模型练习

内容介绍

1. 线性回归模型：
   • 线性回归对于特征的要求；
   • 处理长尾分布；
   • 理解线性回归模型；
2. 模型性能验证：
   • 评价函数与目标函数；
   • 交叉验证方法；
   • 留一验证方法；
   • 针对时间序列问题的验证；
   • 绘制学习率曲线；
   • 绘制验证曲线；
3. 嵌入式特征选择：
   • Lasso回归；
   • Ridge回归；
   • 决策树；
4. 模型对比：
   • 常用线性模型；
   • 常用非线性模型；
5. 模型调参：
   • 贪心调参方法；
   • 网格调参方法；
   • 贝叶斯调参方法；

import pandas as pd
import numpy as np
import warnings
warnings.filterwarnings('ignore')

# reduce_mem_usage 函数通过调整数据类型，帮助我们减少数据在内存中占用的空间
def reduce_mem_usage(df):
    """ iterate through all the columns of a dataframe and modify the data type
        to reduce memory usage.        
    """
    start_mem = df.memory_usage().sum() 
    print('Memory usage of dataframe is {:.2f} MB'.format(start_mem))
    
    for col in df.columns:
        col_type = df[col].dtype
        
        if col_type != object:
            c_min = df[col].min()
            c_max = df[col].max()
            if str(col_type)[:3] == 'int':
                if c_min > np.iinfo(np.int8).min and c_max < np.iinfo(np.int8).max:
                    df[col] = df[col].astype(np.int8)
                elif c_min > np.iinfo(np.int16).min and c_max < np.iinfo(np.int16).max:
                    df[col] = df[col].astype(np.int16)
                elif c_min > np.iinfo(np.int32).min and c_max < np.iinfo(np.int32).max:
                    df[col] = df[col].astype(np.int32)
                elif c_min > np.iinfo(np.int64).min and c_max < np.iinfo(np.int64).max:
                    df[col] = df[col].astype(np.int64)  
            else:
                if c_min > np.finfo(np.float16).min and c_max < np.finfo(np.float16).max:
                    df[col] = df[col].astype(np.float16)
                elif c_min > np.finfo(np.float32).min and c_max < np.finfo(np.float32).max:
                    df[col] = df[col].astype(np.float32)
                else:
                    df[col] = df[col].astype(np.float64)
        else:
            df[col] = df[col].astype('category')

    end_mem = df.memory_usage().sum() 
    print('Memory usage after optimization is: {:.2f} MB'.format(end_mem))
    print('Decreased by {:.1f}%'.format(100 * (start_mem - end_mem) / start_mem))
    return df

sample_feature = reduce_mem_usage(pd.read_csv('data_for_tree.csv'))

Memory usage of dataframe is 66876560.00 MB
Memory usage after optimization is: 18311636.00 MB
Decreased by 72.6%

sample_feature.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 199037 entries, 0 to 199036
Data columns (total 42 columns):
SaleID                 199037 non-null int32
name                   199037 non-null int32
regDate                199037 non-null int32
model                  199036 non-null float16
brand                  199037 non-null int8
bodyType               193130 non-null float16
fuelType               187512 non-null float16
gearbox                191173 non-null float16
power                  199037 non-null int16
kilometer              199037 non-null float16
notRepairedDamage      199037 non-null category
regionCode             199037 non-null int16
seller                 199037 non-null int8
offerType              199037 non-null int8
creatDate              199037 non-null int32
price                  149037 non-null float32
v_0                    199037 non-null float16
v_1                    199037 non-null float16
v_2                    199037 non-null float16
v_3                    199037 non-null float16
v_4                    199037 non-null float16
v_5                    199037 non-null float16
v_6                    199037 non-null float16
v_7                    199037 non-null float16
v_8                    199037 non-null float16
v_9                    199037 non-null float16
v_10                   199037 non-null float16
v_11                   199037 non-null float16
v_12                   199037 non-null float16
v_13                   199037 non-null float16
v_14                   199037 non-null float16
train                  199037 non-null int8
used_time              183965 non-null float16
city                   199037 non-null int8
brand_amount           199037 non-null float16
brand_price_max        199037 non-null float32
brand_price_median     199037 non-null float16
brand_price_min        199037 non-null float16
brand_price_sum        199037 non-null float32
brand_price_std        199037 non-null float16
brand_price_average    199037 non-null float16
power_bin              179090 non-null float16
dtypes: category(1), float16(27), float32(3), int16(2), int32(4), int8(5)
memory usage: 17.5 MB

continuous_feature_names = [x for x in sample_feature.columns if x not in ['price','brand','model']]

