制。
Booster参数:控制每一步的booster(tree/regression)。
学习目标参数:控制训练目标的表现。

1 通用参数

1)booster[默认gbtree]

  • 选择每次迭代的模型,有两种选择:
    gbtree:基于树的模型
    gbliner:线性模型

2)silent[默认0]

  • 当这个参数值为1时,静默模式开启,不会输出任何信息。
  • 一般这个参数就保持默认的0,因为这样能帮我们更好地理解模型。

3)nthread[默认值为最大可能的线程数]

  • 这个参数用来进行多线程控制,应当输入系统的核数。
  • 如果你希望使用CPU全部的核,那就不要输入这个参数,算法会自动检测它。

4)num_feature [set automatically by xgboost, no need to be set by user]

  • boosting过程中用到的特征维数,设置为特征个数。
  • XGBoost会自动设置,不需要手工设置。

2 booster参数

尽管有两种booster可供选择,我这里只介绍tree booster,因为它的表现远远胜过linear booster,所以linear booster很少用到。

1)eta[默认0.3]

  • 和GBM中的 learning rate 参数类似。
  • 通过减少每一步的权重,可以提高模型的鲁棒性。
  • 典型值为0.01-0.2。

2)min_child_weight[默认1]

*决定最小叶子节点样本权重和。

  • 和GBM的 min_child_leaf 参数类似,但不完全一样。XGBoost的这个参数是最小样本权重的和,而GBM参数是最小样本总数。
  • 这个参数用于避免过拟合。当它的值较大时,可以避免模型学习到局部的特殊样本。
  • 但是如果这个值过高,会导致欠拟合。这个参数需要使用CV来调整。

3)max_depth[默认6]

  • 和GBM中的参数相同,这个值为树的最大深度。
  • 这个值也是用来避免过拟合的。max_depth越大,模型会学到更具体更局部的样本。
  • 需要使用CV函数来进行调优。
  • 典型值:3-10

4)max_leaf_nodes

  • 树上最大的节点或叶子的数量。
  • 可以替代max_depth的作用。因为如果生成的是二叉树,一个深度为n的树最多生成
  • 如果定义了这个参数,GBM会忽略max_depth参数。

5)gamma[默认0]

  • 在节点分裂时,只有分裂后损失函数的值下降了,才会分裂这个节点。
  • Gamma指定了节点分裂所需的最小损失函数下降值。这个参数的值越大,算法越保守。这个参数的值和损失函数息息相关,所以是需要调整的。
  • 模型在默认情况下,对于一个节点的划分只有在其loss function 得到结果大于0的情况下才进行,而gamma 给定了所需的最低loss function的值
  • gamma值使得算法更conservation,且其值依赖于loss function ,在模型中应该进行调参.

6)max_delta_step[默认0]

  • 这参数限制每棵树权重改变的最大步长。如果这个参数的值为0,那就意味着没有约束。如果它被赋予了某个正值,那么它会让这个算法更加保守。
  • 通常,这个参数不需要设置。但是当各类别的样本十分不平衡时,它对逻辑回归是很有帮助的。
  • 这个参数一般用不到,但是你可以挖掘出来它更多的用处。

7)subsample[默认1]

  • 和GBM中的subsample参数一模一样。这个参数控制对于每棵树,随机采样的比例。
  • 减小这个参数的值,算法会更加保守,避免过拟合。但是,如果这个值设置得过小,它可能会导致欠拟合。
  • 典型值:0.5-1

8)colsample_bytree[默认1]

  • 和GBM里面的max_features参数类似。用来控制每棵随机采样的列数的占比(每一列是一个特征)。
  • 典型值:0.5-1

9)colsample_bylevel[默认1]

  • 用来控制树的每一级的每一次分裂,对列数的采样的占比。
  • 一般不太用这个参数,因为subsample参数和colsample_bytree参数可以起到相同的作用。但是如果感兴趣,可以挖掘这个参数更多的用处。

10)lambda[默认1]

  • 权重的L2正则化项。(和Ridge regression类似)。
  • 这个参数是用来控制XGBoost的正则化部分的。

11)alpha[默认1]

  • 权重的L1正则化项。(和Lasso regression类似)。
  • 可以应用在很高维度的情况下,使得算法的速度更快。

12)scale_pos_weight[默认1]

  • 在各类别样本十分不平衡时,把这个参数设定为一个正值,可以使算法更快收敛。
  • 大于0的取值可以处理类别不平衡的情况。帮助模型更快收敛。

13) Parameter for Linear Booster

lambda_bias

  • 在偏置上的L2正则。缺省值为0(在L1上没有偏置项的正则,因为L1时偏置不重要)

3 学习目标参数

这个参数用来控制理想的优化目标和每一步结果的度量方法。

1)objective[默认reg:linear]

  • 这个参数定义需要被最小化的损失函数。最常用的值有: 定义学习任务及相应的学习目标,可选的目标函数如下:
  • “reg:linear” –线性回归。
  • “reg:logistic” –逻辑回归。
  • “binary:logistic” –二分类的逻辑回归问题,输出为概率。
  • “binary:logitraw” –二分类的逻辑回归问题,输出的结果为wTx。
  • “count:poisson” –计数问题的poisson回归,输出结果为poisson分布。
  • 在poisson回归中,max_delta_step的缺省值为0.7。(used to safeguard optimization)
  • “multi:softmax” –让XGBoost采用softmax目标函数处理多分类问题,同时需要设置参数num_class(类别个数)
  • “multi:softprob” –和softmax一样,但是输出的是ndata * nclass的向量,可以将该向量reshape成ndata行nclass列的矩阵。每行数据表示样本所属于每个类别的概率。
  • “rank:pairwise” –set XGBoost to do ranking task by minimizing the pairwise loss

2)eval_metric[默认值取决于objective参数的取值]

  • 对于有效数据的度量方法。
  • 对于回归问题,默认值是rmse,对于分类问题,默认值是error。
  • 典型值有:
  • rmse 均方根误差
  • mae 平均绝对误差
  • logloss 负对数似然函数值
  • error 二分类错误率(阈值为0.5)
  • merror 多分类错误率
  • mlogloss 多分类logloss损失函数
  • auc 曲线下面积

3)seed(默认0)

  • 随机数的种子
  • 设置它可以复现随机数据的结果,也可以用于调整参数
  • 如果你比较习惯scikit-learn的参数形式,那么XGBoost的Python 版本也提供了sklearn形式的接口 XGBClassifier。

4)sklearn 参数对照

它使用sklearn形式的参数命名方式,对应关系如下:

  • 1、eta -> learning_rate
  • 2、lambda -> reg_lambda
  • 3、alpha -> reg_alpha

4.平台控制参数 Console Parameters

The following parameters are only used in the console version of xgboost

  • use_buffer [ default=1 ]
    是否为输入创建二进制的缓存文件,缓存文件可以加速计算。缺省值为1
  • num_round
    boosting迭代计算次数。
  • data
    输入数据的路径
  • test:data
    测试数据的路径
  • save_period [default=0]
    表示保存第i*save_period次迭代的模型。例如save_period=10表示每隔10迭代计算XGBoost将会保存中间结果,设置为0表示每次计算的模型都要保持。
  • task [default=train] options: train, pred, eval, dump
    train:训练模型
    pred:对测试数据进行预测
    eval:通过eval[name]=filenam定义评价指标
    dump:将学习模型保存成文本格式
  • model_in [default=NULL]
    指向模型的路径在test, eval, dump都会用到,如果在training中定义XGBoost将会接着输入模型继续训练
  • model_out [default=NULL]
    训练完成后模型的保存路径,如果没有定义则会输出类似0003.model这样的结果,0003是第三次训练的模型结果。
  • model_dir [default=models]
    输出模型所保存的路径。
  • fmap
    feature map, used for dump model
  • name_dump [default=dump.txt]
    name of model dump file
  • name_pred [default=pred.txt]
    预测结果文件
  • pred_margin [default=0]
    输出预测的边界,而不是转换后的概率

二.xgboost 实现

本章以优惠券推荐数据为例对xgboost结合skleran与直接采用xgboost进行实现

1.导入相关包


import pandas as pd, numpy as np
from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn import metrics
import catboost as cb
import xgboost as xgb
from xgboost.sklearn import XGBClassifier
import os
import joblib
from sklearn.preprocessing import LabelEncoder
from collections import defaultdict
data=pd.read_excel('car_coupon.xlsx')
data.head(5)
ID destination passanger toCoupon_GEQ15min toCoupon_GEQ25min direction_same direction_opp gender age maritalStatus ... Bar CoffeeHouse CarryAway RestaurantLessThan20 Restaurant20To50 weather time coupon expiration Y
0 11263 No Urgent Place Friend(s) 0 0 0 1 Male 55 Widowed ... 0 0 1 1 1 Sunny 14 Coffee House 24 1
1 20136 Work Alone 1 0 1 0 Female 26 Married partner ... 0 0 3 3 3 Sunny 7 Bar 24 0
2 14763 Work Alone 1 0 0 1 Female 55 Single ... 0 0 1 1 1 Sunny 7 Coffee House 24 0
3 12612 No Urgent Place Kid(s) 1 0 0 1 Female 41 Married partner ... 0 3 3 3 3 Sunny 10 Carry out & Take away 2 0
4 17850 No Urgent Place Partner 1 0 0 1 Female 31 Married partner ... 1 1 10 10 10 Snowy 14 Coffee House 2 0

5 rows × 23 columns

2.数据处理

    1. 对类别数据进行编码
d = defaultdict(LabelEncoder)
data[['destination', 'passanger','gender','maritalStatus','education', 'occupation', \
    'weather','coupon' ]]=data[['destination', 'passanger','gender','maritalStatus','education', 'occupation', \
        'weather','coupon' ]].apply(lambda x: d[x.name].fit_transform(x))
data.head(5)
ID destination passanger toCoupon_GEQ15min toCoupon_GEQ25min direction_same direction_opp gender age maritalStatus ... Bar CoffeeHouse CarryAway RestaurantLessThan20 Restaurant20To50 weather time coupon expiration Y
0 11263 1 1 0 0 0 1 1 55 4 ... 0 0 1 1 1 2 14 2 24 1
1 20136 2 0 1 0 1 0 0 26 1 ... 0 0 3 3 3 2 7 0 24 0
2 14763 2 0 1 0 0 1 0 55 2 ... 0 0 1 1 1 2 7 2 24 0
3 12612 1 2 1 0 0 1 0 41 1 ... 0 3 3 3 3 2 10 1 2 0
4 17850 1 3 1 0 0 1 0 31 1 ... 1 1 10 10 10 1 14 2 2 0

5 rows × 23 columns

  • 切分训练集与测试集
train, test, y_train, y_test = train_test_split(data.drop(["Y"], axis=1), data["Y"],
                                                random_state=10, test_size=0.3)  
  • 注意下 data ,train, test, y_train, y_test的数据格式
print(type(data))
print(type(train))
print(type( test))
print(type(y_train))
print(type(y_test))





  • 撰写评价函数
def model_eval2(m, train, test):
    print('train_roc_auc_score:',metrics.roc_auc_score(y_train, m.predict_proba(train)[:, 1]))
    print('test_roc_auc_score:',metrics.roc_auc_score(y_test, m.predict_proba(test)[:, 1]))
    print('train_accuracy_score:',metrics.accuracy_score(y_train,  m.predict(train)))
    print('test_accuracy_score:',metrics.accuracy_score(y_test, m.predict(test)))
    print('train_precision_score:',metrics.precision_score(y_train, m.predict(train)))
    print('test__precision_score:',metrics.precision_score(y_test, m.predict(test)))
    print('train_recall_score:',metrics.recall_score(y_train, m.predict(train)))
    print('test_recall_score:',metrics.recall_score(y_test, m.predict(test)))
    print('train_f1_score:',metrics.f1_score(y_train, m.predict(train)))
    print('test_f1_score:',metrics.f1_score(y_test, m.predict(test)))  

3.结合sklearn的xgboot模型

step01-拟合模型

from xgboost.sklearn import XGBClassifier
xgboost_model = XGBClassifier()
eval_set = [(test.values,  y_test.values)]
#拟合模型
xgboost_model.fit(train.values, 
                  y_train.values, 
                  early_stopping_rounds=300, 
                  eval_metric="logloss",  # 损失函数的类型,分类一般都是用对数作为损失函数
                  eval_set=eval_set,
                  verbose=False)
D:\dprograme\Anaconda3\lib\site-packages\xgboost\sklearn.py:793: UserWarning: `eval_metric` in `fit` method is deprecated for better compatibility with scikit-learn, use `eval_metric` in constructor or`set_params` instead.
  warnings.warn(
D:\dprograme\Anaconda3\lib\site-packages\xgboost\sklearn.py:793: UserWarning: `early_stopping_rounds` in `fit` method is deprecated for better compatibility with scikit-learn, use `early_stopping_rounds` in constructor or`set_params` instead.
  warnings.warn(
XGBClassifier(base_score=0.5, booster='gbtree', callbacks=None,
          colsample_bylevel=1, colsample_bynode=1, colsample_bytree=1,
          early_stopping_rounds=None, enable_categorical=False,
          eval_metric=None, gamma=0, gpu_id=-1, grow_policy='depthwise',
          importance_type=None, interaction_constraints='',
          learning_rate=0.300000012, max_bin=256, max_cat_to_onehot=4,
          max_delta_step=0, max_depth=6, max_leaves=0, min_child_weight=1,
          missing=nan, monotone_constraints='()', n_estimators=100,
          n_jobs=0, num_parallel_tree=1, predictor='auto', random_state=0,
          reg_alpha=0, reg_lambda=1, ...)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

step02-评价模型

model_eval2(xgboost_model, train.values, test.values)
train_roc_auc_score: 0.890295988831706
test_roc_auc_score: 0.7178983466569767
train_accuracy_score: 0.8007142857142857
test_accuracy_score: 0.6683333333333333
train_precision_score: 0.7965116279069767
test__precision_score: 0.704225352112676
train_recall_score: 0.8681875792141952
test_recall_score: 0.7267441860465116
train_f1_score: 0.8308065494238933
test_f1_score: 0.7153075822603719

step03-利用模型预测

  • xgboost_model.predict 预测结果是0或1的int型
  • xgboost_model.predict_proba预测结果是0到1之间的float型
y_test_pred = xgboost_model.predict( test.values )
y_trian_prod = xgboost_model.predict_proba( train.values )

step04-保存和调用模型

joblib.dump(xgboost_model , r'D:\Ensemble_Learning\xgboostinfo\xgboostsklearnsingle.model')
load_model=joblib.load(r'D:\Ensemble_Learning\xgboostinfo\xgboostsklearnsingle.model')
load_model.predict( test.values )
array([0, 0, 1, 0, 1, 0, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 1, 1, 0, 1,
       1, 1, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 1, 0, 1,
       1, 0, 1, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 0, 1, 1, 1, 1, 0, 1, 1, 0,
       1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 1, 1, 0, 1, 0, 0, 1, 1, 0,
       1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1,
       0, 0, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1,
       0, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 0, 0,
       1, 1, 1, 0, 0, 1, 1, 0, 1, 0, 1, 1, 1, 0, 0, 0, 1, 1, 0, 1, 0, 1,
       0, 0, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 1, 0, 1, 0,
       1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 0, 1, 1, 0, 1, 0, 1, 1, 0,
       1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 1, 0, 1, 0, 1, 1, 1, 0, 1, 1, 0, 0,
       1, 0, 1, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 0, 1, 1, 0, 0, 0, 0, 1,
       1, 0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0,
       1, 0, 0, 1, 0, 1, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 1, 1, 0, 0, 1,
       1, 0, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0,
       1, 0, 1, 1, 0, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1,
       0, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 1, 1, 0, 0, 1, 0,
       1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 1, 1, 0, 0, 0, 1, 0, 1, 1, 0, 0, 1,
       1, 1, 0, 1, 1, 1, 0, 1, 1, 0, 0, 1, 0, 1, 1, 1, 1, 1, 1, 0, 1, 1,
       1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 1, 0,
       0, 0, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1,
       1, 0, 1, 0, 0, 1, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 0, 0, 0, 0,
       1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0, 0, 0, 1,
       0, 0, 0, 1, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1,
       1, 0, 1, 0, 0, 0, 0, 1, 1, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1,
       1, 0, 1, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 1, 0, 0, 1, 1, 1, 1, 0,
       1, 1, 0, 1, 0, 0, 1, 0, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1,
       0, 1, 1, 0, 1, 0])

注意点

  • 上面xgboost_model.fit传入的是train.values和y_train.values,数据类型为numpy.ndarray
  • 上面* xgboost_model.predict与xgboost_model.predict_proba传入的数据类型为numpy.ndarray
print(type(data.values))
print(type(train.values))
print(type( test.values))
print(type(y_train.values))
print(type(y_test.values))





4.直接采用xgboost的模型

step-01 构建参数

params={'alpha': 0.09,
 'booster': 'gbtree',
 'colsample_bylevel': 0.4,
 'colsample_bytree': 0.7,
 'eval_metric': 'logloss',
 'gamma': 0.85,
 'learning_rate': 0.1,
 'max_depth': 7,
 'min_child_weight': 20,
 'n_estimator': 40,
 'objective': 'binary:logistic',
 'reg_lambda': 0.1,
 'seed': 1,
 'subsample': 0.6}

step-02 处理数据

dtrain = xgb.DMatrix(train, label=y_train,feature_names=list(train.columns))
dtest = xgb.DMatrix(test)
validation = xgb.DMatrix(test,y_test)
watchlist = [(validation,'train')]

step-03 拟合模型

model = xgb.train(params,
                  dtrain,
                  num_boost_round= 2000, # 迭代的次数,及弱学习器的个数
                  evals= watchlist)
[21:06:30] WARNING: C:/Users/administrator/workspace/xgboost-win64_release_1.6.0/src/learner.cc:627: 
Parameters: { "n_estimator" } might not be used.

