天池学习赛：工业蒸汽量预测4——模型验证

最新推荐文章于 2022-10-22 20:00:00 发布

小屋*

最新推荐文章于 2022-10-22 20:00:00 发布

阅读量783

点赞数

分类专栏：数据处理文章标签：机器学习 python 数据分析

本文链接：https://blog.csdn.net/weixin_48994268/article/details/114988342

版权

数据处理专栏收录该内容

9 篇文章 2 订阅

订阅专栏

上一篇《天池学习赛：工业蒸汽量预测3——模型训练》中已经是使用了几种机器学习的模型，接下来将介绍一些模型的评价方法。

1 模型评估的方法

1 欠拟合与过拟合

2 泛化与正则化

3 回归模型评价指标与调用方法

（1）平均绝对误差

from sklearn.metrics import mean_absolute_error
mean_absolute_error(y_test,y_pred)

（2）均方误差

from sklearn.metrics import mean_squared_error
mean_squared_error(y_test,y_pred)

（3）均方根误差

from sklearn.metrics import mean_squared_error
Pred_Error=mean_squared_error(y_test,y_pred)
Sqrt(Pred_Error)

（4）R平方值

from sklearn.metrics import r2_score
r2_score(y_test,y_pred)

4 交叉验证
（1）简单交叉验证：

from sklearn.model_selection import train_test_split
X_train,X_test,Y_train,Y_test=train_test_split(iris.data,iris.target,test_size=.4,random_state=0)

（2）K折交叉验证：

from sklearn.model_selection import KFold
kf=KFold(n_splits=10)

（3）留一法交叉验证：

from sklearn.model_selection import LeaveOneOut
l0o=LeaveOneOut()

（4）留P法交叉验证：

from sklearn.model_selection import LeavePOut
lpo=LeavePOut(p=5)

2 模型调参

1 调参

2 网格搜索
对所有可能的参数组合，依次进行训练，最后找出最好的参数组合

from sklearn.datasets import load_iris
from sklearn.svm import SVC
from sklearn.model_selection import train_test_split

iris=load_iris()
X_train,X_test,Y_train,Y_test=train_test_split(iris.data,iris.target,random_state=0)
print("size of training set:{} size of teating set:{}".format(X_train[0],X_test[0]))

best_score=0
for gamma in[0.001,0.01,0.1,1,10,100]:
    for c in[0.001,0.01,0.1,1,10,100]:
        svm=SVC(gamma=gamma,C=c)
        svm.fit(X_train,Y_train)
        score=svm.score(X_test,Y_test)
        if score >best_score:
            best_score=score
            best_parameters={'gamma':gamma,'C':c}
            
print("Best score:{:.2f}",format(best_score))
print('Best parameters:{}'.format(best_parameters))

可以从输出结果看出gamma=0.001与C=100的组合最好，得分最高。

size of training set:[5.9 3. 4.2 1.5] size of teating set:[5.8 2.8 5.1 2.4]
Best score:0.97
Best parameters:{‘gamma’: 0.001, ‘C’: 100}

3 学习曲线

3 赛题模型验证与调参

3.1 模型过拟合与欠拟合

1 基础代码

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats
from sklearn import preprocessing
from statsmodels.stats.outliers_influence import variance_inflation_factor
from sklearn.decomposition import PCA
import warnings
warnings.filterwarnings("ignore")
from sklearn.model_selection import train_test_split #切分数据
from sklearn.metrics import mean_squared_error  #评价指标
from sklearn.linear_model import LinearRegression   #从sklearn引入线性模型
from sklearn.neighbors import KNeighborsRegressor   #k近邻回归模型
from sklearn.tree import DecisionTreeRegressor     #决策树回归模型
from sklearn.ensemble import RandomForestRegressor    #随机森林回归模型
from lightgbm import LGBMRegressor   #LightGBM回归模型
from sklearn.svm import SVR    #支持向量机
from sklearn.linear_model import SGDRegressor

#读数据
train_data_file = "./zhengqi_train.txt"
test_data_file = "./zhengqi_test.txt"
train_data=pd.read_csv(train_data_file,sep='\t',encoding='utf-8')
test_data=pd.read_csv(test_data_file,sep='\t',encoding='utf-8')