线性回归 & 五折交叉验证 & 模拟真实业务情况

sample_feature = sample_feature.dropna().replace('-', 0).reset_index(drop=True)
sample_feature['notRepairedDamage'] = sample_feature['notRepairedDamage'].astype(np.float32)
train = sample_feature[continuous_feature_names + ['price']]

train_X = train[continuous_feature_names]
train_y = train['price']

简单建模

train_X.columns

Index(['SaleID', 'name', 'regDate', 'bodyType', 'fuelType', 'gearbox', 'power',
       'kilometer', 'notRepairedDamage', 'regionCode', 'seller', 'offerType',
       'creatDate', 'v_0', 'v_1', 'v_2', 'v_3', 'v_4', 'v_5', 'v_6', 'v_7',
       'v_8', 'v_9', 'v_10', 'v_11', 'v_12', 'v_13', 'v_14', 'train',
       'used_time', 'city', 'brand_amount', 'brand_price_max',
       'brand_price_median', 'brand_price_min', 'brand_price_sum',
       'brand_price_std', 'brand_price_average', 'power_bin'],
      dtype='object')

from sklearn.linear_model import LinearRegression
model = LinearRegression(normalize=True)
model = model.fit(train_X, train_y)

# 查看训练的线性回归模型的截距（intercept）与权重(coef)
'intercept:'+ str(model.intercept_)

sorted(dict(zip(continuous_feature_names, model.coef_)).items(), key=lambda x:x[1], reverse=True)

[('v_6', 3445831.1633972577),
 ('v_8', 692770.7678128699),
 ('v_9', 198395.295841116),
 ('v_12', 27507.621428815353),
 ('v_7', 26037.510367580162),
 ('v_3', 21057.286931547267),
 ('v_11', 15092.514215718247),
 ('v_13', 11896.119762518123),
 ('v_10', 6634.124938581224),
 ('gearbox', 818.5776140507646),
 ('fuelType', 319.53469972986306),
 ('bodyType', 191.29134329918656),
 ('power', 26.481934441690022),
 ('city', 1.0274503775584902),
 ('brand_price_median', 0.5250658807342925),
 ('brand_price_std', 0.4481049803719174),
 ('used_time', 0.29417444834786904),
 ('creatDate', 0.1623914298648277),
 ('brand_amount', 0.1440874522380301),
 ('regionCode', 0.0443836108368366),
 ('regDate', 0.010031006447178373),
 ('brand_price_max', 0.004449273956073402),
 ('SaleID', 5.317592799096748e-05),
 ('offerType', 2.8819777071475983e-06),
 ('seller', 2.419576048851013e-06),
 ('train', -1.434236764907837e-07),
 ('brand_price_sum', -2.115967749073978e-05),
 ('name', -0.00014023995730379674),
 ('brand_price_average', -0.4294083622308756),
 ('brand_price_min', -2.2661811055112353),
 ('power_bin', -18.28807828149659),
 ('v_14', -217.31566559778227),
 ('kilometer', -360.2934863036238),
 ('notRepairedDamage', -513.4459404058576),
 ('v_0', -1350.7753342596618),
 ('v_4', -16630.51769044547),
 ('v_2', -33826.13890974259),
 ('v_1', -43804.90907416708),
 ('v_5', -53360.007331716195)]

from matplotlib import pyplot as plt
subsample_index = np.random.randint(low=0, high=len(train_y), size=50)

绘制特征v_9的值与标签的散点图，图片发现模型的预测结果（蓝色点）与真实标签（黑色点）的分布差异较大，且部分预测值出现了小于0的情况，说明我们的模型存在一些问题

plt.scatter(train_X['v_9'][subsample_index], train_y[subsample_index], color='black')
plt.scatter(train_X['v_9'][subsample_index], model.predict(train_X.loc[subsample_index]), color='blue')
plt.xlabel('v_9')
plt.ylabel('price')
plt.legend(['True Price','Predicted Price'],loc='upper right')
print('The predicted price is obvious different from true price')
plt.show()