  This could be a false alarm, with some parameters getting used by language bindings but
  then being mistakenly passed down to XGBoost core, or some parameter actually being used
  but getting flagged wrongly here. Please open an issue if you find any such cases.


[0]	train-logloss:0.68835
[1]	train-logloss:0.68565
[2]	train-logloss:0.68298
[3]	train-logloss:0.67752
[4]	train-logloss:0.67465
[5]	train-logloss:0.67235
[6]	train-logloss:0.66660
[7]	train-logloss:0.66280
[8]	train-logloss:0.66026
[9]	train-logloss:0.65894
[10]	train-logloss:0.65901
[11]	train-logloss:0.65892
[12]	train-logloss:0.65751
[13]	train-logloss:0.65512
[14]	train-logloss:0.65389
[15]	train-logloss:0.65229
[16]	train-logloss:0.64792
[17]	train-logloss:0.64436
[18]	train-logloss:0.64343
[19]	train-logloss:0.64374
[20]	train-logloss:0.64223
[21]	train-logloss:0.63890
[22]	train-logloss:0.63934
[23]	train-logloss:0.63531
[24]	train-logloss:0.63163
[25]	train-logloss:0.63014
[26]	train-logloss:0.62985
[27]	train-logloss:0.62939
[28]	train-logloss:0.62872
[29]	train-logloss:0.62832
[30]	train-logloss:0.62718
[31]	train-logloss:0.62531
[32]	train-logloss:0.62274
[33]	train-logloss:0.62034
[34]	train-logloss:0.61853
[35]	train-logloss:0.61825
[36]	train-logloss:0.61698
[37]	train-logloss:0.61518
[38]	train-logloss:0.61462
[39]	train-logloss:0.61375
[40]	train-logloss:0.61137
[41]	train-logloss:0.61013
[42]	train-logloss:0.61013
[43]	train-logloss:0.61091
[44]	train-logloss:0.60978
[45]	train-logloss:0.60987
[46]	train-logloss:0.60909
[47]	train-logloss:0.60926
[48]	train-logloss:0.60889
[49]	train-logloss:0.60833
[50]	train-logloss:0.60849
[51]	train-logloss:0.60889
[52]	train-logloss:0.60871
[53]	train-logloss:0.60861
[54]	train-logloss:0.60935
[55]	train-logloss:0.60868
[56]	train-logloss:0.60836
[57]	train-logloss:0.60862
[58]	train-logloss:0.60933
[59]	train-logloss:0.60926
[60]	train-logloss:0.60929
[61]	train-logloss:0.60936
[62]	train-logloss:0.60876
[63]	train-logloss:0.60862
[64]	train-logloss:0.60866
[65]	train-logloss:0.60921
[66]	train-logloss:0.60946
[67]	train-logloss:0.60896
[68]	train-logloss:0.60919
[69]	train-logloss:0.60852
[70]	train-logloss:0.60873
[71]	train-logloss:0.60902
[72]	train-logloss:0.60903
[73]	train-logloss:0.60881
[74]	train-logloss:0.60862
[75]	train-logloss:0.60658
[76]	train-logloss:0.60641
[77]	train-logloss:0.60657
[78]	train-logloss:0.60661
[79]	train-logloss:0.60736
[80]	train-logloss:0.60740
[81]	train-logloss:0.60726
[82]	train-logloss:0.60717
[83]	train-logloss:0.60745
[84]	train-logloss:0.60663
[85]	train-logloss:0.60681
[86]	train-logloss:0.60718
[87]	train-logloss:0.60616
[88]	train-logloss:0.60682
[89]	train-logloss:0.60632
[90]	train-logloss:0.60609
[91]	train-logloss:0.60548
[92]	train-logloss:0.60544
[93]	train-logloss:0.60522
[94]	train-logloss:0.60536
[95]	train-logloss:0.60596
[96]	train-logloss:0.60680
[97]	train-logloss:0.60665
[98]	train-logloss:0.60742
[99]	train-logloss:0.60716
[100]	train-logloss:0.60704
[101]	train-logloss:0.60628
[102]	train-logloss:0.60648
[103]	train-logloss:0.60658
[104]	train-logloss:0.60748
[105]	train-logloss:0.60746
[106]	train-logloss:0.60750
[107]	train-logloss:0.60736
[108]	train-logloss:0.60640
[109]	train-logloss:0.60703
[110]	train-logloss:0.60651
[111]	train-logloss:0.60647
[112]	train-logloss:0.60556
[113]	train-logloss:0.60544
[114]	train-logloss:0.60372
[115]	train-logloss:0.60246
[116]	train-logloss:0.60285
[117]	train-logloss:0.60266
[118]	train-logloss:0.60286
[119]	train-logloss:0.60331
[120]	train-logloss:0.60429
[121]	train-logloss:0.60428
[122]	train-logloss:0.60386
[123]	train-logloss:0.60349
[124]	train-logloss:0.60357
[125]	train-logloss:0.60228
[126]	train-logloss:0.60228
[127]	train-logloss:0.60304
[128]	train-logloss:0.60288
[129]	train-logloss:0.60234
[130]	train-logloss:0.60196
[131]	train-logloss:0.60220
[132]	train-logloss:0.60163
[133]	train-logloss:0.60118
[134]	train-logloss:0.60188
[135]	train-logloss:0.60089
[136]	train-logloss:0.60052
[137]	train-logloss:0.60121
[138]	train-logloss:0.60029
[139]	train-logloss:0.59980
[140]	train-logloss:0.60066
[141]	train-logloss:0.60037
[142]	train-logloss:0.60084
[143]	train-logloss:0.60068
[144]	train-logloss:0.60141
[145]	train-logloss:0.60053
[146]	train-logloss:0.60028
[147]	train-logloss:0.60044
[148]	train-logloss:0.59957
[149]	train-logloss:0.60004
[150]	train-logloss:0.59962
[151]	train-logloss:0.59961
[152]	train-logloss:0.59938
[153]	train-logloss:0.59880
[154]	train-logloss:0.59873
[155]	train-logloss:0.59878
[156]	train-logloss:0.59905
[157]	train-logloss:0.59885
[158]	train-logloss:0.59913
[159]	train-logloss:0.59885
[160]	train-logloss:0.59845
[161]	train-logloss:0.59908
[162]	train-logloss:0.59909
[163]	train-logloss:0.59804
[164]	train-logloss:0.59788
[165]	train-logloss:0.59796
[166]	train-logloss:0.59915
[167]	train-logloss:0.59874
[168]	train-logloss:0.59868
[169]	train-logloss:0.59866
[170]	train-logloss:0.59915
[171]	train-logloss:0.59945
[172]	train-logloss:0.59978
[173]	train-logloss:0.59945
[174]	train-logloss:0.59956
[175]	train-logloss:0.59835
[176]	train-logloss:0.59840
[177]	train-logloss:0.59836
[178]	train-logloss:0.59825
[179]	train-logloss:0.59791
[180]	train-logloss:0.59836
[181]	train-logloss:0.59813
[182]	train-logloss:0.59832
[183]	train-logloss:0.59790
[184]	train-logloss:0.59847
[185]	train-logloss:0.59873
[186]	train-logloss:0.59886
[187]	train-logloss:0.59942
[188]	train-logloss:0.59865
[189]	train-logloss:0.59852
[190]	train-logloss:0.59852
[191]	train-logloss:0.59848
[192]	train-logloss:0.59884
[193]	train-logloss:0.59845
[194]	train-logloss:0.59827
[195]	train-logloss:0.59773
[196]	train-logloss:0.59742
[197]	train-logloss:0.59782
[198]	train-logloss:0.59742
[199]	train-logloss:0.59765
[200]	train-logloss:0.59699
[201]	train-logloss:0.59748
[202]	train-logloss:0.59788
[203]	train-logloss:0.59799
[204]	train-logloss:0.59756
[205]	train-logloss:0.59685
[206]	train-logloss:0.59746
[207]	train-logloss:0.59756
[208]	train-logloss:0.59718
[209]	train-logloss:0.59742
[210]	train-logloss:0.59784
[211]	train-logloss:0.59826
[212]	train-logloss:0.59800
[213]	train-logloss:0.59736
[214]	train-logloss:0.59694
[215]	train-logloss:0.59707
[216]	train-logloss:0.59706
[217]	train-logloss:0.59695
[218]	train-logloss:0.59711
[219]	train-logloss:0.59697
[220]	train-logloss:0.59773
[221]	train-logloss:0.59839
[222]	train-logloss:0.59860
[223]	train-logloss:0.59783
[224]	train-logloss:0.59776
[225]	train-logloss:0.59783
[226]	train-logloss:0.59780
[227]	train-logloss:0.59815
[228]	train-logloss:0.59765
[229]	train-logloss:0.59831
[230]	train-logloss:0.59830
[231]	train-logloss:0.59818
[232]	train-logloss:0.59829
[233]	train-logloss:0.59806
[234]	train-logloss:0.59734
[235]	train-logloss:0.59763
[236]	train-logloss:0.59748
[237]	train-logloss:0.59630
[238]	train-logloss:0.59615
[239]	train-logloss:0.59571
[240]	train-logloss:0.59605
[241]	train-logloss:0.59521
[242]	train-logloss:0.59485
[243]	train-logloss:0.59427
[244]	train-logloss:0.59476
[245]	train-logloss:0.59555
[246]	train-logloss:0.59568
[247]	train-logloss:0.59555
[248]	train-logloss:0.59653
[249]	train-logloss:0.59710
[250]	train-logloss:0.59722
[251]	train-logloss:0.59678
[252]	train-logloss:0.59689
[253]	train-logloss:0.59721
[254]	train-logloss:0.59773
[255]	train-logloss:0.59789
[256]	train-logloss:0.59814
[257]	train-logloss:0.59722
[258]	train-logloss:0.59697
[259]	train-logloss:0.59736
[260]	train-logloss:0.59678
[261]	train-logloss:0.59661
[262]	train-logloss:0.59701
[263]	train-logloss:0.59634
[264]	train-logloss:0.59628
[265]	train-logloss:0.59599
[266]	train-logloss:0.59570
[267]	train-logloss:0.59623
[268]	train-logloss:0.59656
[269]	train-logloss:0.59578
[270]	train-logloss:0.59617
[271]	train-logloss:0.59549
[272]	train-logloss:0.59521
[273]	train-logloss:0.59510
[274]	train-logloss:0.59484
[275]	train-logloss:0.59461
[276]	train-logloss:0.59496
[277]	train-logloss:0.59509
[278]	train-logloss:0.59511
[279]	train-logloss:0.59475
[280]	train-logloss:0.59425
[281]	train-logloss:0.59337
[282]	train-logloss:0.59408
[283]	train-logloss:0.59440
[284]	train-logloss:0.59461
[285]	train-logloss:0.59478
[286]	train-logloss:0.59540
[287]	train-logloss:0.59601
[288]	train-logloss:0.59565
[289]	train-logloss:0.59641
[290]	train-logloss:0.59619
[291]	train-logloss:0.59652
[292]	train-logloss:0.59666
[293]	train-logloss:0.59647
[294]	train-logloss:0.59690
[295]	train-logloss:0.59681
[296]	train-logloss:0.59674
[297]	train-logloss:0.59613
[298]	train-logloss:0.59633
[299]	train-logloss:0.59615
[300]	train-logloss:0.59657
[301]	train-logloss:0.59685
[302]	train-logloss:0.59679
[303]	train-logloss:0.59676
[304]	train-logloss:0.59651
[305]	train-logloss:0.59599
[306]	train-logloss:0.59591
[307]	train-logloss:0.59589
[308]	train-logloss:0.59606
[309]	train-logloss:0.59680
[310]	train-logloss:0.59755
[311]	train-logloss:0.59776
[312]	train-logloss:0.59839
[313]	train-logloss:0.59982
[314]	train-logloss:0.60061
[315]	train-logloss:0.60068
[316]	train-logloss:0.60074
[317]	train-logloss:0.60003
[318]	train-logloss:0.59996
[319]	train-logloss:0.59952
[320]	train-logloss:0.59922
[321]	train-logloss:0.59896
[322]	train-logloss:0.59843
[323]	train-logloss:0.59792
[324]	train-logloss:0.59771
[325]	train-logloss:0.59799
[326]	train-logloss:0.59850
[327]	train-logloss:0.59840
[328]	train-logloss:0.59858
[329]	train-logloss:0.59830
[330]	train-logloss:0.59859
[331]	train-logloss:0.59892
[332]	train-logloss:0.59962
[333]	train-logloss:0.59948
[334]	train-logloss:0.59957
[335]	train-logloss:0.59921
[336]	train-logloss:0.59992
[337]	train-logloss:0.60011
[338]	train-logloss:0.60025
[339]	train-logloss:0.60013
[340]	train-logloss:0.59981
[341]	train-logloss:0.59978
[342]	train-logloss:0.59933
[343]	train-logloss:0.59936
[344]	train-logloss:0.59835
[345]	train-logloss:0.59806
[346]	train-logloss:0.59652
[347]	train-logloss:0.59686
[348]	train-logloss:0.59685
[349]	train-logloss:0.59660
[350]	train-logloss:0.59550
[351]	train-logloss:0.59544
[352]	train-logloss:0.59591
[353]	train-logloss:0.59621
[354]	train-logloss:0.59615
[355]	train-logloss:0.59651
[356]	train-logloss:0.59627
[357]	train-logloss:0.59743
[358]	train-logloss:0.59777
[359]	train-logloss:0.59810
[360]	train-logloss:0.59777
[361]	train-logloss:0.59743
[362]	train-logloss:0.59659
[363]	train-logloss:0.59644
[364]	train-logloss:0.59640
[365]	train-logloss:0.59634
[366]	train-logloss:0.59636
[367]	train-logloss:0.59684
[368]	train-logloss:0.59731
[369]	train-logloss:0.59742
[370]	train-logloss:0.59739
[371]	train-logloss:0.59784
[372]	train-logloss:0.59729
[373]	train-logloss:0.59773
[374]	train-logloss:0.59768
[375]	train-logloss:0.59806
[376]	train-logloss:0.59811
[377]	train-logloss:0.59777
[378]	train-logloss:0.59874
[379]	train-logloss:0.59870
[380]	train-logloss:0.59868
[381]	train-logloss:0.59937
[382]	train-logloss:0.59917
[383]	train-logloss:0.59956
[384]	train-logloss:0.59952
[385]	train-logloss:0.59952
[386]	train-logloss:0.59907
[387]	train-logloss:0.59934
[388]	train-logloss:0.59920
[389]	train-logloss:0.59938
[390]	train-logloss:0.59972
[391]	train-logloss:0.59959
[392]	train-logloss:0.59966
[393]	train-logloss:0.59993
[394]	train-logloss:0.59983
[395]	train-logloss:0.60023
[396]	train-logloss:0.60025
[397]	train-logloss:0.60012
[398]	train-logloss:0.59959
[399]	train-logloss:0.59971
[400]	train-logloss:0.59964
[401]	train-logloss:0.59952
[402]	train-logloss:0.59944
[403]	train-logloss:0.59939
[404]	train-logloss:0.59934
[405]	train-logloss:0.59978
[406]	train-logloss:0.59954
[407]	train-logloss:0.59956
[408]	train-logloss:0.59985
[409]	train-logloss:0.59924
[410]	train-logloss:0.59999
[411]	train-logloss:0.60040
[412]	train-logloss:0.60098
[413]	train-logloss:0.60030
[414]	train-logloss:0.60028
[415]	train-logloss:0.59985
[416]	train-logloss:0.60055
[417]	train-logloss:0.60067
[418]	train-logloss:0.60093
[419]	train-logloss:0.60046
[420]	train-logloss:0.60099
[421]	train-logloss:0.60128
[422]	train-logloss:0.60063
[423]	train-logloss:0.60044
[424]	train-logloss:0.60062
[425]	train-logloss:0.60075
[426]	train-logloss:0.60039
[427]	train-logloss:0.60039
[428]	train-logloss:0.60120
[429]	train-logloss:0.60134
[430]	train-logloss:0.60121
[431]	train-logloss:0.60132
[432]	train-logloss:0.60147
[433]	train-logloss:0.60110
[434]	train-logloss:0.60113
[435]	train-logloss:0.60103
[436]	train-logloss:0.60065
[437]	train-logloss:0.60031
[438]	train-logloss:0.60043
[439]	train-logloss:0.60048
[440]	train-logloss:0.60005
[441]	train-logloss:0.59975
[442]	train-logloss:0.59958
[443]	train-logloss:0.59946
[444]	train-logloss:0.59932
[445]	train-logloss:0.59964
[446]	train-logloss:0.59884
[447]	train-logloss:0.59847
[448]	train-logloss:0.59863
[449]	train-logloss:0.59869
[450]	train-logloss:0.59856
[451]	train-logloss:0.59894
[452]	train-logloss:0.59901
[453]	train-logloss:0.59873
[454]	train-logloss:0.59953
[455]	train-logloss:0.59953
[456]	train-logloss:0.59972
[457]	train-logloss:0.59962
[458]	train-logloss:0.59994
[459]	train-logloss:0.60006
[460]	train-logloss:0.60028
[461]	train-logloss:0.60110
[462]	train-logloss:0.60111
[463]	train-logloss:0.60122
[464]	train-logloss:0.60074
[465]	train-logloss:0.60093
[466]	train-logloss:0.60080
[467]	train-logloss:0.60120
[468]	train-logloss:0.60122
[469]	train-logloss:0.60124
[470]	train-logloss:0.60122
[471]	train-logloss:0.60116
[472]	train-logloss:0.60101
[473]	train-logloss:0.60090
[474]	train-logloss:0.60111
[475]	train-logloss:0.60109
[476]	train-logloss:0.60151
[477]	train-logloss:0.60201
[478]	train-logloss:0.60160
[479]	train-logloss:0.60101
[480]	train-logloss:0.60132
[481]	train-logloss:0.60067
[482]	train-logloss:0.60054
[483]	train-logloss:0.60041
[484]	train-logloss:0.60017
[485]	train-logloss:0.60025
[486]	train-logloss:0.60024
[487]	train-logloss:0.59967
[488]	train-logloss:0.59935
[489]	train-logloss:0.59868
[490]	train-logloss:0.59907
[491]	train-logloss:0.59912
[492]	train-logloss:0.59919
[493]	train-logloss:0.59890
[494]	train-logloss:0.59955
[495]	train-logloss:0.59947
[496]	train-logloss:0.59907
[497]	train-logloss:0.59937
[498]	train-logloss:0.59933
[499]	train-logloss:0.59960
[500]	train-logloss:0.60029
[501]	train-logloss:0.60047
[502]	train-logloss:0.60013
[503]	train-logloss:0.59989
[504]	train-logloss:0.60059
[505]	train-logloss:0.60072
[506]	train-logloss:0.60102
[507]	train-logloss:0.60086
[508]	train-logloss:0.60060
[509]	train-logloss:0.60126
[510]	train-logloss:0.60112
[511]	train-logloss:0.60126
[512]	train-logloss:0.60129
[513]	train-logloss:0.60059
[514]	train-logloss:0.59989
[515]	train-logloss:0.60005
[516]	train-logloss:0.59968
[517]	train-logloss:0.60008
[518]	train-logloss:0.60084
[519]	train-logloss:0.60062
[520]	train-logloss:0.60111
[521]	train-logloss:0.60070
[522]	train-logloss:0.60063
[523]	train-logloss:0.60065
[524]	train-logloss:0.60044
[525]	train-logloss:0.60053
[526]	train-logloss:0.60099
[527]	train-logloss:0.60125
[528]	train-logloss:0.60105
[529]	train-logloss:0.60155
[530]	train-logloss:0.60176
[531]	train-logloss:0.60249
[532]	train-logloss:0.60304
[533]	train-logloss:0.60372
[534]	train-logloss:0.60326
[535]	train-logloss:0.60391
[536]	train-logloss:0.60371
[537]	train-logloss:0.60472
[538]	train-logloss:0.60431
[539]	train-logloss:0.60337
[540]	train-logloss:0.60355
[541]	train-logloss:0.60365
[542]	train-logloss:0.60295
[543]	train-logloss:0.60268
[544]	train-logloss:0.60312
[545]	train-logloss:0.60293
[546]	train-logloss:0.60275
[547]	train-logloss:0.60344
[548]	train-logloss:0.60334
[549]	train-logloss:0.60411
[550]	train-logloss:0.60460
[551]	train-logloss:0.60409
[552]	train-logloss:0.60423
[553]	train-logloss:0.60366
[554]	train-logloss:0.60341
[555]	train-logloss:0.60364
[556]	train-logloss:0.60365
[557]	train-logloss:0.60316
[558]	train-logloss:0.60353
[559]	train-logloss:0.60382
[560]	train-logloss:0.60396
[561]	train-logloss:0.60426
[562]	train-logloss:0.60465
[563]	train-logloss:0.60500
[564]	train-logloss:0.60502
[565]	train-logloss:0.60465
[566]	train-logloss:0.60496
[567]	train-logloss:0.60545
[568]	train-logloss:0.60523
[569]	train-logloss:0.60451
[570]	train-logloss:0.60424
[571]	train-logloss:0.60479
[572]	train-logloss:0.60501
[573]	train-logloss:0.60472
[574]	train-logloss:0.60399
[575]	train-logloss:0.60399
[576]	train-logloss:0.60339
[577]	train-logloss:0.60306
[578]	train-logloss:0.60286
[579]	train-logloss:0.60302
[580]	train-logloss:0.60266
[581]	train-logloss:0.60206
[582]	train-logloss:0.60216
[583]	train-logloss:0.60172
[584]	train-logloss:0.60186
[585]	train-logloss:0.60173
[586]	train-logloss:0.60159
[587]	train-logloss:0.60130
[588]	train-logloss:0.60173
[589]	train-logloss:0.60182
[590]	train-logloss:0.60176
[591]	train-logloss:0.60224
[592]	train-logloss:0.60242
[593]	train-logloss:0.60209
[594]	train-logloss:0.60148
[595]	train-logloss:0.60173
[596]	train-logloss:0.60187
[597]	train-logloss:0.60157
[598]	train-logloss:0.60219
[599]	train-logloss:0.60211
[600]	train-logloss:0.60197
[601]	train-logloss:0.60217
[602]	train-logloss:0.60158
[603]	train-logloss:0.60171
[604]	train-logloss:0.60143
[605]	train-logloss:0.60067
[606]	train-logloss:0.60052
[607]	train-logloss:0.60008
[608]	train-logloss:0.59992
[609]	train-logloss:0.60023
[610]	train-logloss:0.60063
[611]	train-logloss:0.60079
[612]	train-logloss:0.60056
[613]	train-logloss:0.60045
[614]	train-logloss:0.60035
[615]	train-logloss:0.60040
[616]	train-logloss:0.60038
[617]	train-logloss:0.60047
[618]	train-logloss:0.60006
[619]	train-logloss:0.60058
[620]	train-logloss:0.60048
[621]	train-logloss:0.60130
[622]	train-logloss:0.60134
[623]	train-logloss:0.60108
[624]	train-logloss:0.60107
[625]	train-logloss:0.60103
[626]	train-logloss:0.60110
[627]	train-logloss:0.60111
[628]	train-logloss:0.60118
[629]	train-logloss:0.60107
[630]	train-logloss:0.60026
[631]	train-logloss:0.60035
[632]	train-logloss:0.60089
[633]	train-logloss:0.60139
[634]	train-logloss:0.60136
[635]	train-logloss:0.60107
[636]	train-logloss:0.60094
[637]	train-logloss:0.60075
[638]	train-logloss:0.60102
[639]	train-logloss:0.60164
[640]	train-logloss:0.60075
[641]	train-logloss:0.60064
[642]	train-logloss:0.60051
[643]	train-logloss:0.60076
[644]	train-logloss:0.60053
[645]	train-logloss:0.60062
[646]	train-logloss:0.60055
[647]	train-logloss:0.60115
[648]	train-logloss:0.60093
[649]	train-logloss:0.60052
[650]	train-logloss:0.60054
[651]	train-logloss:0.60064
[652]	train-logloss:0.60126
[653]	train-logloss:0.60113
[654]	train-logloss:0.60096
[655]	train-logloss:0.60108
[656]	train-logloss:0.60129
[657]	train-logloss:0.60122
[658]	train-logloss:0.60162
[659]	train-logloss:0.60155
[660]	train-logloss:0.60163
[661]	train-logloss:0.60166
[662]	train-logloss:0.60170
[663]	train-logloss:0.60317
[664]	train-logloss:0.60358
[665]	train-logloss:0.60430
[666]	train-logloss:0.60406
[667]	train-logloss:0.60419
[668]	train-logloss:0.60394
[669]	train-logloss:0.60423
[670]	train-logloss:0.60479
[671]	train-logloss:0.60492
[672]	train-logloss:0.60493
[673]	train-logloss:0.60458
[674]	train-logloss:0.60413