#归一化
features_columns=[col for col in train_data.columns if col not in ['target']]
min_max_scaler=preprocessing.MinMaxScaler()
min_max_scaler=min_max_scaler.fit(train_data[features_columns])
train_data_scaler=min_max_scaler.transform(train_data[features_columns])
test_data_scaler=min_max_scaler.transform(test_data[features_columns])
train_data_scaler=pd.DataFrame(train_data_scaler)
train_data_scaler.columns=features_columns
test_data_scaler=pd.DataFrame(test_data_scaler)
test_data_scaler.columns=features_columns
train_data_scaler['target']=train_data['target']

#PCA降维
pca=PCA(n_components=0.9)
new_train_pca_16=pca.fit_transform(train_data_scaler.iloc[:,0:-1])
new_test_pca_16=pca.transform(test_data_scaler)
new_train_pca_16=pd.DataFrame(new_train_pca_16)
new_test_pca_16=pd.DataFrame(new_test_pca_16)
new_train_pca_16['target']=train_data_scaler['target']

#采用PCA保留的16维特征的数据
new_train_pca_16=new_train_pca_16.fillna(0)
train=new_train_pca_16[new_train_pca_16.columns]
target=new_train_pca_16['target']

#划分数据集   训练集80%验证机20%
train_data,test_data,train_target,test_target=train_test_split(train,target,\
                                                               test_size=0.2,random_state=0)

2 欠拟合

clf=SGDRegressor(max_iter=500,tol=1e-2)
clf.fit(train_data,train_target)
score_train=mean_squared_error(train_target,clf.predict(train_data))
score_test=mean_squared_error(test_target,clf.predict(test_data))
print("SGDRegressor train MSE:",score_train)
print("SGDRegressor test MSE:",score_test)

3 过拟合

from sklearn.preprocessing import PolynomialFeatures
poly=PolynomialFeatures(5)
train_data_poly=poly.fit_transform(train_data)
test_data_poly=poly.fit_transform(test_data)
clf=SGDRegressor(max_iter=1000,tol=1e-3)
clf.fit(train_data_poly,train_target)
score_train=mean_squared_error(train_target,clf.predict(train_data_poly))
score_test=mean_squared_error(test_target,clf.predict(test_data_poly))
print("SGDRegressor train MSE:",score_train)
print("SGDRegressor test MSE:",score_test)

4 正常拟合

from sklearn.preprocessing import PolynomialFeatures
poly=PolynomialFeatures(3)
train_data_poly=poly.fit_transform(train_data)
test_data_poly=poly.fit_transform(test_data)
clf=SGDRegressor(max_iter=1000,tol=1e-3)
clf.fit(train_data_poly,train_target)
score_train=mean_squared_error(train_target,clf.predict(train_data_poly))
score_test=mean_squared_error(test_target,clf.predict(test_data_poly))
print("SGDRegressor train MSE:",score_train)
print("SGDRegressor test MSE:",score_test)

3.2 模型正则化

1 L2正则化

在上面正常拟合的代码中加入正则化项

from sklearn.preprocessing import PolynomialFeatures
poly=PolynomialFeatures(3)
train_data_poly=poly.fit_transform(train_data)
test_data_poly=poly.fit_transform(test_data)
clf=SGDRegressor(max_iter=1000,tol=1e-3,penalty='L2',alpha=0.0001)
clf.fit(train_data_poly,train_target)
score_train=mean_squared_error(train_target,clf.predict(train_data_poly))
score_test=mean_squared_error(test_target,clf.predict(test_data_poly))
print("SGDRegressor train MSE:",score_train)
print("SGDRegressor test MSE:",score_test)

2 L1正则化

同上

from sklearn.preprocessing import PolynomialFeatures
poly=PolynomialFeatures(3)
train_data_poly=poly.fit_transform(train_data)
test_data_poly=poly.fit_transform(test_data)
clf=SGDRegressor(max_iter=1000,tol=1e-3,penalty='L1',alpha=0.0001)
clf.fit(train_data_poly,train_target)
score_train=mean_squared_error(train_target,clf.predict(train_data_poly))
score_test=mean_squared_error(test_target,clf.predict(test_data_poly))
print("SGDRegressor train MSE:",score_train)
print("SGDRegressor test MSE:",score_test)