The predicted price is obvious different from true price

import seaborn as sns
print('It is clear to see the price shows a typical exponential distribution')
plt.figure(figsize=(15,5))
plt.subplot(1,2,1)
sns.distplot(train_y)
plt.subplot(1,2,2)
sns.distplot(train_y[train_y < np.quantile(train_y, 0.9)])

It is clear to see the price shows a typical exponential distribution





<matplotlib.axes._subplots.AxesSubplot at 0x2123e2c8fc8>

# 在这里我们对标签进行了  𝑙𝑜𝑔(𝑥+1)  变换，使标签贴近于正态分布
train_y_ln = np.log(train_y + 1)
import seaborn as sns
print('The transformed price seems like normal distribution')
plt.figure(figsize=(15,5))
plt.subplot(1,2,1)
sns.distplot(train_y_ln)
plt.subplot(1,2,2)
sns.distplot(train_y_ln[train_y_ln < np.quantile(train_y_ln, 0.9)])

The transformed price seems like normal distribution





<matplotlib.axes._subplots.AxesSubplot at 0x2123e8c5308>

model = model.fit(train_X, train_y_ln)

print('intercept:'+ str(model.intercept_))
sorted(dict(zip(continuous_feature_names, model.coef_)).items(), key=lambda x:x[1], reverse=True)

intercept:-323.56388425612806





[('v_9', 7.077907185884061),
 ('v_5', 1.7860278461338932),
 ('v_12', 1.418400813080686),
 ('v_1', 1.3905712817315665),
 ('v_13', 1.0453959507409174),
 ('v_11', 0.8904032829993984),
 ('v_3', 0.6938311223489996),
 ('gearbox', 0.006998901777640581),
 ('power_bin', 0.005446388835649742),
 ('fuelType', 0.004926197345716262),
 ('bodyType', 0.004194306130198107),
 ('power', 0.0010051552773711423),
 ('creatDate', 1.8622079316177753e-05),
 ('brand_price_min', 1.5212532404786894e-05),
 ('brand_price_average', 3.6057617242970683e-06),
 ('brand_amount', 2.7506356959918595e-06),
 ('brand_price_max', 6.349379527535381e-07),
 ('SaleID', 1.787886046176962e-08),
 ('offerType', 9.667644462751923e-11),
 ('seller', 2.610534011182608e-11),
 ('train', 1.3642420526593924e-12),
 ('brand_price_sum', -1.2880256126660146e-10),
 ('name', -5.5090285160730094e-08),
 ('brand_price_median', -1.1892421114927995e-07),
 ('regionCode', -1.0320248688382342e-06),
 ('regDate', -1.3726795201954239e-06),
 ('brand_price_std', -2.236334197647584e-06),
 ('used_time', -4.196247894423456e-05),
 ('city', -0.0003906159940914123),
 ('kilometer', -0.013589777358806768),
 ('v_14', -0.014714513584856967),
 ('v_7', -0.06508871105975078),
 ('v_0', -0.08178982770228543),
 ('notRepairedDamage', -0.26628790017203813),
 ('v_2', -0.5529658177746759),
 ('v_4', -0.9218923136921164),
 ('v_10', -1.878068611441135),
 ('v_8', -42.513212411879785),
 ('v_6', -243.07783875427796)]

# 再次进行可视化，发现预测结果与真实值较为接近，且未出现异常状况
plt.scatter(train_X['v_9'][subsample_index], train_y[subsample_index], color='black')
plt.scatter(train_X['v_9'][subsample_index], np.exp(model.predict(train_X.loc[subsample_index])), color='blue')
plt.xlabel('v_9')
plt.ylabel('price')
plt.legend(['True Price','Predicted Price'],loc='upper right')
print('The predicted price seems normal after np.log transforming')
plt.show()

The predicted price seems normal after np.log transforming

五折交叉验证

在使用训练集对参数进行训练的时候，经常会发现人们通常会将一整个训练集分为三个部分（比如mnist手写训练集）。一般分为：训练集（train_set），评估集（valid_set），测试集（test_set）这三个部分。这其实是为了保证训练效果而特意设置的。其中测试集很好理解，其实就是完全不参与训练的数据，仅仅用来观测测试效果的数据。而训练集和评估集则牵涉到下面的知识了。