[675]	train-logloss:0.60381
[676]	train-logloss:0.60380
[677]	train-logloss:0.60329
[678]	train-logloss:0.60327
[679]	train-logloss:0.60334
[680]	train-logloss:0.60352
[681]	train-logloss:0.60370
[682]	train-logloss:0.60361
[683]	train-logloss:0.60389
[684]	train-logloss:0.60361
[685]	train-logloss:0.60419
[686]	train-logloss:0.60502
[687]	train-logloss:0.60500
[688]	train-logloss:0.60507
[689]	train-logloss:0.60466
[690]	train-logloss:0.60461
[691]	train-logloss:0.60461
[692]	train-logloss:0.60505
[693]	train-logloss:0.60527
[694]	train-logloss:0.60532
[695]	train-logloss:0.60534
[696]	train-logloss:0.60565
[697]	train-logloss:0.60592
[698]	train-logloss:0.60541
[699]	train-logloss:0.60534
[700]	train-logloss:0.60509
[701]	train-logloss:0.60491
[702]	train-logloss:0.60503
[703]	train-logloss:0.60507
[704]	train-logloss:0.60564
[705]	train-logloss:0.60548
[706]	train-logloss:0.60611
[707]	train-logloss:0.60603
[708]	train-logloss:0.60553
[709]	train-logloss:0.60522
[710]	train-logloss:0.60433
[711]	train-logloss:0.60431
[712]	train-logloss:0.60441
[713]	train-logloss:0.60433
[714]	train-logloss:0.60479
[715]	train-logloss:0.60464
[716]	train-logloss:0.60522
[717]	train-logloss:0.60565
[718]	train-logloss:0.60521
[719]	train-logloss:0.60472
[720]	train-logloss:0.60502
[721]	train-logloss:0.60541
[722]	train-logloss:0.60551
[723]	train-logloss:0.60531
[724]	train-logloss:0.60464
[725]	train-logloss:0.60453
[726]	train-logloss:0.60449
[727]	train-logloss:0.60426
[728]	train-logloss:0.60378
[729]	train-logloss:0.60523
[730]	train-logloss:0.60574
[731]	train-logloss:0.60550
[732]	train-logloss:0.60547
[733]	train-logloss:0.60580
[734]	train-logloss:0.60546
[735]	train-logloss:0.60541
[736]	train-logloss:0.60566
[737]	train-logloss:0.60568
[738]	train-logloss:0.60556
[739]	train-logloss:0.60546
[740]	train-logloss:0.60533
[741]	train-logloss:0.60570
[742]	train-logloss:0.60580
[743]	train-logloss:0.60562
[744]	train-logloss:0.60563
[745]	train-logloss:0.60553
[746]	train-logloss:0.60570
[747]	train-logloss:0.60584
[748]	train-logloss:0.60632
[749]	train-logloss:0.60628
[750]	train-logloss:0.60637
[751]	train-logloss:0.60680
[752]	train-logloss:0.60716
[753]	train-logloss:0.60663
[754]	train-logloss:0.60630
[755]	train-logloss:0.60617
[756]	train-logloss:0.60614
[757]	train-logloss:0.60527
[758]	train-logloss:0.60568
[759]	train-logloss:0.60560
[760]	train-logloss:0.60595
[761]	train-logloss:0.60631
[762]	train-logloss:0.60588
[763]	train-logloss:0.60584
[764]	train-logloss:0.60627
[765]	train-logloss:0.60617
[766]	train-logloss:0.60665
[767]	train-logloss:0.60641
[768]	train-logloss:0.60655
[769]	train-logloss:0.60689
[770]	train-logloss:0.60710
[771]	train-logloss:0.60707
[772]	train-logloss:0.60664
[773]	train-logloss:0.60689
[774]	train-logloss:0.60732
[775]	train-logloss:0.60677
[776]	train-logloss:0.60677
[777]	train-logloss:0.60719
[778]	train-logloss:0.60771
[779]	train-logloss:0.60774
[780]	train-logloss:0.60803
[781]	train-logloss:0.60886
[782]	train-logloss:0.60919
[783]	train-logloss:0.60931
[784]	train-logloss:0.60956
[785]	train-logloss:0.60928
[786]	train-logloss:0.60890
[787]	train-logloss:0.60871
[788]	train-logloss:0.60884
[789]	train-logloss:0.60840
[790]	train-logloss:0.60815
[791]	train-logloss:0.60824
[792]	train-logloss:0.60808
[793]	train-logloss:0.60843
[794]	train-logloss:0.60818
[795]	train-logloss:0.60906
[796]	train-logloss:0.60931
[797]	train-logloss:0.60894
[798]	train-logloss:0.60874
[799]	train-logloss:0.60895
[800]	train-logloss:0.60818
[801]	train-logloss:0.60806
[802]	train-logloss:0.60856
[803]	train-logloss:0.60939
[804]	train-logloss:0.60937
[805]	train-logloss:0.60924
[806]	train-logloss:0.60880
[807]	train-logloss:0.60893
[808]	train-logloss:0.60851
[809]	train-logloss:0.60872
[810]	train-logloss:0.60823
[811]	train-logloss:0.60924
[812]	train-logloss:0.60916
[813]	train-logloss:0.60913
[814]	train-logloss:0.60906
[815]	train-logloss:0.60876
[816]	train-logloss:0.60875
[817]	train-logloss:0.60929
[818]	train-logloss:0.60952
[819]	train-logloss:0.60933
[820]	train-logloss:0.60891
[821]	train-logloss:0.60856
[822]	train-logloss:0.60921
[823]	train-logloss:0.60961
[824]	train-logloss:0.60921
[825]	train-logloss:0.60899
[826]	train-logloss:0.60953
[827]	train-logloss:0.61011
[828]	train-logloss:0.60985
[829]	train-logloss:0.60952
[830]	train-logloss:0.60889
[831]	train-logloss:0.60909
[832]	train-logloss:0.60925
[833]	train-logloss:0.60953
[834]	train-logloss:0.60918
[835]	train-logloss:0.60896
[836]	train-logloss:0.60951
[837]	train-logloss:0.60939
[838]	train-logloss:0.60935
[839]	train-logloss:0.60904
[840]	train-logloss:0.60951
[841]	train-logloss:0.61017
[842]	train-logloss:0.61034
[843]	train-logloss:0.61009
[844]	train-logloss:0.61010
[845]	train-logloss:0.61063
[846]	train-logloss:0.61112
[847]	train-logloss:0.61078
[848]	train-logloss:0.61036
[849]	train-logloss:0.61058
[850]	train-logloss:0.61066
[851]	train-logloss:0.61041
[852]	train-logloss:0.61029
[853]	train-logloss:0.60977
[854]	train-logloss:0.60990
[855]	train-logloss:0.60954
[856]	train-logloss:0.60964
[857]	train-logloss:0.60979
[858]	train-logloss:0.60995
[859]	train-logloss:0.60974
[860]	train-logloss:0.60945
[861]	train-logloss:0.60979
[862]	train-logloss:0.61024
[863]	train-logloss:0.61075
[864]	train-logloss:0.61087
[865]	train-logloss:0.61062
[866]	train-logloss:0.61108
[867]	train-logloss:0.61132
[868]	train-logloss:0.61127
[869]	train-logloss:0.61123
[870]	train-logloss:0.61163
[871]	train-logloss:0.61160
[872]	train-logloss:0.61153
[873]	train-logloss:0.61156
[874]	train-logloss:0.61207
[875]	train-logloss:0.61186
[876]	train-logloss:0.61301
[877]	train-logloss:0.61300
[878]	train-logloss:0.61276
[879]	train-logloss:0.61250
[880]	train-logloss:0.61269
[881]	train-logloss:0.61302
[882]	train-logloss:0.61330
[883]	train-logloss:0.61256
[884]	train-logloss:0.61219
[885]	train-logloss:0.61190
[886]	train-logloss:0.61175
[887]	train-logloss:0.61211
[888]	train-logloss:0.61195
[889]	train-logloss:0.61177
[890]	train-logloss:0.61180
[891]	train-logloss:0.61172
[892]	train-logloss:0.61242
[893]	train-logloss:0.61320
[894]	train-logloss:0.61337
[895]	train-logloss:0.61354
[896]	train-logloss:0.61354
[897]	train-logloss:0.61361
[898]	train-logloss:0.61390
[899]	train-logloss:0.61390
[900]	train-logloss:0.61439
[901]	train-logloss:0.61473
[902]	train-logloss:0.61455
[903]	train-logloss:0.61482
[904]	train-logloss:0.61491
[905]	train-logloss:0.61608
[906]	train-logloss:0.61604
[907]	train-logloss:0.61654
[908]	train-logloss:0.61628
[909]	train-logloss:0.61609
[910]	train-logloss:0.61661
[911]	train-logloss:0.61665
[912]	train-logloss:0.61649
[913]	train-logloss:0.61661
[914]	train-logloss:0.61669
[915]	train-logloss:0.61661
[916]	train-logloss:0.61669
[917]	train-logloss:0.61610
[918]	train-logloss:0.61622
[919]	train-logloss:0.61678
[920]	train-logloss:0.61674
[921]	train-logloss:0.61652
[922]	train-logloss:0.61651
[923]	train-logloss:0.61610
[924]	train-logloss:0.61625
[925]	train-logloss:0.61607
[926]	train-logloss:0.61634
[927]	train-logloss:0.61619
[928]	train-logloss:0.61594
[929]	train-logloss:0.61565
[930]	train-logloss:0.61541
[931]	train-logloss:0.61557
[932]	train-logloss:0.61549
[933]	train-logloss:0.61504
[934]	train-logloss:0.61500
[935]	train-logloss:0.61530
[936]	train-logloss:0.61608
[937]	train-logloss:0.61571
[938]	train-logloss:0.61553
[939]	train-logloss:0.61567
[940]	train-logloss:0.61549
[941]	train-logloss:0.61562
[942]	train-logloss:0.61594
[943]	train-logloss:0.61611
[944]	train-logloss:0.61579
[945]	train-logloss:0.61624
[946]	train-logloss:0.61548
[947]	train-logloss:0.61579
[948]	train-logloss:0.61570
[949]	train-logloss:0.61623
[950]	train-logloss:0.61624
[951]	train-logloss:0.61583
[952]	train-logloss:0.61581
[953]	train-logloss:0.61566
[954]	train-logloss:0.61573
[955]	train-logloss:0.61590
[956]	train-logloss:0.61602
[957]	train-logloss:0.61595
[958]	train-logloss:0.61607
[959]	train-logloss:0.61633
[960]	train-logloss:0.61581
[961]	train-logloss:0.61588
[962]	train-logloss:0.61593
[963]	train-logloss:0.61603
[964]	train-logloss:0.61550
[965]	train-logloss:0.61553
[966]	train-logloss:0.61595
[967]	train-logloss:0.61583
[968]	train-logloss:0.61558
[969]	train-logloss:0.61575
[970]	train-logloss:0.61599
[971]	train-logloss:0.61579
[972]	train-logloss:0.61623
[973]	train-logloss:0.61584
[974]	train-logloss:0.61529
[975]	train-logloss:0.61515
[976]	train-logloss:0.61492
[977]	train-logloss:0.61465
[978]	train-logloss:0.61481
[979]	train-logloss:0.61462
[980]	train-logloss:0.61420
[981]	train-logloss:0.61395
[982]	train-logloss:0.61406
[983]	train-logloss:0.61360
[984]	train-logloss:0.61340
[985]	train-logloss:0.61345
[986]	train-logloss:0.61342
[987]	train-logloss:0.61302
[988]	train-logloss:0.61285
[989]	train-logloss:0.61300
[990]	train-logloss:0.61285
[991]	train-logloss:0.61253
[992]	train-logloss:0.61262
[993]	train-logloss:0.61249
[994]	train-logloss:0.61250
[995]	train-logloss:0.61245
[996]	train-logloss:0.61260
[997]	train-logloss:0.61251
[998]	train-logloss:0.61306
[999]	train-logloss:0.61383
[1000]	train-logloss:0.61397
[1001]	train-logloss:0.61455
[1002]	train-logloss:0.61472
[1003]	train-logloss:0.61494
[1004]	train-logloss:0.61473
[1005]	train-logloss:0.61453
[1006]	train-logloss:0.61421
[1007]	train-logloss:0.61468
[1008]	train-logloss:0.61430
[1009]	train-logloss:0.61480
[1010]	train-logloss:0.61528
[1011]	train-logloss:0.61538
[1012]	train-logloss:0.61550
[1013]	train-logloss:0.61584
[1014]	train-logloss:0.61590
[1015]	train-logloss:0.61605
[1016]	train-logloss:0.61570
[1017]	train-logloss:0.61538
[1018]	train-logloss:0.61533
[1019]	train-logloss:0.61534
[1020]	train-logloss:0.61527
[1021]	train-logloss:0.61568
[1022]	train-logloss:0.61605
[1023]	train-logloss:0.61607
[1024]	train-logloss:0.61542
[1025]	train-logloss:0.61558
[1026]	train-logloss:0.61556
[1027]	train-logloss:0.61553
[1028]	train-logloss:0.61594
[1029]	train-logloss:0.61582
[1030]	train-logloss:0.61594
[1031]	train-logloss:0.61604
[1032]	train-logloss:0.61639
[1033]	train-logloss:0.61661
[1034]	train-logloss:0.61689
[1035]	train-logloss:0.61686
[1036]	train-logloss:0.61699
[1037]	train-logloss:0.61677
[1038]	train-logloss:0.61704
[1039]	train-logloss:0.61679
[1040]	train-logloss:0.61639
[1041]	train-logloss:0.61661
[1042]	train-logloss:0.61671
[1043]	train-logloss:0.61707
[1044]	train-logloss:0.61705
[1045]	train-logloss:0.61700
[1046]	train-logloss:0.61702
[1047]	train-logloss:0.61658
[1048]	train-logloss:0.61620
[1049]	train-logloss:0.61636
[1050]	train-logloss:0.61652
[1051]	train-logloss:0.61664
[1052]	train-logloss:0.61641
[1053]	train-logloss:0.61597
[1054]	train-logloss:0.61604
[1055]	train-logloss:0.61616
[1056]	train-logloss:0.61564
[1057]	train-logloss:0.61594
[1058]	train-logloss:0.61626
[1059]	train-logloss:0.61589
[1060]	train-logloss:0.61572
[1061]	train-logloss:0.61588
[1062]	train-logloss:0.61573
[1063]	train-logloss:0.61585
[1064]	train-logloss:0.61614
[1065]	train-logloss:0.61631
[1066]	train-logloss:0.61634
[1067]	train-logloss:0.61673
[1068]	train-logloss:0.61688
[1069]	train-logloss:0.61712
[1070]	train-logloss:0.61709
[1071]	train-logloss:0.61696
[1072]	train-logloss:0.61791
[1073]	train-logloss:0.61820
[1074]	train-logloss:0.61861
[1075]	train-logloss:0.61900
[1076]	train-logloss:0.61834
[1077]	train-logloss:0.61826
[1078]	train-logloss:0.61791
[1079]	train-logloss:0.61792
[1080]	train-logloss:0.61756
[1081]	train-logloss:0.61741
[1082]	train-logloss:0.61676
[1083]	train-logloss:0.61664
[1084]	train-logloss:0.61645
[1085]	train-logloss:0.61573
[1086]	train-logloss:0.61622
[1087]	train-logloss:0.61672
[1088]	train-logloss:0.61692
[1089]	train-logloss:0.61723
[1090]	train-logloss:0.61650
[1091]	train-logloss:0.61586
[1092]	train-logloss:0.61588
[1093]	train-logloss:0.61634
[1094]	train-logloss:0.61671
[1095]	train-logloss:0.61643
[1096]	train-logloss:0.61593
[1097]	train-logloss:0.61576
[1098]	train-logloss:0.61546
[1099]	train-logloss:0.61495
[1100]	train-logloss:0.61523
[1101]	train-logloss:0.61544
[1102]	train-logloss:0.61590
[1103]	train-logloss:0.61593
[1104]	train-logloss:0.61564
[1105]	train-logloss:0.61594
[1106]	train-logloss:0.61570
[1107]	train-logloss:0.61605
[1108]	train-logloss:0.61652
[1109]	train-logloss:0.61626
[1110]	train-logloss:0.61620
[1111]	train-logloss:0.61637
[1112]	train-logloss:0.61701
[1113]	train-logloss:0.61639
[1114]	train-logloss:0.61580
[1115]	train-logloss:0.61562
[1116]	train-logloss:0.61616
[1117]	train-logloss:0.61612
[1118]	train-logloss:0.61586
[1119]	train-logloss:0.61648
[1120]	train-logloss:0.61633
[1121]	train-logloss:0.61633
[1122]	train-logloss:0.61712
[1123]	train-logloss:0.61759
[1124]	train-logloss:0.61791
[1125]	train-logloss:0.61720
[1126]	train-logloss:0.61710
[1127]	train-logloss:0.61720
[1128]	train-logloss:0.61675
[1129]	train-logloss:0.61666
[1130]	train-logloss:0.61628
[1131]	train-logloss:0.61601
[1132]	train-logloss:0.61628
[1133]	train-logloss:0.61608
[1134]	train-logloss:0.61602
[1135]	train-logloss:0.61527
[1136]	train-logloss:0.61503
[1137]	train-logloss:0.61488
[1138]	train-logloss:0.61479
[1139]	train-logloss:0.61432
[1140]	train-logloss:0.61408
[1141]	train-logloss:0.61431
[1142]	train-logloss:0.61440
[1143]	train-logloss:0.61479
[1144]	train-logloss:0.61484
[1145]	train-logloss:0.61439
[1146]	train-logloss:0.61438
[1147]	train-logloss:0.61478
[1148]	train-logloss:0.61462
[1149]	train-logloss:0.61460
[1150]	train-logloss:0.61440
[1151]	train-logloss:0.61477
[1152]	train-logloss:0.61534
[1153]	train-logloss:0.61534
[1154]	train-logloss:0.61508
[1155]	train-logloss:0.61530
[1156]	train-logloss:0.61556
[1157]	train-logloss:0.61549
[1158]	train-logloss:0.61548
[1159]	train-logloss:0.61577
[1160]	train-logloss:0.61552
[1161]	train-logloss:0.61577
[1162]	train-logloss:0.61566
[1163]	train-logloss:0.61610
[1164]	train-logloss:0.61608
[1165]	train-logloss:0.61612
[1166]	train-logloss:0.61637
[1167]	train-logloss:0.61638
[1168]	train-logloss:0.61655
[1169]	train-logloss:0.61646
[1170]	train-logloss:0.61632
[1171]	train-logloss:0.61654
[1172]	train-logloss:0.61617
[1173]	train-logloss:0.61593
[1174]	train-logloss:0.61582
[1175]	train-logloss:0.61604
[1176]	train-logloss:0.61593
[1177]	train-logloss:0.61602
[1178]	train-logloss:0.61590
[1179]	train-logloss:0.61559
[1180]	train-logloss:0.61554
[1181]	train-logloss:0.61582
[1182]	train-logloss:0.61582
[1183]	train-logloss:0.61576
[1184]	train-logloss:0.61592
[1185]	train-logloss:0.61615
[1186]	train-logloss:0.61567
[1187]	train-logloss:0.61549
[1188]	train-logloss:0.61548
[1189]	train-logloss:0.61619
[1190]	train-logloss:0.61626
[1191]	train-logloss:0.61679
[1192]	train-logloss:0.61673
[1193]	train-logloss:0.61731
[1194]	train-logloss:0.61746
[1195]	train-logloss:0.61761
[1196]	train-logloss:0.61761
[1197]	train-logloss:0.61751
[1198]	train-logloss:0.61805
[1199]	train-logloss:0.61834
[1200]	train-logloss:0.61812
[1201]	train-logloss:0.61811
[1202]	train-logloss:0.61823
[1203]	train-logloss:0.61798
[1204]	train-logloss:0.61777
[1205]	train-logloss:0.61818
[1206]	train-logloss:0.61818
[1207]	train-logloss:0.61824
[1208]	train-logloss:0.61831
[1209]	train-logloss:0.61811
[1210]	train-logloss:0.61812
[1211]	train-logloss:0.61833
[1212]	train-logloss:0.61835
[1213]	train-logloss:0.61837
[1214]	train-logloss:0.61841
[1215]	train-logloss:0.61840
[1216]	train-logloss:0.61836
[1217]	train-logloss:0.61805
[1218]	train-logloss:0.61808
[1219]	train-logloss:0.61835
[1220]	train-logloss:0.61845
[1221]	train-logloss:0.61870
[1222]	train-logloss:0.61850
[1223]	train-logloss:0.61854
[1224]	train-logloss:0.61863
[1225]	train-logloss:0.61899
[1226]	train-logloss:0.61892
[1227]	train-logloss:0.61846
[1228]	train-logloss:0.61747
[1229]	train-logloss:0.61741
[1230]	train-logloss:0.61723
[1231]	train-logloss:0.61720
[1232]	train-logloss:0.61760
[1233]	train-logloss:0.61721
[1234]	train-logloss:0.61750
[1235]	train-logloss:0.61749
[1236]	train-logloss:0.61791
[1237]	train-logloss:0.61784
[1238]	train-logloss:0.61782
[1239]	train-logloss:0.61761
[1240]	train-logloss:0.61788
[1241]	train-logloss:0.61803
[1242]	train-logloss:0.61798
[1243]	train-logloss:0.61792
[1244]	train-logloss:0.61842
[1245]	train-logloss:0.61798
[1246]	train-logloss:0.61819
[1247]	train-logloss:0.61888
[1248]	train-logloss:0.61904
[1249]	train-logloss:0.61933
[1250]	train-logloss:0.61934
[1251]	train-logloss:0.61989
[1252]	train-logloss:0.61986
[1253]	train-logloss:0.61987
[1254]	train-logloss:0.62028
[1255]	train-logloss:0.62067
[1256]	train-logloss:0.62057
[1257]	train-logloss:0.62052
[1258]	train-logloss:0.62099
[1259]	train-logloss:0.62093
[1260]	train-logloss:0.62084
[1261]	train-logloss:0.62128
[1262]	train-logloss:0.62201
[1263]	train-logloss:0.62241
[1264]	train-logloss:0.62245
[1265]	train-logloss:0.62252
[1266]	train-logloss:0.62243
[1267]	train-logloss:0.62244
[1268]	train-logloss:0.62245
[1269]	train-logloss:0.62248
[1270]	train-logloss:0.62249
[1271]	train-logloss:0.62313
[1272]	train-logloss:0.62362
[1273]	train-logloss:0.62363
[1274]	train-logloss:0.62333
[1275]	train-logloss:0.62393
[1276]	train-logloss:0.62373
[1277]	train-logloss:0.62412
[1278]	train-logloss:0.62350
[1279]	train-logloss:0.62284
[1280]	train-logloss:0.62233
[1281]	train-logloss:0.62190
[1282]	train-logloss:0.62219
[1283]	train-logloss:0.62188
[1284]	train-logloss:0.62152
[1285]	train-logloss:0.62160
[1286]	train-logloss:0.62161
[1287]	train-logloss:0.62144
[1288]	train-logloss:0.62174
[1289]	train-logloss:0.62205
[1290]	train-logloss:0.62258
[1291]	train-logloss:0.62214
[1292]	train-logloss:0.62211
[1293]	train-logloss:0.62220
[1294]	train-logloss:0.62162
[1295]	train-logloss:0.62190
[1296]	train-logloss:0.62167
[1297]	train-logloss:0.62130
[1298]	train-logloss:0.62131
[1299]	train-logloss:0.62069
[1300]	train-logloss:0.62077
[1301]	train-logloss:0.62085
[1302]	train-logloss:0.62065
[1303]	train-logloss:0.62093
[1304]	train-logloss:0.62098
[1305]	train-logloss:0.62133
[1306]	train-logloss:0.62180
[1307]	train-logloss:0.62205
[1308]	train-logloss:0.62153
[1309]	train-logloss:0.62135
[1310]	train-logloss:0.62109
[1311]	train-logloss:0.62135
[1312]	train-logloss:0.62126
[1313]	train-logloss:0.62143
[1314]	train-logloss:0.62136
[1315]	train-logloss:0.62137
[1316]	train-logloss:0.62184
[1317]	train-logloss:0.62164
[1318]	train-logloss:0.62177
[1319]	train-logloss:0.62198
[1320]	train-logloss:0.62296
[1321]	train-logloss:0.62289
[1322]	train-logloss:0.62195
[1323]	train-logloss:0.62224
[1324]	train-logloss:0.62239
[1325]	train-logloss:0.62226
[1326]	train-logloss:0.62231
[1327]	train-logloss:0.62226
[1328]	train-logloss:0.62208
[1329]	train-logloss:0.62160
[1330]	train-logloss:0.62211
[1331]	train-logloss:0.62208
[1332]	train-logloss:0.62155
[1333]	train-logloss:0.62138
[1334]	train-logloss:0.62145
[1335]	train-logloss:0.62141
[1336]	train-logloss:0.62144
[1337]	train-logloss:0.62210
[1338]	train-logloss:0.62197
[1339]	train-logloss:0.62169
[1340]	train-logloss:0.62142
[1341]	train-logloss:0.62128
[1342]	train-logloss:0.62129