3 ElasticNet联合L1和L2范数加权正则化

from sklearn.preprocessing import PolynomialFeatures
poly=PolynomialFeatures(3)
train_data_poly=poly.fit_transform(train_data)
test_data_poly=poly.fit_transform(test_data)
clf=SGDRegressor(max_iter=1000,tol=1e-3,penalty='elasticnet',l1_ratio=0.9,alpha=.0001)
clf.fit(train_data_poly,train_target)
score_train=mean_squared_error(train_target,clf.predict(train_data_poly))
score_test=mean_squared_error(test_target,clf.predict(test_data_poly))
print("SGDRegressor train MSE:",score_train)
print("SGDRegressor test MSE:",score_test)

3.3 模型交叉验证

1 简单交叉

2 K折交叉

from sklearn.model_selection import KFold
kf=KFold(n_splits=5)
for k,(train_index,test_index) in enumerate(kf.split(train)):
    train_data,test_data,train_target,test_target=train.values[train_index],\
                                                  train.values[test_index],\
                                                  target[train_index],\
                                                  target[test_index]
    clf = SGDRegressor(max_iter=1000, tol=1e-3)
    clf.fit(train_data, train_target)
    score_train = mean_squared_error(train_target, clf.predict(train_data))
    score_test = mean_squared_error(test_target, clf.predict(test_data))
    print(k,"折","SGDRegressor train MSE:", score_train)
    print(k,"折","SGDRegressor test MSE:", score_test,"\n")

3.留一法交叉

from sklearn.model_selection import LeaveOneOut
loo=LeaveOneOut()
num=100
for k,(train_index,test_index) in enumerate(loo.split(train)):
    train_data, test_data, train_target, test_target = train.values[train_index], \
                                                       train.values[test_index], \
                                                       target[train_index], \
                                                       target[test_index]
    clf = SGDRegressor(max_iter=1000, tol=1e-3)
    clf.fit(train_data, train_target)
    score_train = mean_squared_error(train_target, clf.predict(train_data))
    score_test = mean_squared_error(test_target, clf.predict(test_data))
    print(k, "个", "SGDRegressor train MSE:", score_train)
    print(k, "个", "SGDRegressor test MSE:", score_test, "\n")
    if k>=9:
        break

4 留P法交叉

from sklearn.model_selection import LeavePOut
lpo=LeavePOut(p=10)
num=100
for k,(train_index,test_index) in enumerate(lpo.split(train)):
    train_data, test_data, train_target, test_target = train.values[train_index], \
                                                       train.values[test_index], \
                                                       target[train_index], \
                                                       target[test_index]
    clf = SGDRegressor(max_iter=1000, tol=1e-3)
    clf.fit(train_data, train_target)
    score_train = mean_squared_error(train_target, clf.predict(train_data))
    score_test = mean_squared_error(test_target, clf.predict(test_data))
    print(k, "10个", "SGDRegressor train MSE:", score_train)
    print(k, "10个", "SGDRegressor test MSE:", score_test, "\n")
    if k >= 9:
        break

3.4 模型超参空间及调参

1 穷举网格搜索

#使用数据训练随机森林模型，采用穷举网格搜索方法调参
from sklearn.model_selection import GridSearchCV
from sklearn.ensemble import RandomForestRegressor

RandomForestRegressor=RandomForestRegressor()
parameters={'n_estimators':[50,100,200],'max_depth':[1,2,3]}
clf=GridSearchCV(RandomForestRegressor,parameters,cv=5)
clf.fit(train_data,train_target)
score_test=mean_squared_error(test_target,clf.predict(test_data))
print("RandomForestRegressor GridSearchCV test MSE:",score_test)
print(sorted(clf.cv_results_.keys()))    #包含训练时间和验证指标的一些信息

2 随机参数优化

#使用数据训练随机森林模型，采用随即参数优化方法调参
from sklearn.model_selection import RandomizedSearchCV
from sklearn.ensemble import RandomForestRegressor
RandomForestRegressor=RandomForestRegressor()
parameters={
    'n_estimators':[50,100,200,300],
    'max_depth':[1,2,3,4,5]
}
clf=RandomizedSearchCV(RandomForestRegressor,parameters,cv=5)
clf.fit(train_data,train_target)
print('Best parameters found are:',clf.best_params_)
score_test=mean_squared_error(test_target,clf.predict(test_data))
print("RandomForestRegressor RandomizedSearchCV GridSearchCV test MSE:",score_test)
print(sorted(clf.cv_results_.keys()))   #包含训练时间和验证指标的一些信息