因为在实际的训练中，训练的结果对于训练集的拟合程度通常还是挺好的（初始条件敏感），但是对于训练集之外的数据的拟合程度通常就不那么令人满意了。因此我们通常并不会把所有的数据集都拿来训练，而是分出一部分来（这一部分不参加训练）对训练集生成的参数进行测试，相对客观的判断这些参数对训练集之外的数据的符合程度。这种思想就称为交叉验证（Cross Validation）

from sklearn.model_selection import cross_val_score
from sklearn.metrics import mean_absolute_error,  make_scorer

def log_transfer(func):
    def wrapper(y, yhat):
        result = func(np.log(y), np.nan_to_num(np.log(yhat)))
        return result
    return wrapper

scores = cross_val_score(model, X=train_X, y=train_y, verbose=1, cv = 5, scoring=make_scorer(log_transfer(mean_absolute_error)))

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:    1.0s finished

# 使用线性回归模型，对未处理标签的特征数据进行五折交叉验证（Error 1.36）
print('AVG:', np.mean(scores))

AVG: 1.3509622154531906

# 使用线性回归模型，对处理过标签的特征数据进行五折交叉验证（Error 0.19）
scores = cross_val_score(model, X=train_X, y=train_y_ln, verbose=1, cv = 5, scoring=make_scorer(mean_absolute_error))
print('AVG:', np.mean(scores))

[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.


AVG: 0.19183751589722967


[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:    0.9s finished

scores = pd.DataFrame(scores.reshape(1,-1))
scores.columns = ['cv' + str(x) for x in range(1, 6)]
scores.index = ['MAE']
scores

	cv1	cv2	cv3	cv4	cv5
MAE	0.190636	0.192098	0.19275	0.190861	0.192843

模拟真实业务情况

但在事实上，由于我们并不具有预知未来的能力，五折交叉验证在某些与时间相关的数据集上反而反映了不真实的情况。通过2018年的二手车价格预测2017年的二手车价格，这显然是不合理的，因此我们还可以采用时间顺序对数据集进行分隔。在本例中，我们选用靠前时间的4/5样本当作训练集，靠后时间的1/5当作验证集，最终结果与五折交叉验证差距不大

import datetime
sample_feature = sample_feature.reset_index(drop=True)
split_point = len(sample_feature) // 5 * 4
train = sample_feature.loc[:split_point].dropna()
val = sample_feature.loc[split_point:].dropna()

train_X = train[continuous_feature_names]
train_y_ln = np.log(train['price'] + 1)
val_X = val[continuous_feature_names]
val_y_ln = np.log(val['price'] + 1)
model = model.fit(train_X, train_y_ln)
mean_absolute_error(val_y_ln, model.predict(val_X))

0.19283569397506634

绘制学习率曲线与验证曲线

from sklearn.model_selection import learning_curve, validation_curve

def plot_learning_curve(estimator, title, X, y, ylim=None, cv=None,n_jobs=1, train_size=np.linspace(.1, 1.0, 5 )):  
    plt.figure()  
    plt.title(title)  
    if ylim is not None:  
        plt.ylim(*ylim)  
    plt.xlabel('Training example')  
    plt.ylabel('score')  
    train_sizes, train_scores, test_scores = learning_curve(estimator, X, y, cv=cv, n_jobs=n_jobs, train_sizes=train_size, scoring = make_scorer(mean_absolute_error))  
    train_scores_mean = np.mean(train_scores, axis=1)  
    train_scores_std = np.std(train_scores, axis=1)  
    test_scores_mean = np.mean(test_scores, axis=1)  
    test_scores_std = np.std(test_scores, axis=1)  
    plt.grid()#区域  
    plt.fill_between(train_sizes, train_scores_mean - train_scores_std,  
                     train_scores_mean + train_scores_std, alpha=0.1,  
                     color="r")  
    plt.fill_between(train_sizes, test_scores_mean - test_scores_std,  
                     test_scores_mean + test_scores_std, alpha=0.1,  
                     color="g")  
    plt.plot(train_sizes, train_scores_mean, 'o-', color='r',  
             label="Training score")  
    plt.plot(train_sizes, test_scores_mean,'o-',color="g",  
             label="Cross-validation score")  
    plt.legend(loc="best")  
    return plt  

plot_learning_curve(LinearRegression(), 'Liner_model', train_X[:1000], train_y_ln[:1000], ylim=(0.0, 0.5), cv=5, n_jobs=1)  

<module 'matplotlib.pyplot' from 'D:\\anaconda\\Anaconda3\\lib\\site-packages\\matplotlib\\pyplot.py'>