[1343]	train-logloss:0.62180
[1344]	train-logloss:0.62237
[1345]	train-logloss:0.62215
[1346]	train-logloss:0.62250
[1347]	train-logloss:0.62197
[1348]	train-logloss:0.62196
[1349]	train-logloss:0.62166
[1350]	train-logloss:0.62169
[1351]	train-logloss:0.62127
[1352]	train-logloss:0.62157
[1353]	train-logloss:0.62163
[1354]	train-logloss:0.62116
[1355]	train-logloss:0.62129
[1356]	train-logloss:0.62164
[1357]	train-logloss:0.62179
[1358]	train-logloss:0.62193
[1359]	train-logloss:0.62255
[1360]	train-logloss:0.62253
[1361]	train-logloss:0.62186
[1362]	train-logloss:0.62189
[1363]	train-logloss:0.62179
[1364]	train-logloss:0.62182
[1365]	train-logloss:0.62170
[1366]	train-logloss:0.62147
[1367]	train-logloss:0.62138
[1368]	train-logloss:0.62146
[1369]	train-logloss:0.62147
[1370]	train-logloss:0.62220
[1371]	train-logloss:0.62200
[1372]	train-logloss:0.62165
[1373]	train-logloss:0.62146
[1374]	train-logloss:0.62162
[1375]	train-logloss:0.62167
[1376]	train-logloss:0.62154
[1377]	train-logloss:0.62150
[1378]	train-logloss:0.62163
[1379]	train-logloss:0.62158
[1380]	train-logloss:0.62126
[1381]	train-logloss:0.62109
[1382]	train-logloss:0.62034
[1383]	train-logloss:0.62063
[1384]	train-logloss:0.61993
[1385]	train-logloss:0.62037
[1386]	train-logloss:0.62061
[1387]	train-logloss:0.62109
[1388]	train-logloss:0.62067
[1389]	train-logloss:0.62111
[1390]	train-logloss:0.62117
[1391]	train-logloss:0.62114
[1392]	train-logloss:0.62100
[1393]	train-logloss:0.62126
[1394]	train-logloss:0.62121
[1395]	train-logloss:0.62034
[1396]	train-logloss:0.62015
[1397]	train-logloss:0.61977
[1398]	train-logloss:0.61984
[1399]	train-logloss:0.61980
[1400]	train-logloss:0.62001
[1401]	train-logloss:0.62021
[1402]	train-logloss:0.61998
[1403]	train-logloss:0.61985
[1404]	train-logloss:0.62000
[1405]	train-logloss:0.61983
[1406]	train-logloss:0.62019
[1407]	train-logloss:0.62021
[1408]	train-logloss:0.62011
[1409]	train-logloss:0.62013
[1410]	train-logloss:0.62020
[1411]	train-logloss:0.62035
[1412]	train-logloss:0.62013
[1413]	train-logloss:0.62051
[1414]	train-logloss:0.62023
[1415]	train-logloss:0.61969
[1416]	train-logloss:0.61964
[1417]	train-logloss:0.62012
[1418]	train-logloss:0.61977
[1419]	train-logloss:0.62004
[1420]	train-logloss:0.61985
[1421]	train-logloss:0.62022
[1422]	train-logloss:0.62018
[1423]	train-logloss:0.62115
[1424]	train-logloss:0.62131
[1425]	train-logloss:0.62105
[1426]	train-logloss:0.62091
[1427]	train-logloss:0.62092
[1428]	train-logloss:0.62157
[1429]	train-logloss:0.62142
[1430]	train-logloss:0.62116
[1431]	train-logloss:0.62139
[1432]	train-logloss:0.62133
[1433]	train-logloss:0.62163
[1434]	train-logloss:0.62205
[1435]	train-logloss:0.62173
[1436]	train-logloss:0.62203
[1437]	train-logloss:0.62223
[1438]	train-logloss:0.62139
[1439]	train-logloss:0.62153
[1440]	train-logloss:0.62179
[1441]	train-logloss:0.62182
[1442]	train-logloss:0.62184
[1443]	train-logloss:0.62181
[1444]	train-logloss:0.62181
[1445]	train-logloss:0.62172
[1446]	train-logloss:0.62191
[1447]	train-logloss:0.62234
[1448]	train-logloss:0.62249
[1449]	train-logloss:0.62289
[1450]	train-logloss:0.62289
[1451]	train-logloss:0.62240
[1452]	train-logloss:0.62203
[1453]	train-logloss:0.62179
[1454]	train-logloss:0.62148
[1455]	train-logloss:0.62208
[1456]	train-logloss:0.62211
[1457]	train-logloss:0.62210
[1458]	train-logloss:0.62212
[1459]	train-logloss:0.62243
[1460]	train-logloss:0.62176
[1461]	train-logloss:0.62173
[1462]	train-logloss:0.62273
[1463]	train-logloss:0.62274
[1464]	train-logloss:0.62264
[1465]	train-logloss:0.62251
[1466]	train-logloss:0.62216
[1467]	train-logloss:0.62178
[1468]	train-logloss:0.62177
[1469]	train-logloss:0.62123
[1470]	train-logloss:0.62158
[1471]	train-logloss:0.62149
[1472]	train-logloss:0.62120
[1473]	train-logloss:0.62089
[1474]	train-logloss:0.62088
[1475]	train-logloss:0.62042
[1476]	train-logloss:0.62060
[1477]	train-logloss:0.62094
[1478]	train-logloss:0.62070
[1479]	train-logloss:0.62138
[1480]	train-logloss:0.62191
[1481]	train-logloss:0.62263
[1482]	train-logloss:0.62314
[1483]	train-logloss:0.62297
[1484]	train-logloss:0.62304
[1485]	train-logloss:0.62302
[1486]	train-logloss:0.62320
[1487]	train-logloss:0.62371
[1488]	train-logloss:0.62408
[1489]	train-logloss:0.62425
[1490]	train-logloss:0.62483
[1491]	train-logloss:0.62470
[1492]	train-logloss:0.62468
[1493]	train-logloss:0.62445
[1494]	train-logloss:0.62364
[1495]	train-logloss:0.62281
[1496]	train-logloss:0.62235
[1497]	train-logloss:0.62246
[1498]	train-logloss:0.62299
[1499]	train-logloss:0.62292
[1500]	train-logloss:0.62292
[1501]	train-logloss:0.62397
[1502]	train-logloss:0.62421
[1503]	train-logloss:0.62474
[1504]	train-logloss:0.62482
[1505]	train-logloss:0.62449
[1506]	train-logloss:0.62440
[1507]	train-logloss:0.62389
[1508]	train-logloss:0.62370
[1509]	train-logloss:0.62357
[1510]	train-logloss:0.62330
[1511]	train-logloss:0.62317
[1512]	train-logloss:0.62402
[1513]	train-logloss:0.62354
[1514]	train-logloss:0.62335
[1515]	train-logloss:0.62294
[1516]	train-logloss:0.62292
[1517]	train-logloss:0.62292
[1518]	train-logloss:0.62291
[1519]	train-logloss:0.62241
[1520]	train-logloss:0.62281
[1521]	train-logloss:0.62292
[1522]	train-logloss:0.62264
[1523]	train-logloss:0.62284
[1524]	train-logloss:0.62344
[1525]	train-logloss:0.62342
[1526]	train-logloss:0.62341
[1527]	train-logloss:0.62322
[1528]	train-logloss:0.62380
[1529]	train-logloss:0.62396
[1530]	train-logloss:0.62362
[1531]	train-logloss:0.62355
[1532]	train-logloss:0.62339
[1533]	train-logloss:0.62331
[1534]	train-logloss:0.62320
[1535]	train-logloss:0.62286
[1536]	train-logloss:0.62317
[1537]	train-logloss:0.62443
[1538]	train-logloss:0.62493
[1539]	train-logloss:0.62527
[1540]	train-logloss:0.62483
[1541]	train-logloss:0.62509
[1542]	train-logloss:0.62480
[1543]	train-logloss:0.62506
[1544]	train-logloss:0.62635
[1545]	train-logloss:0.62708
[1546]	train-logloss:0.62721
[1547]	train-logloss:0.62686
[1548]	train-logloss:0.62723
[1549]	train-logloss:0.62748
[1550]	train-logloss:0.62745
[1551]	train-logloss:0.62808
[1552]	train-logloss:0.62749
[1553]	train-logloss:0.62703
[1554]	train-logloss:0.62705
[1555]	train-logloss:0.62714
[1556]	train-logloss:0.62733
[1557]	train-logloss:0.62796
[1558]	train-logloss:0.62826
[1559]	train-logloss:0.62826
[1560]	train-logloss:0.62829
[1561]	train-logloss:0.62839
[1562]	train-logloss:0.62812
[1563]	train-logloss:0.62794
[1564]	train-logloss:0.62794
[1565]	train-logloss:0.62733
[1566]	train-logloss:0.62713
[1567]	train-logloss:0.62760
[1568]	train-logloss:0.62765
[1569]	train-logloss:0.62734
[1570]	train-logloss:0.62715
[1571]	train-logloss:0.62716
[1572]	train-logloss:0.62697
[1573]	train-logloss:0.62685
[1574]	train-logloss:0.62616
[1575]	train-logloss:0.62604
[1576]	train-logloss:0.62584
[1577]	train-logloss:0.62552
[1578]	train-logloss:0.62563
[1579]	train-logloss:0.62520
[1580]	train-logloss:0.62522
[1581]	train-logloss:0.62523
[1582]	train-logloss:0.62511
[1583]	train-logloss:0.62505
[1584]	train-logloss:0.62541
[1585]	train-logloss:0.62588
[1586]	train-logloss:0.62578
[1587]	train-logloss:0.62553
[1588]	train-logloss:0.62557
[1589]	train-logloss:0.62467
[1590]	train-logloss:0.62473
[1591]	train-logloss:0.62508
[1592]	train-logloss:0.62497
[1593]	train-logloss:0.62453
[1594]	train-logloss:0.62384
[1595]	train-logloss:0.62420
[1596]	train-logloss:0.62446
[1597]	train-logloss:0.62479
[1598]	train-logloss:0.62449
[1599]	train-logloss:0.62449
[1600]	train-logloss:0.62423
[1601]	train-logloss:0.62411
[1602]	train-logloss:0.62388
[1603]	train-logloss:0.62411
[1604]	train-logloss:0.62443
[1605]	train-logloss:0.62469
[1606]	train-logloss:0.62507
[1607]	train-logloss:0.62572
[1608]	train-logloss:0.62554
[1609]	train-logloss:0.62555
[1610]	train-logloss:0.62558
[1611]	train-logloss:0.62570
[1612]	train-logloss:0.62653
[1613]	train-logloss:0.62706
[1614]	train-logloss:0.62691
[1615]	train-logloss:0.62700
[1616]	train-logloss:0.62672
[1617]	train-logloss:0.62688
[1618]	train-logloss:0.62700
[1619]	train-logloss:0.62699
[1620]	train-logloss:0.62742
[1621]	train-logloss:0.62767
[1622]	train-logloss:0.62734
[1623]	train-logloss:0.62717
[1624]	train-logloss:0.62756
[1625]	train-logloss:0.62705
[1626]	train-logloss:0.62695
[1627]	train-logloss:0.62633
[1628]	train-logloss:0.62619
[1629]	train-logloss:0.62691
[1630]	train-logloss:0.62652
[1631]	train-logloss:0.62642
[1632]	train-logloss:0.62627
[1633]	train-logloss:0.62633
[1634]	train-logloss:0.62699
[1635]	train-logloss:0.62705
[1636]	train-logloss:0.62704
[1637]	train-logloss:0.62736
[1638]	train-logloss:0.62731
[1639]	train-logloss:0.62708
[1640]	train-logloss:0.62668
[1641]	train-logloss:0.62663
[1642]	train-logloss:0.62660
[1643]	train-logloss:0.62673
[1644]	train-logloss:0.62695
[1645]	train-logloss:0.62719
[1646]	train-logloss:0.62804
[1647]	train-logloss:0.62804
[1648]	train-logloss:0.62861
[1649]	train-logloss:0.62823
[1650]	train-logloss:0.62817
[1651]	train-logloss:0.62793
[1652]	train-logloss:0.62743
[1653]	train-logloss:0.62737
[1654]	train-logloss:0.62774
[1655]	train-logloss:0.62777
[1656]	train-logloss:0.62778
[1657]	train-logloss:0.62840
[1658]	train-logloss:0.62773
[1659]	train-logloss:0.62748
[1660]	train-logloss:0.62749
[1661]	train-logloss:0.62737
[1662]	train-logloss:0.62715
[1663]	train-logloss:0.62719
[1664]	train-logloss:0.62730
[1665]	train-logloss:0.62723
[1666]	train-logloss:0.62722
[1667]	train-logloss:0.62713
[1668]	train-logloss:0.62705
[1669]	train-logloss:0.62717
[1670]	train-logloss:0.62800
[1671]	train-logloss:0.62689
[1672]	train-logloss:0.62649
[1673]	train-logloss:0.62711
[1674]	train-logloss:0.62687
[1675]	train-logloss:0.62650
[1676]	train-logloss:0.62633
[1677]	train-logloss:0.62623
[1678]	train-logloss:0.62646
[1679]	train-logloss:0.62636
[1680]	train-logloss:0.62612
[1681]	train-logloss:0.62655
[1682]	train-logloss:0.62635
[1683]	train-logloss:0.62605
[1684]	train-logloss:0.62646
[1685]	train-logloss:0.62708
[1686]	train-logloss:0.62742
[1687]	train-logloss:0.62785
[1688]	train-logloss:0.62789
[1689]	train-logloss:0.62822
[1690]	train-logloss:0.62799
[1691]	train-logloss:0.62868
[1692]	train-logloss:0.62901
[1693]	train-logloss:0.62901
[1694]	train-logloss:0.62914
[1695]	train-logloss:0.62889
[1696]	train-logloss:0.62889
[1697]	train-logloss:0.62943
[1698]	train-logloss:0.63000
[1699]	train-logloss:0.63004
[1700]	train-logloss:0.63026
[1701]	train-logloss:0.63075
[1702]	train-logloss:0.63076
[1703]	train-logloss:0.63148
[1704]	train-logloss:0.63152
[1705]	train-logloss:0.63151
[1706]	train-logloss:0.63170
[1707]	train-logloss:0.63178
[1708]	train-logloss:0.63160
[1709]	train-logloss:0.63154
[1710]	train-logloss:0.63216
[1711]	train-logloss:0.63176
[1712]	train-logloss:0.63144
[1713]	train-logloss:0.63144
[1714]	train-logloss:0.63135
[1715]	train-logloss:0.63146
[1716]	train-logloss:0.63145
[1717]	train-logloss:0.63156
[1718]	train-logloss:0.63085
[1719]	train-logloss:0.63143
[1720]	train-logloss:0.63115
[1721]	train-logloss:0.63196
[1722]	train-logloss:0.63176
[1723]	train-logloss:0.63173
[1724]	train-logloss:0.63226
[1725]	train-logloss:0.63247
[1726]	train-logloss:0.63249
[1727]	train-logloss:0.63195
[1728]	train-logloss:0.63201
[1729]	train-logloss:0.63176
[1730]	train-logloss:0.63183
[1731]	train-logloss:0.63172
[1732]	train-logloss:0.63126
[1733]	train-logloss:0.63168
[1734]	train-logloss:0.63187
[1735]	train-logloss:0.63190
[1736]	train-logloss:0.63153
[1737]	train-logloss:0.63155
[1738]	train-logloss:0.63142
[1739]	train-logloss:0.63193
[1740]	train-logloss:0.63273
[1741]	train-logloss:0.63286
[1742]	train-logloss:0.63302
[1743]	train-logloss:0.63290
[1744]	train-logloss:0.63289
[1745]	train-logloss:0.63304
[1746]	train-logloss:0.63250
[1747]	train-logloss:0.63258
[1748]	train-logloss:0.63193
[1749]	train-logloss:0.63185
[1750]	train-logloss:0.63220
[1751]	train-logloss:0.63252
[1752]	train-logloss:0.63256
[1753]	train-logloss:0.63246
[1754]	train-logloss:0.63227
[1755]	train-logloss:0.63293
[1756]	train-logloss:0.63271
[1757]	train-logloss:0.63324
[1758]	train-logloss:0.63333
[1759]	train-logloss:0.63328
[1760]	train-logloss:0.63319
[1761]	train-logloss:0.63320
[1762]	train-logloss:0.63344
[1763]	train-logloss:0.63367
[1764]	train-logloss:0.63343
[1765]	train-logloss:0.63387
[1766]	train-logloss:0.63410
[1767]	train-logloss:0.63494
[1768]	train-logloss:0.63479
[1769]	train-logloss:0.63492
[1770]	train-logloss:0.63518
[1771]	train-logloss:0.63433
[1772]	train-logloss:0.63369
[1773]	train-logloss:0.63367
[1774]	train-logloss:0.63371
[1775]	train-logloss:0.63405
[1776]	train-logloss:0.63410
[1777]	train-logloss:0.63479
[1778]	train-logloss:0.63420
[1779]	train-logloss:0.63421
[1780]	train-logloss:0.63344
[1781]	train-logloss:0.63337
[1782]	train-logloss:0.63343
[1783]	train-logloss:0.63341
[1784]	train-logloss:0.63357
[1785]	train-logloss:0.63359
[1786]	train-logloss:0.63375
[1787]	train-logloss:0.63367
[1788]	train-logloss:0.63314
[1789]	train-logloss:0.63308
[1790]	train-logloss:0.63310
[1791]	train-logloss:0.63399
[1792]	train-logloss:0.63392
[1793]	train-logloss:0.63406
[1794]	train-logloss:0.63405
[1795]	train-logloss:0.63456
[1796]	train-logloss:0.63486
[1797]	train-logloss:0.63499
[1798]	train-logloss:0.63507
[1799]	train-logloss:0.63509
[1800]	train-logloss:0.63491
[1801]	train-logloss:0.63487
[1802]	train-logloss:0.63536
[1803]	train-logloss:0.63584
[1804]	train-logloss:0.63591
[1805]	train-logloss:0.63588
[1806]	train-logloss:0.63546
[1807]	train-logloss:0.63529
[1808]	train-logloss:0.63565
[1809]	train-logloss:0.63558
[1810]	train-logloss:0.63572
[1811]	train-logloss:0.63561
[1812]	train-logloss:0.63598
[1813]	train-logloss:0.63634
[1814]	train-logloss:0.63634
[1815]	train-logloss:0.63663
[1816]	train-logloss:0.63615
[1817]	train-logloss:0.63646
[1818]	train-logloss:0.63635
[1819]	train-logloss:0.63620
[1820]	train-logloss:0.63593
[1821]	train-logloss:0.63538
[1822]	train-logloss:0.63517
[1823]	train-logloss:0.63479
[1824]	train-logloss:0.63480
[1825]	train-logloss:0.63417
[1826]	train-logloss:0.63417
[1827]	train-logloss:0.63357
[1828]	train-logloss:0.63291
[1829]	train-logloss:0.63237
[1830]	train-logloss:0.63229
[1831]	train-logloss:0.63239
[1832]	train-logloss:0.63236
[1833]	train-logloss:0.63249
[1834]	train-logloss:0.63265
[1835]	train-logloss:0.63263
[1836]	train-logloss:0.63264
[1837]	train-logloss:0.63256
[1838]	train-logloss:0.63256
[1839]	train-logloss:0.63252
[1840]	train-logloss:0.63264
[1841]	train-logloss:0.63257
[1842]	train-logloss:0.63266
[1843]	train-logloss:0.63267
[1844]	train-logloss:0.63223
[1845]	train-logloss:0.63223
[1846]	train-logloss:0.63218
[1847]	train-logloss:0.63234
[1848]	train-logloss:0.63234
[1849]	train-logloss:0.63235
[1850]	train-logloss:0.63175
[1851]	train-logloss:0.63204
[1852]	train-logloss:0.63210
[1853]	train-logloss:0.63177
[1854]	train-logloss:0.63243
[1855]	train-logloss:0.63226
[1856]	train-logloss:0.63271
[1857]	train-logloss:0.63206
[1858]	train-logloss:0.63206
[1859]	train-logloss:0.63191
[1860]	train-logloss:0.63220
[1861]	train-logloss:0.63236
[1862]	train-logloss:0.63214
[1863]	train-logloss:0.63248
[1864]	train-logloss:0.63216
[1865]	train-logloss:0.63245
[1866]	train-logloss:0.63247
[1867]	train-logloss:0.63262
[1868]	train-logloss:0.63261
[1869]	train-logloss:0.63266
[1870]	train-logloss:0.63278
[1871]	train-logloss:0.63256
[1872]	train-logloss:0.63322
[1873]	train-logloss:0.63320
[1874]	train-logloss:0.63290
[1875]	train-logloss:0.63291
[1876]	train-logloss:0.63290
[1877]	train-logloss:0.63275
[1878]	train-logloss:0.63277
[1879]	train-logloss:0.63280
[1880]	train-logloss:0.63254
[1881]	train-logloss:0.63225
[1882]	train-logloss:0.63286
[1883]	train-logloss:0.63271
[1884]	train-logloss:0.63270
[1885]	train-logloss:0.63268
[1886]	train-logloss:0.63268
[1887]	train-logloss:0.63276
[1888]	train-logloss:0.63250
[1889]	train-logloss:0.63276
[1890]	train-logloss:0.63270
[1891]	train-logloss:0.63247
[1892]	train-logloss:0.63222
[1893]	train-logloss:0.63252
[1894]	train-logloss:0.63280
[1895]	train-logloss:0.63284
[1896]	train-logloss:0.63253
[1897]	train-logloss:0.63241
[1898]	train-logloss:0.63218
[1899]	train-logloss:0.63219
[1900]	train-logloss:0.63192
[1901]	train-logloss:0.63223
[1902]	train-logloss:0.63201
[1903]	train-logloss:0.63173
[1904]	train-logloss:0.63202
[1905]	train-logloss:0.63222
[1906]	train-logloss:0.63181
[1907]	train-logloss:0.63178
[1908]	train-logloss:0.63213
[1909]	train-logloss:0.63178
[1910]	train-logloss:0.63225
[1911]	train-logloss:0.63274
[1912]	train-logloss:0.63294
[1913]	train-logloss:0.63338
[1914]	train-logloss:0.63338
[1915]	train-logloss:0.63338
[1916]	train-logloss:0.63341
[1917]	train-logloss:0.63340
[1918]	train-logloss:0.63349
[1919]	train-logloss:0.63310
[1920]	train-logloss:0.63315
[1921]	train-logloss:0.63328
[1922]	train-logloss:0.63319
[1923]	train-logloss:0.63287
[1924]	train-logloss:0.63251
[1925]	train-logloss:0.63272
[1926]	train-logloss:0.63240
[1927]	train-logloss:0.63280
[1928]	train-logloss:0.63241
[1929]	train-logloss:0.63241
[1930]	train-logloss:0.63241
[1931]	train-logloss:0.63229
[1932]	train-logloss:0.63205
[1933]	train-logloss:0.63170
[1934]	train-logloss:0.63269
[1935]	train-logloss:0.63312
[1936]	train-logloss:0.63253
[1937]	train-logloss:0.63222
[1938]	train-logloss:0.63223
[1939]	train-logloss:0.63224
[1940]	train-logloss:0.63252
[1941]	train-logloss:0.63260
[1942]	train-logloss:0.63329
[1943]	train-logloss:0.63331
[1944]	train-logloss:0.63432
[1945]	train-logloss:0.63457
[1946]	train-logloss:0.63454
[1947]	train-logloss:0.63421
[1948]	train-logloss:0.63418
[1949]	train-logloss:0.63412
[1950]	train-logloss:0.63373
[1951]	train-logloss:0.63307
[1952]	train-logloss:0.63306
[1953]	train-logloss:0.63307
[1954]	train-logloss:0.63296
[1955]	train-logloss:0.63289
[1956]	train-logloss:0.63286
[1957]	train-logloss:0.63286
[1958]	train-logloss:0.63286
[1959]	train-logloss:0.63268
[1960]	train-logloss:0.63289
[1961]	train-logloss:0.63299
[1962]	train-logloss:0.63288
[1963]	train-logloss:0.63288
[1964]	train-logloss:0.63280
[1965]	train-logloss:0.63254
[1966]	train-logloss:0.63272
[1967]	train-logloss:0.63287
[1968]	train-logloss:0.63327
[1969]	train-logloss:0.63324
[1970]	train-logloss:0.63324
[1971]	train-logloss:0.63336
[1972]	train-logloss:0.63382
[1973]	train-logloss:0.63386
[1974]	train-logloss:0.63427
[1975]	train-logloss:0.63428
[1976]	train-logloss:0.63462
[1977]	train-logloss:0.63443
[1978]	train-logloss:0.63445
[1979]	train-logloss:0.63453
[1980]	train-logloss:0.63466
[1981]	train-logloss:0.63527
[1982]	train-logloss:0.63546
[1983]	train-logloss:0.63513
[1984]	train-logloss:0.63484
[1985]	train-logloss:0.63482
[1986]	train-logloss:0.63484
[1987]	train-logloss:0.63513
[1988]	train-logloss:0.63536
[1989]	train-logloss:0.63516
[1990]	train-logloss:0.63468
[1991]	train-logloss:0.63452
[1992]	train-logloss:0.63448
[1993]	train-logloss:0.63460
[1994]	train-logloss:0.63451
[1995]	train-logloss:0.63422
[1996]	train-logloss:0.63409
[1997]	train-logloss:0.63412
[1998]	train-logloss:0.63406
[1999]	train-logloss:0.63402