3 LGB调参

import lightgbm as lgb

clf=lgb.LGBMRegressor(num_leaves=31)
parameters={'learning_rate':[0.01,0.1,1],'n_estimators':[20,40]}
clf.fit(train_data,train_target)
score_test=mean_squared_error(test_target,clf.predict(test_data))
print("LGBMRegressor GridSearchCV test MSE:",score_test)

3.5 学习曲线和验证曲线

1 学习曲线

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection import ShuffleSplit
from sklearn.linear_model import SGDRegressor
from sklearn.model_selection import learning_curve

train_data_file = "./zhengqi_train.txt"
test_data_file = "./zhengqi_test.txt"
train_data=pd.read_csv(train_data_file,sep='\t',encoding='utf-8')
test_data=pd.read_csv(test_data_file,sep='\t',encoding='utf-8')

plt.figure(figsize=(18,10),dpi=150)

def plot_learning_curve(estimator,title,x,y,ylim=None,cv=None,n_jobs=1,train_sizes=np.linspace(.1,1.0,5)):
    plt.figure()
    plt.title(title)
    if ylim is not None:
        plt.ylim(*ylim)
    plt.xlabel("Training examples")
    plt.ylabel("Score")
    train_sizes,train_scores,test_scores=learning_curve(estimator,x,y,cv=cv,n_jobs=n_jobs,train_sizes=train_sizes)
    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()
    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="corss-validation score")
    plt.legend(loc="best")
    return plt
x=train_data[test_data.columns].values
y=train_data['target'].values
title="LinearRegression"
cv=ShuffleSplit(n_splits=100,test_size=0.2,random_state=0)
estimator=SGDRegressor()
plot_learning_curve(estimator,title,x,y,ylim=(0.7,1.01),cv=cv,n_jobs=-1).show()

在这里插入图片描述

2 验证曲线

绘制数据训练SGDRegressor模型的曲线

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.linear_model import SGDRegressor
from sklearn.model_selection import validation_curve

train_data_file = "./zhengqi_train.txt"
test_data_file = "./zhengqi_test.txt"
train_data=pd.read_csv(train_data_file,sep='\t',encoding='utf-8')
test_data=pd.read_csv(test_data_file,sep='\t',encoding='utf-8')
x=train_data[test_data.columns].values
y=train_data['target'].values

param_range=[0.1,0.01,0.001,0.0001,0.00001,0.000001]
train_scores,test_scores=validation_curve(SGDRegressor(max_iter=1000,tol=1e-3,penalty='L1')\
                                          ,x,y,param_name='alpha',param_range=param_range,\
                                          cv=10,scoring='r2',n_jobs=1)
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.title("Validation Curve with SCDRegressor")
plt.xlabel("alpha")
plt.ylabel("Score")
plt.ylim(0.0,1.1)
plt.semilogx(param_range,train_scores_mean,label="Training scores",color='r')
plt.fill_between(param_range,train_scores_mean-train_scores_std,train_scores_mean+train_scores_std,alpha=0.2,color='r')
plt.semilogx(param_range,test_scores_mean,label="Sross_validation score",color='g')
plt.fill_between(param_range,train_scores_mean-train_scores_std,train_scores_mean+train_scores_std,alpha=0.2,color='g')
plt.legend(loc="best")
plt.show()

在这里插入图片描述

下一篇《天池学习赛：工业蒸汽量预测5——特征优化》

小屋*

关注

0
点赞
踩
8

收藏

觉得还不错? 一键收藏
打赏
0
评论
天池学习赛：工业蒸汽量预测4——模型验证

上一篇《天池学习赛：工业蒸汽量预测3——模型训练》中已经是使用了几种机器学习的模型，接下来将介绍一些模型的评价方法。目录1 模型评估的方法2 模型调参3 赛题模型验证与调参1 模型评估的方法1 欠拟合与过拟合2 泛化与正则化3 回归模型评价指标与调用方法（1）平均绝对误差from sklearn.metrics import mean_absolute_errormean_absolute_error(y_test,y_pred)（2）均方误差from sklearn.metrics i
复制链接

扫一扫