多种模型对比

train = sample_feature[continuous_feature_names + ['price']].dropna()

train_X = train[continuous_feature_names]
train_y = train['price']
train_y_ln = np.log(train_y + 1)

线性模型 & 嵌入式特征选择

在过滤式和包裹式特征选择方法中，特征选择过程与学习器训练过程有明显的分别。而嵌入式特征选择在学习器训练过程中自动地进行特征选择。嵌入式选择最常用的是L1正则化与L2正则化。在对线性回归模型加入两种正则化方法后，他们分别变成了岭回归与Lasso回归。

from sklearn.linear_model import LinearRegression
from sklearn.linear_model import Ridge
from sklearn.linear_model import Lasso

models = [LinearRegression(),
          Ridge(),
          Lasso()]

result = dict()
for model in models:
    model_name = str(model).split('(')[0]
    scores = cross_val_score(model, X=train_X, y=train_y_ln, verbose=0, cv = 5, scoring=make_scorer(mean_absolute_error))
    result[model_name] = scores
    print(model_name + ' is finished')

LinearRegression is finished
Ridge is finished
Lasso is finished

result = pd.DataFrame(result)
result.index = ['cv' + str(x) for x in range(1, 6)]
result

	LinearRegression	Ridge	Lasso
cv1	0.190636	0.194434	0.381675
cv2	0.192098	0.195851	0.378666
cv3	0.192750	0.196521	0.382266
cv4	0.190861	0.194865	0.378322
cv5	0.192843	0.196609	0.381600

model = LinearRegression().fit(train_X, train_y_ln)
print('intercept:'+ str(model.intercept_))
sns.barplot(abs(model.coef_), continuous_feature_names)

intercept:-323.5638824355704





<matplotlib.axes._subplots.AxesSubplot at 0x2123f0e1b88>

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-nKhoEXPD-1585749119864)(output_39_2.png)]

L2正则化在拟合过程中通常都倾向于让权值尽可能小，最后构造一个所有参数都比较小的模型。因为一般认为参数值小的模型比较简单，能适应不同的数据集，也在一定程度上避免了过拟合现象。可以设想一下对于一个线性回归方程，若参数很大，那么只要数据偏移一点点，就会对结果造成很大的影响；但如果参数足够小，数据偏移得多一点也不会对结果造成什么影响，专业一点的说法是『抗扰动能力强』

model = Ridge().fit(train_X, train_y_ln)
print('intercept:'+ str(model.intercept_))
sns.barplot(abs(model.coef_), continuous_feature_names)

intercept:-313.08053786017797





<matplotlib.axes._subplots.AxesSubplot at 0x2123f266548>

L1正则化有助于生成一个稀疏权值矩阵，进而可以用于特征选择。如下图，我们发现power与userd_time特征非常重要。

model = Lasso().fit(train_X, train_y_ln)
print('intercept:'+ str(model.intercept_))
sns.barplot(abs(model.coef_), continuous_feature_names)

intercept:-262.76278370429543





<matplotlib.axes._subplots.AxesSubplot at 0x21242b42588>

除此之外，决策树通过信息熵或GINI指数选择分裂节点时，优先选择的分裂特征也更加重要，这同样是一种特征选择的方法。XGBoost与LightGBM模型中的model_importance指标正是基于此计算的

非线性模型

除了线性模型以外，还有许多我们常用的非线性模型如下，在此篇幅有限不再一一讲解原理。我们选择了部分常用模型与线性模型进行效果比对。

from sklearn.linear_model import LinearRegression
from sklearn.svm import SVC
from sklearn.tree import DecisionTreeRegressor
from sklearn.ensemble import RandomForestRegressor
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.neural_network import MLPRegressor
from xgboost.sklearn import XGBRegressor
from lightgbm.sklearn import LGBMRegressor

models = [LinearRegression(),
          DecisionTreeRegressor(),
          RandomForestRegressor(),
          GradientBoostingRegressor(),
          MLPRegressor(solver='lbfgs', max_iter=100), 
          XGBRegressor(n_estimators = 100, objective='reg:squarederror'), 
          LGBMRegressor(n_estimators = 100)]

result = dict()
for model in models:
    model_name = str(model).split('(')[0]
    scores = cross_val_score(model, X=train_X, y=train_y_ln, verbose=0, cv = 5, scoring=make_scorer(mean_absolute_error))
    result[model_name] = scores
    print(model_name + ' is finished')