step-04 用模型预测

ytrain=model.predict(dtrain)

注意:

  • 这里model.predict()预测得到的是概率值,而不是0或者1的结果
  • 下面将结果转换为0或者1
ytrain_class = (ytrain>= 0.5)*1
ytest=model.predict(dtest)
y_pred = (ytest >= 0.5)*1

step-05 评价模型效果

print(‘train_roc_auc_score:’,metrics.roc_auc_score(y_train,ytrain))
print(‘test_roc_auc_score:’,metrics.roc_auc_score(y_test, ytest))
print(‘train_accuracy_score:’,metrics.accuracy_score(y_train, ytrain_class))
print(‘test_accuracy_score:’,metrics.accuracy_score(y_test,y_pred ))

step-06 保存模型并调用

joblib.dump(model , r'D:\Ensemble_Learning\xgboostinfo\xgboost01.model')
load_model=joblib.load(r'D:\Ensemble_Learning\xgboostinfo\xgboost01.model')
ytest=load_model.predict(dtest)
ytest[0:5]
array([0.265046  , 0.39359182, 0.82298654, 0.07664716, 0.28468448],
      dtype=float32)

三. 网格搜索最优xgboost参数

1.step-01 配置参数列表

from sklearn.model_selection import GridSearchCV
## 定义参数取值范围
learning_rate = [0.1] #0.15,0.11
subsample = [ 0.65] #0.7,0.8
colsample_bytree = [0.6] #0.7, 0.5
colsample_bylevel=[0.7] #0.8,
colsample_bynode=[0.7] #0.8,
max_depth = [6] #,7
n_estimators=[1000] #,900
gamma=[0,0.1]
reg_alpha=[1,2]
reg_lambda=[2,3]
min_child_weight=[30,50]
max_bin=[12,16]
base_score=[0.4,0.5,0.6]

parameters = { 
              'learning_rate': learning_rate,
              'subsample': subsample,
              'colsample_bytree':colsample_bytree,
              'colsample_bylevel':colsample_bylevel,
              'colsample_bynode':colsample_bynode,
              'max_depth': max_depth,
              'n_estimators':n_estimators,
               'gamma':gamma,
               'reg_alpha':reg_alpha,
               'reg_lambda':reg_lambda,
               'min_child_weight':min_child_weight,
               'max_bin':max_bin,
               'base_score':base_score,
              }

step-02 选择待优化模型

model = XGBClassifier( eval_metric="logloss")

step-03 进行网格搜索 拟合模型

clf = GridSearchCV(model, parameters, cv=2, scoring='accuracy',verbose=1,n_jobs=-1)
clf = clf.fit(train.values, y_train.values,eval_set=eval_set)
Fitting 2 folds for each of 96 candidates, totalling 192 fits
[0]	validation_0-logloss:0.68822
[1]	validation_0-logloss:0.68488
[2]	validation_0-logloss:0.67979
[3]	validation_0-logloss:0.67770
[4]	validation_0-logloss:0.67431
[5]	validation_0-logloss:0.67095
[6]	validation_0-logloss:0.66894
[7]	validation_0-logloss:0.66736
[8]	validation_0-logloss:0.66269
[9]	validation_0-logloss:0.65911
[10]	validation_0-logloss:0.65691
[11]	validation_0-logloss:0.65429
[12]	validation_0-logloss:0.64994
[13]	validation_0-logloss:0.64843
[14]	validation_0-logloss:0.64748
[15]	validation_0-logloss:0.64628
[16]	validation_0-logloss:0.64424
[17]	validation_0-logloss:0.64260
[18]	validation_0-logloss:0.64172
[19]	validation_0-logloss:0.64020
[20]	validation_0-logloss:0.63933
[21]	validation_0-logloss:0.63795
[22]	validation_0-logloss:0.63296
[23]	validation_0-logloss:0.63192
[24]	validation_0-logloss:0.63157
[25]	validation_0-logloss:0.63006
[26]	validation_0-logloss:0.62925
[27]	validation_0-logloss:0.62915
[28]	validation_0-logloss:0.62914
[29]	validation_0-logloss:0.62940
[30]	validation_0-logloss:0.62872
[31]	validation_0-logloss:0.62866
[32]	validation_0-logloss:0.62860
[33]	validation_0-logloss:0.62812
[34]	validation_0-logloss:0.62823
[35]	validation_0-logloss:0.62819
[36]	validation_0-logloss:0.62489
[37]	validation_0-logloss:0.62490
[38]	validation_0-logloss:0.62293
[39]	validation_0-logloss:0.62222
[40]	validation_0-logloss:0.62102
[41]	validation_0-logloss:0.61937
[42]	validation_0-logloss:0.61839
[43]	validation_0-logloss:0.61829
[44]	validation_0-logloss:0.61782
[45]	validation_0-logloss:0.61781
[46]	validation_0-logloss:0.61763
[47]	validation_0-logloss:0.61733
[48]	validation_0-logloss:0.61704
[49]	validation_0-logloss:0.61602
[50]	validation_0-logloss:0.61585
[51]	validation_0-logloss:0.61632
[52]	validation_0-logloss:0.61601
[53]	validation_0-logloss:0.61658
[54]	validation_0-logloss:0.61598
[55]	validation_0-logloss:0.61581
[56]	validation_0-logloss:0.61530
[57]	validation_0-logloss:0.61455
[58]	validation_0-logloss:0.61557
[59]	validation_0-logloss:0.61533
[60]	validation_0-logloss:0.61390
[61]	validation_0-logloss:0.61426
[62]	validation_0-logloss:0.61365
[63]	validation_0-logloss:0.61269
[64]	validation_0-logloss:0.61244
[65]	validation_0-logloss:0.61196
[66]	validation_0-logloss:0.61196
[67]	validation_0-logloss:0.61175
[68]	validation_0-logloss:0.61179
[69]	validation_0-logloss:0.61195
[70]	validation_0-logloss:0.61165
[71]	validation_0-logloss:0.61130
[72]	validation_0-logloss:0.61112
[73]	validation_0-logloss:0.61133
[74]	validation_0-logloss:0.61152
[75]	validation_0-logloss:0.61118
[76]	validation_0-logloss:0.61160
[77]	validation_0-logloss:0.61167
[78]	validation_0-logloss:0.61175
[79]	validation_0-logloss:0.61156
[80]	validation_0-logloss:0.61164
[81]	validation_0-logloss:0.61126
[82]	validation_0-logloss:0.61166
[83]	validation_0-logloss:0.61163
[84]	validation_0-logloss:0.61156
[85]	validation_0-logloss:0.61177
[86]	validation_0-logloss:0.61271
[87]	validation_0-logloss:0.61074
[88]	validation_0-logloss:0.61048
[89]	validation_0-logloss:0.60983
[90]	validation_0-logloss:0.60992
[91]	validation_0-logloss:0.60904
[92]	validation_0-logloss:0.60858
[93]	validation_0-logloss:0.60805
[94]	validation_0-logloss:0.60787
[95]	validation_0-logloss:0.60836
[96]	validation_0-logloss:0.60857
[97]	validation_0-logloss:0.60862
[98]	validation_0-logloss:0.60874
[99]	validation_0-logloss:0.60815
[100]	validation_0-logloss:0.60815
[101]	validation_0-logloss:0.60762
[102]	validation_0-logloss:0.60721
[103]	validation_0-logloss:0.60722
[104]	validation_0-logloss:0.60713
[105]	validation_0-logloss:0.60712
[106]	validation_0-logloss:0.60659
[107]	validation_0-logloss:0.60623
[108]	validation_0-logloss:0.60603
[109]	validation_0-logloss:0.60549
[110]	validation_0-logloss:0.60546
[111]	validation_0-logloss:0.60535
[112]	validation_0-logloss:0.60451
[113]	validation_0-logloss:0.60451
[114]	validation_0-logloss:0.60397
[115]	validation_0-logloss:0.60426
[116]	validation_0-logloss:0.60452
[117]	validation_0-logloss:0.60424
[118]	validation_0-logloss:0.60428
[119]	validation_0-logloss:0.60379
[120]	validation_0-logloss:0.60408
[121]	validation_0-logloss:0.60420
[122]	validation_0-logloss:0.60399
[123]	validation_0-logloss:0.60389
[124]	validation_0-logloss:0.60441
[125]	validation_0-logloss:0.60494
[126]	validation_0-logloss:0.60457
[127]	validation_0-logloss:0.60444
[128]	validation_0-logloss:0.60442
[129]	validation_0-logloss:0.60438
[130]	validation_0-logloss:0.60436
[131]	validation_0-logloss:0.60378
[132]	validation_0-logloss:0.60310
[133]	validation_0-logloss:0.60328
[134]	validation_0-logloss:0.60349
[135]	validation_0-logloss:0.60336
[136]	validation_0-logloss:0.60355
[137]	validation_0-logloss:0.60356
[138]	validation_0-logloss:0.60385
[139]	validation_0-logloss:0.60383
[140]	validation_0-logloss:0.60363
[141]	validation_0-logloss:0.60288
[142]	validation_0-logloss:0.60319
[143]	validation_0-logloss:0.60344
[144]	validation_0-logloss:0.60350
[145]	validation_0-logloss:0.60393
[146]	validation_0-logloss:0.60399
[147]	validation_0-logloss:0.60408
[148]	validation_0-logloss:0.60428
[149]	validation_0-logloss:0.60439
[150]	validation_0-logloss:0.60444
[151]	validation_0-logloss:0.60460
[152]	validation_0-logloss:0.60519
[153]	validation_0-logloss:0.60553
[154]	validation_0-logloss:0.60516
[155]	validation_0-logloss:0.60552
[156]	validation_0-logloss:0.60554
[157]	validation_0-logloss:0.60521
[158]	validation_0-logloss:0.60540
[159]	validation_0-logloss:0.60549
[160]	validation_0-logloss:0.60561
[161]	validation_0-logloss:0.60576
[162]	validation_0-logloss:0.60609
[163]	validation_0-logloss:0.60591
[164]	validation_0-logloss:0.60582
[165]	validation_0-logloss:0.60576
[166]	validation_0-logloss:0.60607
[167]	validation_0-logloss:0.60569
[168]	validation_0-logloss:0.60565
[169]	validation_0-logloss:0.60612
[170]	validation_0-logloss:0.60641
[171]	validation_0-logloss:0.60640
[172]	validation_0-logloss:0.60609
[173]	validation_0-logloss:0.60584
[174]	validation_0-logloss:0.60604
[175]	validation_0-logloss:0.60608
[176]	validation_0-logloss:0.60609
[177]	validation_0-logloss:0.60606
[178]	validation_0-logloss:0.60660
[179]	validation_0-logloss:0.60601
[180]	validation_0-logloss:0.60543
[181]	validation_0-logloss:0.60482
[182]	validation_0-logloss:0.60460
[183]	validation_0-logloss:0.60465
[184]	validation_0-logloss:0.60453
[185]	validation_0-logloss:0.60450
[186]	validation_0-logloss:0.60447
[187]	validation_0-logloss:0.60442
[188]	validation_0-logloss:0.60432
[189]	validation_0-logloss:0.60451
[190]	validation_0-logloss:0.60469
[191]	validation_0-logloss:0.60473
[192]	validation_0-logloss:0.60455
[193]	validation_0-logloss:0.60426
[194]	validation_0-logloss:0.60474
[195]	validation_0-logloss:0.60463
[196]	validation_0-logloss:0.60473
[197]	validation_0-logloss:0.60477
[198]	validation_0-logloss:0.60532
[199]	validation_0-logloss:0.60515
[200]	validation_0-logloss:0.60518
[201]	validation_0-logloss:0.60515
[202]	validation_0-logloss:0.60500
[203]	validation_0-logloss:0.60524
[204]	validation_0-logloss:0.60522
[205]	validation_0-logloss:0.60516
[206]	validation_0-logloss:0.60473
[207]	validation_0-logloss:0.60459
[208]	validation_0-logloss:0.60468
[209]	validation_0-logloss:0.60497
[210]	validation_0-logloss:0.60538
[211]	validation_0-logloss:0.60584
[212]	validation_0-logloss:0.60534
[213]	validation_0-logloss:0.60530
[214]	validation_0-logloss:0.60557
[215]	validation_0-logloss:0.60565
[216]	validation_0-logloss:0.60610
[217]	validation_0-logloss:0.60636
[218]	validation_0-logloss:0.60650
[219]	validation_0-logloss:0.60661
[220]	validation_0-logloss:0.60655
[221]	validation_0-logloss:0.60701
[222]	validation_0-logloss:0.60714
[223]	validation_0-logloss:0.60700
[224]	validation_0-logloss:0.60750
[225]	validation_0-logloss:0.60757
[226]	validation_0-logloss:0.60762
[227]	validation_0-logloss:0.60722
[228]	validation_0-logloss:0.60706
[229]	validation_0-logloss:0.60686
[230]	validation_0-logloss:0.60654
[231]	validation_0-logloss:0.60657
[232]	validation_0-logloss:0.60676
[233]	validation_0-logloss:0.60664
[234]	validation_0-logloss:0.60668
[235]	validation_0-logloss:0.60694
[236]	validation_0-logloss:0.60680
[237]	validation_0-logloss:0.60677
[238]	validation_0-logloss:0.60649
[239]	validation_0-logloss:0.60630
[240]	validation_0-logloss:0.60609
[241]	validation_0-logloss:0.60574
[242]	validation_0-logloss:0.60603
[243]	validation_0-logloss:0.60609
[244]	validation_0-logloss:0.60588
[245]	validation_0-logloss:0.60599
[246]	validation_0-logloss:0.60576
[247]	validation_0-logloss:0.60621
[248]	validation_0-logloss:0.60669
[249]	validation_0-logloss:0.60657
[250]	validation_0-logloss:0.60696
[251]	validation_0-logloss:0.60693
[252]	validation_0-logloss:0.60653
[253]	validation_0-logloss:0.60678
[254]	validation_0-logloss:0.60658
[255]	validation_0-logloss:0.60608
[256]	validation_0-logloss:0.60590
[257]	validation_0-logloss:0.60587
[258]	validation_0-logloss:0.60539
[259]	validation_0-logloss:0.60528
[260]	validation_0-logloss:0.60510
[261]	validation_0-logloss:0.60560
[262]	validation_0-logloss:0.60583
[263]	validation_0-logloss:0.60592
[264]	validation_0-logloss:0.60591
[265]	validation_0-logloss:0.60541
[266]	validation_0-logloss:0.60535
[267]	validation_0-logloss:0.60566
[268]	validation_0-logloss:0.60543
[269]	validation_0-logloss:0.60562
[270]	validation_0-logloss:0.60554
[271]	validation_0-logloss:0.60535
[272]	validation_0-logloss:0.60563
[273]	validation_0-logloss:0.60566
[274]	validation_0-logloss:0.60529
[275]	validation_0-logloss:0.60534
[276]	validation_0-logloss:0.60551
[277]	validation_0-logloss:0.60549
[278]	validation_0-logloss:0.60546
[279]	validation_0-logloss:0.60526
[280]	validation_0-logloss:0.60515
[281]	validation_0-logloss:0.60527
[282]	validation_0-logloss:0.60511
[283]	validation_0-logloss:0.60428
[284]	validation_0-logloss:0.60414
[285]	validation_0-logloss:0.60400
[286]	validation_0-logloss:0.60428
[287]	validation_0-logloss:0.60393
[288]	validation_0-logloss:0.60395
[289]	validation_0-logloss:0.60418
[290]	validation_0-logloss:0.60400
[291]	validation_0-logloss:0.60397
[292]	validation_0-logloss:0.60400
[293]	validation_0-logloss:0.60457
[294]	validation_0-logloss:0.60491
[295]	validation_0-logloss:0.60482
[296]	validation_0-logloss:0.60503
[297]	validation_0-logloss:0.60526
[298]	validation_0-logloss:0.60520
[299]	validation_0-logloss:0.60509
[300]	validation_0-logloss:0.60484
[301]	validation_0-logloss:0.60457
[302]	validation_0-logloss:0.60474
[303]	validation_0-logloss:0.60462
[304]	validation_0-logloss:0.60472
[305]	validation_0-logloss:0.60515
[306]	validation_0-logloss:0.60481
[307]	validation_0-logloss:0.60471
[308]	validation_0-logloss:0.60469
[309]	validation_0-logloss:0.60460
[310]	validation_0-logloss:0.60466
[311]	validation_0-logloss:0.60474
[312]	validation_0-logloss:0.60487
[313]	validation_0-logloss:0.60508
[314]	validation_0-logloss:0.60515
[315]	validation_0-logloss:0.60525
[316]	validation_0-logloss:0.60464
[317]	validation_0-logloss:0.60475
[318]	validation_0-logloss:0.60480
[319]	validation_0-logloss:0.60429
[320]	validation_0-logloss:0.60425
[321]	validation_0-logloss:0.60446
[322]	validation_0-logloss:0.60442
[323]	validation_0-logloss:0.60446
[324]	validation_0-logloss:0.60472
[325]	validation_0-logloss:0.60480
[326]	validation_0-logloss:0.60463
[327]	validation_0-logloss:0.60456
[328]	validation_0-logloss:0.60465
[329]	validation_0-logloss:0.60469
[330]	validation_0-logloss:0.60477
[331]	validation_0-logloss:0.60517
[332]	validation_0-logloss:0.60530
[333]	validation_0-logloss:0.60528
[334]	validation_0-logloss:0.60485
[335]	validation_0-logloss:0.60464
[336]	validation_0-logloss:0.60450
[337]	validation_0-logloss:0.60485
[338]	validation_0-logloss:0.60507
[339]	validation_0-logloss:0.60503
[340]	validation_0-logloss:0.60486
[341]	validation_0-logloss:0.60507
[342]	validation_0-logloss:0.60502
[343]	validation_0-logloss:0.60454
[344]	validation_0-logloss:0.60476
[345]	validation_0-logloss:0.60511
[346]	validation_0-logloss:0.60532
[347]	validation_0-logloss:0.60501
[348]	validation_0-logloss:0.60510
[349]	validation_0-logloss:0.60524
[350]	validation_0-logloss:0.60553
[351]	validation_0-logloss:0.60552
[352]	validation_0-logloss:0.60485
[353]	validation_0-logloss:0.60502
[354]	validation_0-logloss:0.60475
[355]	validation_0-logloss:0.60484
[356]	validation_0-logloss:0.60499
[357]	validation_0-logloss:0.60494
[358]	validation_0-logloss:0.60474
[359]	validation_0-logloss:0.60461
[360]	validation_0-logloss:0.60477
[361]	validation_0-logloss:0.60355
[362]	validation_0-logloss:0.60340
[363]	validation_0-logloss:0.60368
[364]	validation_0-logloss:0.60373
[365]	validation_0-logloss:0.60382
[366]	validation_0-logloss:0.60382
[367]	validation_0-logloss:0.60366
[368]	validation_0-logloss:0.60367
[369]	validation_0-logloss:0.60350
[370]	validation_0-logloss:0.60348
[371]	validation_0-logloss:0.60336
[372]	validation_0-logloss:0.60300
[373]	validation_0-logloss:0.60334
[374]	validation_0-logloss:0.60330
[375]	validation_0-logloss:0.60371
[376]	validation_0-logloss:0.60409
[377]	validation_0-logloss:0.60424
[378]	validation_0-logloss:0.60393
[379]	validation_0-logloss:0.60401
[380]	validation_0-logloss:0.60403
[381]	validation_0-logloss:0.60395
[382]	validation_0-logloss:0.60366
[383]	validation_0-logloss:0.60358
[384]	validation_0-logloss:0.60356
[385]	validation_0-logloss:0.60394
[386]	validation_0-logloss:0.60367
[387]	validation_0-logloss:0.60399
[388]	validation_0-logloss:0.60392
[389]	validation_0-logloss:0.60449
[390]	validation_0-logloss:0.60467
[391]	validation_0-logloss:0.60516
[392]	validation_0-logloss:0.60514
[393]	validation_0-logloss:0.60507
[394]	validation_0-logloss:0.60519
[395]	validation_0-logloss:0.60530
[396]	validation_0-logloss:0.60509
[397]	validation_0-logloss:0.60484
[398]	validation_0-logloss:0.60473
[399]	validation_0-logloss:0.60446
[400]	validation_0-logloss:0.60440
[401]	validation_0-logloss:0.60455
[402]	validation_0-logloss:0.60452
[403]	validation_0-logloss:0.60424
[404]	validation_0-logloss:0.60409
[405]	validation_0-logloss:0.60405
[406]	validation_0-logloss:0.60397
[407]	validation_0-logloss:0.60402
[408]	validation_0-logloss:0.60391
[409]	validation_0-logloss:0.60378
[410]	validation_0-logloss:0.60382
[411]	validation_0-logloss:0.60386
[412]	validation_0-logloss:0.60359
[413]	validation_0-logloss:0.60344
[414]	validation_0-logloss:0.60370
[415]	validation_0-logloss:0.60382
[416]	validation_0-logloss:0.60394
[417]	validation_0-logloss:0.60401
[418]	validation_0-logloss:0.60385
[419]	validation_0-logloss:0.60374
[420]	validation_0-logloss:0.60382
[421]	validation_0-logloss:0.60395
[422]	validation_0-logloss:0.60394
[423]	validation_0-logloss:0.60395
[424]	validation_0-logloss:0.60385
[425]	validation_0-logloss:0.60374
[426]	validation_0-logloss:0.60343
[427]	validation_0-logloss:0.60384
[428]	validation_0-logloss:0.60435
[429]	validation_0-logloss:0.60471
[430]	validation_0-logloss:0.60426
[431]	validation_0-logloss:0.60393
[432]	validation_0-logloss:0.60411
[433]	validation_0-logloss:0.60418
[434]	validation_0-logloss:0.60446
[435]	validation_0-logloss:0.60360
[436]	validation_0-logloss:0.60333
[437]	validation_0-logloss:0.60326
[438]	validation_0-logloss:0.60335
[439]	validation_0-logloss:0.60329
[440]	validation_0-logloss:0.60312
[441]	validation_0-logloss:0.60343
[442]	validation_0-logloss:0.60387
[443]	validation_0-logloss:0.60386
[444]	validation_0-logloss:0.60377
[445]	validation_0-logloss:0.60369
[446]	validation_0-logloss:0.60395
[447]	validation_0-logloss:0.60427
[448]	validation_0-logloss:0.60443
[449]	validation_0-logloss:0.60459
[450]	validation_0-logloss:0.60452
[451]	validation_0-logloss:0.60487
[452]	validation_0-logloss:0.60499
[453]	validation_0-logloss:0.60422
[454]	validation_0-logloss:0.60429
[455]	validation_0-logloss:0.60423
[456]	validation_0-logloss:0.60457
[457]	validation_0-logloss:0.60458
[458]	validation_0-logloss:0.60459
[459]	validation_0-logloss:0.60461
[460]	validation_0-logloss:0.60487
[461]	validation_0-logloss:0.60523
[462]	validation_0-logloss:0.60522
[463]	validation_0-logloss:0.60511
[464]	validation_0-logloss:0.60496
[465]	validation_0-logloss:0.60522
[466]	validation_0-logloss:0.60537
[467]	validation_0-logloss:0.60529
[468]	validation_0-logloss:0.60488
[469]	validation_0-logloss:0.60495
[470]	validation_0-logloss:0.60476
[471]	validation_0-logloss:0.60436
[472]	validation_0-logloss:0.60453
[473]	validation_0-logloss:0.60423
[474]	validation_0-logloss:0.60389
[475]	validation_0-logloss:0.60389
[476]	validation_0-logloss:0.60365
[477]	validation_0-logloss:0.60376
[478]	validation_0-logloss:0.60377
[479]	validation_0-logloss:0.60350
[480]	validation_0-logloss:0.60341
[481]	validation_0-logloss:0.60335
[482]	validation_0-logloss:0.60350
[483]	validation_0-logloss:0.60303
[484]	validation_0-logloss:0.60329
[485]	validation_0-logloss:0.60326
[486]	validation_0-logloss:0.60336
[487]	validation_0-logloss:0.60346
[488]	validation_0-logloss:0.60365
[489]	validation_0-logloss:0.60350
[490]	validation_0-logloss:0.60350
[491]	validation_0-logloss:0.60373
[492]	validation_0-logloss:0.60363
[493]	validation_0-logloss:0.60397
[494]	validation_0-logloss:0.60403
[495]	validation_0-logloss:0.60371
[496]	validation_0-logloss:0.60382
[497]	validation_0-logloss:0.60378