LinearRegression is finished
DecisionTreeRegressor is finished
RandomForestRegressor is finished
GradientBoostingRegressor is finished
MLPRegressor is finished
XGBRegressor is finished
LGBMRegressor is finished

# 可以看到随机森林模型在每一个fold中均取得了更好的效果
result = pd.DataFrame(result)
result.index = ['cv' + str(x) for x in range(1, 6)]
result

	LinearRegression	DecisionTreeRegressor	RandomForestRegressor	GradientBoostingRegressor	MLPRegressor	XGBRegressor	LGBMRegressor
cv1	0.190636	0.196053	0.140345	0.168761	11996.412741	0.136414	0.141475
cv2	0.192098	0.187891	0.139094	0.170193	10906.332250	0.136383	0.142744
cv3	0.192750	0.186494	0.138625	0.170948	5952.913318	0.136034	0.142839
cv4	0.190861	0.186894	0.136944	0.169144	10415.738128	0.136390	0.140496
cv5	0.192843	0.199474	0.141740	0.171634	9386.509133	0.146308	0.141699

模型调参

在此我们介绍了三种常用的调参方法如下：

• 贪心算法
• 网格调参
• 贝叶斯调参

## LGB的参数集合：

objective = ['regression', 'regression_l1', 'mape', 'huber', 'fair']

num_leaves = [3,5,10,15,20,40, 55]
max_depth = [3,5,10,15,20,40, 55]
bagging_fraction = []
feature_fraction = []
drop_rate = []

贪心调参

best_obj = dict()
for obj in objective:
    model = LGBMRegressor(objective=obj)
    score = np.mean(cross_val_score(model, X=train_X, y=train_y_ln, verbose=0, cv = 5, scoring=make_scorer(mean_absolute_error)))
    best_obj[obj] = score
    
best_leaves = dict()
for leaves in num_leaves:
    model = LGBMRegressor(objective=min(best_obj.items(), key=lambda x:x[1])[0], num_leaves=leaves)
    score = np.mean(cross_val_score(model, X=train_X, y=train_y_ln, verbose=0, cv = 5, scoring=make_scorer(mean_absolute_error)))
    best_leaves[leaves] = score
    
best_depth = dict()
for depth in max_depth:
    model = LGBMRegressor(objective=min(best_obj.items(), key=lambda x:x[1])[0],
                          num_leaves=min(best_leaves.items(), key=lambda x:x[1])[0],
                          max_depth=depth)
    score = np.mean(cross_val_score(model, X=train_X, y=train_y_ln, verbose=0, cv = 5, scoring=make_scorer(mean_absolute_error)))
    best_depth[depth] = score

sns.lineplot(x=['0_initial','1_turning_obj','2_turning_leaves','3_turning_depth'], y=[0.143 ,min(best_obj.values()), min(best_leaves.values()), min(best_depth.values())])

<matplotlib.axes._subplots.AxesSubplot at 0x2123679fcc8>

Grid Search 调参

from sklearn.model_selection import GridSearchCV
parameters = {'objective': objective , 'num_leaves': num_leaves, 'max_depth': max_depth}
model = LGBMRegressor()
clf = GridSearchCV(model, parameters, cv=5)
clf = clf.fit(train_X, train_y)
clf.best_params_

model = LGBMRegressor(objective='regression',
                          num_leaves=55,
                          max_depth=15)
np.mean(cross_val_score(model, X=train_X, y=train_y_ln, verbose=0, cv = 5, scoring=make_scorer(mean_absolute_error)))

贝叶斯调参

from bayes_opt import BayesianOptimization
def rf_cv(num_leaves, max_depth, subsample, min_child_samples):
    val = cross_val_score(
        LGBMRegressor(objective = 'regression_l1',
            num_leaves=int(num_leaves),
            max_depth=int(max_depth),
            subsample = subsample,
            min_child_samples = int(min_child_samples)
        ),
        X=train_X, y=train_y_ln, verbose=0, cv = 5, scoring=make_scorer(mean_absolute_error)
    ).mean()
    return 1 - val

rf_bo = BayesianOptimization(
    rf_cv,
    {
    'num_leaves': (2, 10),
    'max_depth': (2, 10),
    'subsample': (0.1, 0.5),
    'min_child_samples' : (2, 10)
    }
)

rf_bo.maximize()

1 - rf_bo.max['target']