[498]	validation_0-logloss:0.60390
[499]	validation_0-logloss:0.60406
[500]	validation_0-logloss:0.60411
[501]	validation_0-logloss:0.60401
[502]	validation_0-logloss:0.60416
[503]	validation_0-logloss:0.60469
[504]	validation_0-logloss:0.60466
[505]	validation_0-logloss:0.60460
[506]	validation_0-logloss:0.60480
[507]	validation_0-logloss:0.60445
[508]	validation_0-logloss:0.60471
[509]	validation_0-logloss:0.60446
[510]	validation_0-logloss:0.60447
[511]	validation_0-logloss:0.60452
[512]	validation_0-logloss:0.60432
[513]	validation_0-logloss:0.60395
[514]	validation_0-logloss:0.60411
[515]	validation_0-logloss:0.60397
[516]	validation_0-logloss:0.60418
[517]	validation_0-logloss:0.60432
[518]	validation_0-logloss:0.60424
[519]	validation_0-logloss:0.60419
[520]	validation_0-logloss:0.60442
[521]	validation_0-logloss:0.60408
[522]	validation_0-logloss:0.60413
[523]	validation_0-logloss:0.60399
[524]	validation_0-logloss:0.60416
[525]	validation_0-logloss:0.60426
[526]	validation_0-logloss:0.60448
[527]	validation_0-logloss:0.60472
[528]	validation_0-logloss:0.60455
[529]	validation_0-logloss:0.60461
[530]	validation_0-logloss:0.60446
[531]	validation_0-logloss:0.60432
[532]	validation_0-logloss:0.60416
[533]	validation_0-logloss:0.60405
[534]	validation_0-logloss:0.60423
[535]	validation_0-logloss:0.60428
[536]	validation_0-logloss:0.60378
[537]	validation_0-logloss:0.60372
[538]	validation_0-logloss:0.60382
[539]	validation_0-logloss:0.60379
[540]	validation_0-logloss:0.60388
[541]	validation_0-logloss:0.60372
[542]	validation_0-logloss:0.60382
[543]	validation_0-logloss:0.60378
[544]	validation_0-logloss:0.60367
[545]	validation_0-logloss:0.60397
[546]	validation_0-logloss:0.60379
[547]	validation_0-logloss:0.60401
[548]	validation_0-logloss:0.60416
[549]	validation_0-logloss:0.60442
[550]	validation_0-logloss:0.60443
[551]	validation_0-logloss:0.60432
[552]	validation_0-logloss:0.60414
[553]	validation_0-logloss:0.60427
[554]	validation_0-logloss:0.60457
[555]	validation_0-logloss:0.60423
[556]	validation_0-logloss:0.60474
[557]	validation_0-logloss:0.60459
[558]	validation_0-logloss:0.60463
[559]	validation_0-logloss:0.60445
[560]	validation_0-logloss:0.60412
[561]	validation_0-logloss:0.60404
[562]	validation_0-logloss:0.60418
[563]	validation_0-logloss:0.60409
[564]	validation_0-logloss:0.60425
[565]	validation_0-logloss:0.60470
[566]	validation_0-logloss:0.60461
[567]	validation_0-logloss:0.60490
[568]	validation_0-logloss:0.60464
[569]	validation_0-logloss:0.60456
[570]	validation_0-logloss:0.60474
[571]	validation_0-logloss:0.60472
[572]	validation_0-logloss:0.60466
[573]	validation_0-logloss:0.60453
[574]	validation_0-logloss:0.60497
[575]	validation_0-logloss:0.60498
[576]	validation_0-logloss:0.60512
[577]	validation_0-logloss:0.60532
[578]	validation_0-logloss:0.60528
[579]	validation_0-logloss:0.60516
[580]	validation_0-logloss:0.60537
[581]	validation_0-logloss:0.60552
[582]	validation_0-logloss:0.60537
[583]	validation_0-logloss:0.60543
[584]	validation_0-logloss:0.60534
[585]	validation_0-logloss:0.60534
[586]	validation_0-logloss:0.60523
[587]	validation_0-logloss:0.60507
[588]	validation_0-logloss:0.60517
[589]	validation_0-logloss:0.60532
[590]	validation_0-logloss:0.60511
[591]	validation_0-logloss:0.60522
[592]	validation_0-logloss:0.60522
[593]	validation_0-logloss:0.60500
[594]	validation_0-logloss:0.60504
[595]	validation_0-logloss:0.60453
[596]	validation_0-logloss:0.60472
[597]	validation_0-logloss:0.60476
[598]	validation_0-logloss:0.60454
[599]	validation_0-logloss:0.60482
[600]	validation_0-logloss:0.60493
[601]	validation_0-logloss:0.60508
[602]	validation_0-logloss:0.60498
[603]	validation_0-logloss:0.60468
[604]	validation_0-logloss:0.60489
[605]	validation_0-logloss:0.60471
[606]	validation_0-logloss:0.60445
[607]	validation_0-logloss:0.60449
[608]	validation_0-logloss:0.60416
[609]	validation_0-logloss:0.60470
[610]	validation_0-logloss:0.60475
[611]	validation_0-logloss:0.60463
[612]	validation_0-logloss:0.60459
[613]	validation_0-logloss:0.60463
[614]	validation_0-logloss:0.60483
[615]	validation_0-logloss:0.60463
[616]	validation_0-logloss:0.60455
[617]	validation_0-logloss:0.60469
[618]	validation_0-logloss:0.60512
[619]	validation_0-logloss:0.60497
[620]	validation_0-logloss:0.60498
[621]	validation_0-logloss:0.60506
[622]	validation_0-logloss:0.60505
[623]	validation_0-logloss:0.60511
[624]	validation_0-logloss:0.60516
[625]	validation_0-logloss:0.60471
[626]	validation_0-logloss:0.60465
[627]	validation_0-logloss:0.60462
[628]	validation_0-logloss:0.60465
[629]	validation_0-logloss:0.60461
[630]	validation_0-logloss:0.60509
[631]	validation_0-logloss:0.60494
[632]	validation_0-logloss:0.60538
[633]	validation_0-logloss:0.60578
[634]	validation_0-logloss:0.60573
[635]	validation_0-logloss:0.60580
[636]	validation_0-logloss:0.60596
[637]	validation_0-logloss:0.60593
[638]	validation_0-logloss:0.60586
[639]	validation_0-logloss:0.60597
[640]	validation_0-logloss:0.60609
[641]	validation_0-logloss:0.60606
[642]	validation_0-logloss:0.60550
[643]	validation_0-logloss:0.60544
[644]	validation_0-logloss:0.60542
[645]	validation_0-logloss:0.60576
[646]	validation_0-logloss:0.60561
[647]	validation_0-logloss:0.60587
[648]	validation_0-logloss:0.60584
[649]	validation_0-logloss:0.60494
[650]	validation_0-logloss:0.60505
[651]	validation_0-logloss:0.60494
[652]	validation_0-logloss:0.60488
[653]	validation_0-logloss:0.60494
[654]	validation_0-logloss:0.60439
[655]	validation_0-logloss:0.60448
[656]	validation_0-logloss:0.60448
[657]	validation_0-logloss:0.60455
[658]	validation_0-logloss:0.60459
[659]	validation_0-logloss:0.60436
[660]	validation_0-logloss:0.60424
[661]	validation_0-logloss:0.60412
[662]	validation_0-logloss:0.60409
[663]	validation_0-logloss:0.60410
[664]	validation_0-logloss:0.60421
[665]	validation_0-logloss:0.60425
[666]	validation_0-logloss:0.60453
[667]	validation_0-logloss:0.60444
[668]	validation_0-logloss:0.60434
[669]	validation_0-logloss:0.60442
[670]	validation_0-logloss:0.60437
[671]	validation_0-logloss:0.60456
[672]	validation_0-logloss:0.60458
[673]	validation_0-logloss:0.60443
[674]	validation_0-logloss:0.60407
[675]	validation_0-logloss:0.60402
[676]	validation_0-logloss:0.60406
[677]	validation_0-logloss:0.60406
[678]	validation_0-logloss:0.60412
[679]	validation_0-logloss:0.60435
[680]	validation_0-logloss:0.60433
[681]	validation_0-logloss:0.60408
[682]	validation_0-logloss:0.60389
[683]	validation_0-logloss:0.60368
[684]	validation_0-logloss:0.60364
[685]	validation_0-logloss:0.60370
[686]	validation_0-logloss:0.60360
[687]	validation_0-logloss:0.60370
[688]	validation_0-logloss:0.60363
[689]	validation_0-logloss:0.60367
[690]	validation_0-logloss:0.60391
[691]	validation_0-logloss:0.60374
[692]	validation_0-logloss:0.60393
[693]	validation_0-logloss:0.60394
[694]	validation_0-logloss:0.60422
[695]	validation_0-logloss:0.60424
[696]	validation_0-logloss:0.60417
[697]	validation_0-logloss:0.60411
[698]	validation_0-logloss:0.60426
[699]	validation_0-logloss:0.60473
[700]	validation_0-logloss:0.60487
[701]	validation_0-logloss:0.60560
[702]	validation_0-logloss:0.60577
[703]	validation_0-logloss:0.60570
[704]	validation_0-logloss:0.60535
[705]	validation_0-logloss:0.60524
[706]	validation_0-logloss:0.60532
[707]	validation_0-logloss:0.60555
[708]	validation_0-logloss:0.60548
[709]	validation_0-logloss:0.60556
[710]	validation_0-logloss:0.60569
[711]	validation_0-logloss:0.60592
[712]	validation_0-logloss:0.60615
[713]	validation_0-logloss:0.60617
[714]	validation_0-logloss:0.60631
[715]	validation_0-logloss:0.60655
[716]	validation_0-logloss:0.60684
[717]	validation_0-logloss:0.60676
[718]	validation_0-logloss:0.60646
[719]	validation_0-logloss:0.60614
[720]	validation_0-logloss:0.60583
[721]	validation_0-logloss:0.60571
[722]	validation_0-logloss:0.60550
[723]	validation_0-logloss:0.60545
[724]	validation_0-logloss:0.60471
[725]	validation_0-logloss:0.60475
[726]	validation_0-logloss:0.60462
[727]	validation_0-logloss:0.60456
[728]	validation_0-logloss:0.60422
[729]	validation_0-logloss:0.60413
[730]	validation_0-logloss:0.60415
[731]	validation_0-logloss:0.60436
[732]	validation_0-logloss:0.60453
[733]	validation_0-logloss:0.60435
[734]	validation_0-logloss:0.60413
[735]	validation_0-logloss:0.60428
[736]	validation_0-logloss:0.60421
[737]	validation_0-logloss:0.60376
[738]	validation_0-logloss:0.60376
[739]	validation_0-logloss:0.60379
[740]	validation_0-logloss:0.60400
[741]	validation_0-logloss:0.60416
[742]	validation_0-logloss:0.60410
[743]	validation_0-logloss:0.60400
[744]	validation_0-logloss:0.60408
[745]	validation_0-logloss:0.60419
[746]	validation_0-logloss:0.60411
[747]	validation_0-logloss:0.60401
[748]	validation_0-logloss:0.60395
[749]	validation_0-logloss:0.60409
[750]	validation_0-logloss:0.60397
[751]	validation_0-logloss:0.60388
[752]	validation_0-logloss:0.60448
[753]	validation_0-logloss:0.60439
[754]	validation_0-logloss:0.60436
[755]	validation_0-logloss:0.60419
[756]	validation_0-logloss:0.60411
[757]	validation_0-logloss:0.60439
[758]	validation_0-logloss:0.60456
[759]	validation_0-logloss:0.60472
[760]	validation_0-logloss:0.60418
[761]	validation_0-logloss:0.60395
[762]	validation_0-logloss:0.60395
[763]	validation_0-logloss:0.60384
[764]	validation_0-logloss:0.60380
[765]	validation_0-logloss:0.60412
[766]	validation_0-logloss:0.60415
[767]	validation_0-logloss:0.60427
[768]	validation_0-logloss:0.60411
[769]	validation_0-logloss:0.60426
[770]	validation_0-logloss:0.60430
[771]	validation_0-logloss:0.60455
[772]	validation_0-logloss:0.60482
[773]	validation_0-logloss:0.60490
[774]	validation_0-logloss:0.60482
[775]	validation_0-logloss:0.60506
[776]	validation_0-logloss:0.60499
[777]	validation_0-logloss:0.60479
[778]	validation_0-logloss:0.60462
[779]	validation_0-logloss:0.60462
[780]	validation_0-logloss:0.60461
[781]	validation_0-logloss:0.60505
[782]	validation_0-logloss:0.60512
[783]	validation_0-logloss:0.60534
[784]	validation_0-logloss:0.60552
[785]	validation_0-logloss:0.60558
[786]	validation_0-logloss:0.60575
[787]	validation_0-logloss:0.60570
[788]	validation_0-logloss:0.60578
[789]	validation_0-logloss:0.60564
[790]	validation_0-logloss:0.60568
[791]	validation_0-logloss:0.60587
[792]	validation_0-logloss:0.60602
[793]	validation_0-logloss:0.60574
[794]	validation_0-logloss:0.60576
[795]	validation_0-logloss:0.60569
[796]	validation_0-logloss:0.60569
[797]	validation_0-logloss:0.60633
[798]	validation_0-logloss:0.60678
[799]	validation_0-logloss:0.60706
[800]	validation_0-logloss:0.60701
[801]	validation_0-logloss:0.60686
[802]	validation_0-logloss:0.60681
[803]	validation_0-logloss:0.60680
[804]	validation_0-logloss:0.60670
[805]	validation_0-logloss:0.60700
[806]	validation_0-logloss:0.60709
[807]	validation_0-logloss:0.60697
[808]	validation_0-logloss:0.60676
[809]	validation_0-logloss:0.60660
[810]	validation_0-logloss:0.60628
[811]	validation_0-logloss:0.60646
[812]	validation_0-logloss:0.60627
[813]	validation_0-logloss:0.60681
[814]	validation_0-logloss:0.60678
[815]	validation_0-logloss:0.60702
[816]	validation_0-logloss:0.60666
[817]	validation_0-logloss:0.60681
[818]	validation_0-logloss:0.60716
[819]	validation_0-logloss:0.60757
[820]	validation_0-logloss:0.60738
[821]	validation_0-logloss:0.60758
[822]	validation_0-logloss:0.60761
[823]	validation_0-logloss:0.60766
[824]	validation_0-logloss:0.60746
[825]	validation_0-logloss:0.60728
[826]	validation_0-logloss:0.60736
[827]	validation_0-logloss:0.60739
[828]	validation_0-logloss:0.60743
[829]	validation_0-logloss:0.60748
[830]	validation_0-logloss:0.60727
[831]	validation_0-logloss:0.60745
[832]	validation_0-logloss:0.60717
[833]	validation_0-logloss:0.60697
[834]	validation_0-logloss:0.60676
[835]	validation_0-logloss:0.60640
[836]	validation_0-logloss:0.60708
[837]	validation_0-logloss:0.60744
[838]	validation_0-logloss:0.60775
[839]	validation_0-logloss:0.60798
[840]	validation_0-logloss:0.60808
[841]	validation_0-logloss:0.60765
[842]	validation_0-logloss:0.60776
[843]	validation_0-logloss:0.60782
[844]	validation_0-logloss:0.60783
[845]	validation_0-logloss:0.60776
[846]	validation_0-logloss:0.60800
[847]	validation_0-logloss:0.60782
[848]	validation_0-logloss:0.60815
[849]	validation_0-logloss:0.60799
[850]	validation_0-logloss:0.60784
[851]	validation_0-logloss:0.60796
[852]	validation_0-logloss:0.60805
[853]	validation_0-logloss:0.60803
[854]	validation_0-logloss:0.60794
[855]	validation_0-logloss:0.60811
[856]	validation_0-logloss:0.60789
[857]	validation_0-logloss:0.60779
[858]	validation_0-logloss:0.60777
[859]	validation_0-logloss:0.60769
[860]	validation_0-logloss:0.60778
[861]	validation_0-logloss:0.60765
[862]	validation_0-logloss:0.60734
[863]	validation_0-logloss:0.60729
[864]	validation_0-logloss:0.60720
[865]	validation_0-logloss:0.60696
[866]	validation_0-logloss:0.60701
[867]	validation_0-logloss:0.60726
[868]	validation_0-logloss:0.60718
[869]	validation_0-logloss:0.60698
[870]	validation_0-logloss:0.60683
[871]	validation_0-logloss:0.60689
[872]	validation_0-logloss:0.60708
[873]	validation_0-logloss:0.60722
[874]	validation_0-logloss:0.60703
[875]	validation_0-logloss:0.60677
[876]	validation_0-logloss:0.60664
[877]	validation_0-logloss:0.60656
[878]	validation_0-logloss:0.60645
[879]	validation_0-logloss:0.60644
[880]	validation_0-logloss:0.60642
[881]	validation_0-logloss:0.60628
[882]	validation_0-logloss:0.60623
[883]	validation_0-logloss:0.60586
[884]	validation_0-logloss:0.60563
[885]	validation_0-logloss:0.60562
[886]	validation_0-logloss:0.60593
[887]	validation_0-logloss:0.60599
[888]	validation_0-logloss:0.60578
[889]	validation_0-logloss:0.60594
[890]	validation_0-logloss:0.60606
[891]	validation_0-logloss:0.60617
[892]	validation_0-logloss:0.60616
[893]	validation_0-logloss:0.60620
[894]	validation_0-logloss:0.60611
[895]	validation_0-logloss:0.60604
[896]	validation_0-logloss:0.60608
[897]	validation_0-logloss:0.60654
[898]	validation_0-logloss:0.60656
[899]	validation_0-logloss:0.60647
[900]	validation_0-logloss:0.60649
[901]	validation_0-logloss:0.60647
[902]	validation_0-logloss:0.60646
[903]	validation_0-logloss:0.60673
[904]	validation_0-logloss:0.60678
[905]	validation_0-logloss:0.60708
[906]	validation_0-logloss:0.60672
[907]	validation_0-logloss:0.60680
[908]	validation_0-logloss:0.60665
[909]	validation_0-logloss:0.60660
[910]	validation_0-logloss:0.60646
[911]	validation_0-logloss:0.60655
[912]	validation_0-logloss:0.60660
[913]	validation_0-logloss:0.60635
[914]	validation_0-logloss:0.60667
[915]	validation_0-logloss:0.60676
[916]	validation_0-logloss:0.60678
[917]	validation_0-logloss:0.60682
[918]	validation_0-logloss:0.60632
[919]	validation_0-logloss:0.60579
[920]	validation_0-logloss:0.60602
[921]	validation_0-logloss:0.60611
[922]	validation_0-logloss:0.60623
[923]	validation_0-logloss:0.60628
[924]	validation_0-logloss:0.60643
[925]	validation_0-logloss:0.60628
[926]	validation_0-logloss:0.60611
[927]	validation_0-logloss:0.60583
[928]	validation_0-logloss:0.60574
[929]	validation_0-logloss:0.60544
[930]	validation_0-logloss:0.60559
[931]	validation_0-logloss:0.60561
[932]	validation_0-logloss:0.60572
[933]	validation_0-logloss:0.60564
[934]	validation_0-logloss:0.60589
[935]	validation_0-logloss:0.60591
[936]	validation_0-logloss:0.60569
[937]	validation_0-logloss:0.60572
[938]	validation_0-logloss:0.60552
[939]	validation_0-logloss:0.60558
[940]	validation_0-logloss:0.60522
[941]	validation_0-logloss:0.60468
[942]	validation_0-logloss:0.60427
[943]	validation_0-logloss:0.60452
[944]	validation_0-logloss:0.60500
[945]	validation_0-logloss:0.60481
[946]	validation_0-logloss:0.60507
[947]	validation_0-logloss:0.60503
[948]	validation_0-logloss:0.60505
[949]	validation_0-logloss:0.60494
[950]	validation_0-logloss:0.60439
[951]	validation_0-logloss:0.60454
[952]	validation_0-logloss:0.60453
[953]	validation_0-logloss:0.60467
[954]	validation_0-logloss:0.60456
[955]	validation_0-logloss:0.60452
[956]	validation_0-logloss:0.60464
[957]	validation_0-logloss:0.60494
[958]	validation_0-logloss:0.60493
[959]	validation_0-logloss:0.60518
[960]	validation_0-logloss:0.60535
[961]	validation_0-logloss:0.60534
[962]	validation_0-logloss:0.60530
[963]	validation_0-logloss:0.60515
[964]	validation_0-logloss:0.60497
[965]	validation_0-logloss:0.60475
[966]	validation_0-logloss:0.60487
[967]	validation_0-logloss:0.60496
[968]	validation_0-logloss:0.60503
[969]	validation_0-logloss:0.60510
[970]	validation_0-logloss:0.60502
[971]	validation_0-logloss:0.60511
[972]	validation_0-logloss:0.60512
[973]	validation_0-logloss:0.60506
[974]	validation_0-logloss:0.60495
[975]	validation_0-logloss:0.60517
[976]	validation_0-logloss:0.60527
[977]	validation_0-logloss:0.60520
[978]	validation_0-logloss:0.60499
[979]	validation_0-logloss:0.60524
[980]	validation_0-logloss:0.60502
[981]	validation_0-logloss:0.60549
[982]	validation_0-logloss:0.60578
[983]	validation_0-logloss:0.60528
[984]	validation_0-logloss:0.60477
[985]	validation_0-logloss:0.60478
[986]	validation_0-logloss:0.60509
[987]	validation_0-logloss:0.60460
[988]	validation_0-logloss:0.60440
[989]	validation_0-logloss:0.60463
[990]	validation_0-logloss:0.60491
[991]	validation_0-logloss:0.60490
[992]	validation_0-logloss:0.60493
[993]	validation_0-logloss:0.60501
[994]	validation_0-logloss:0.60499
[995]	validation_0-logloss:0.60497
[996]	validation_0-logloss:0.60508
[997]	validation_0-logloss:0.60511
[998]	validation_0-logloss:0.60555
[999]	validation_0-logloss:0.60554

step-04 确认最优参数

print(clf.best_params_)
{'base_score': 0.5, 'colsample_bylevel': 0.7, 'colsample_bynode': 0.7, 'colsample_bytree': 0.6, 'gamma': 0, 'learning_rate': 0.1, 'max_bin': 12, 'max_depth': 6, 'min_child_weight': 30, 'n_estimators': 1000, 'reg_alpha': 2, 'reg_lambda': 3, 'subsample': 0.65}

step-05 选取最优模型

best_model=clf.best_estimator_

step-06 评价最优模型

model_eval2(best_model, train.values, test.values)
train_roc_auc_score: 0.8766644056264636
test_roc_auc_score: 0.7278343023255814
train_accuracy_score: 0.8
test_accuracy_score: 0.6833333333333333
train_precision_score: 0.8069963811821471
test__precision_score: 0.7162921348314607
train_recall_score: 0.8479087452471483
test_recall_score: 0.7412790697674418
train_f1_score: 0.8269468479604452
test_f1_score: 0.7285714285714285

step-07 保存并调用模型

joblib.dump(best_model ,  r'D:\Ensemble_Learning\xgboostinfo\xgboostgridbest.model')
best_model=joblib.load( r'D:\Ensemble_Learning\xgboostinfo\xgboostgridbest.model')
model_eval2(best_model, train.values, test.values)
train_roc_auc_score: 0.8766644056264636
test_roc_auc_score: 0.7278343023255814
train_accuracy_score: 0.8
test_accuracy_score: 0.6833333333333333
train_precision_score: 0.8069963811821471
test__precision_score: 0.7162921348314607
train_recall_score: 0.8479087452471483
test_recall_score: 0.7412790697674418
train_f1_score: 0.8269468479604452
test_f1_score: 0.7285714285714285

你可能感兴趣的:(python机器学习,数据挖掘,算法)