学习曲线显示过度拟合吗？

Question

我试图知道我的分类模型（二进制）是否有过度拟合，我得到了学习曲线。数据集为：6836个实例，其中有1006个正面类别。

1）如果我使用SMOTE来平衡类和RandomForest作为技术，我得到这个曲线，这些比率：TPR = 0.887 y FPR = 0.041：

请注意训练错误且几乎为0。

2）如果我使用函数＆＃34; balanced_subsample＆＃34; （附在最后）以平衡类和RandomForest作为技术，我获得此曲线，并且这些比率：TPR = 0.866 y FPR = 0.14：

请注意，在这种情况下，测试错误是平的。

• 模型是否会过度拟合？
• 哪一个更有意义？

函数＆＃34; balanced_subsample＆＃34;：

def balanced_subsample(x,y,subsample_size):

class_xs = []
min_elems = None

for yi in np.unique(y):
    elems = x[(y == yi)]
    class_xs.append((yi, elems))
    if min_elems == None or elems.shape[0] < min_elems:
        min_elems = elems.shape[0]

use_elems = min_elems
if subsample_size < 1:
    use_elems = int(min_elems*subsample_size)

xs = []
ys = []

for ci,this_xs in class_xs:
    if len(this_xs) > use_elems:
        np.random.shuffle(this_xs)

    x_ = this_xs[:use_elems]
    y_ = np.empty(use_elems)
    y_.fill(ci)

    xs.append(x_)
    ys.append(y_)

xs = np.concatenate(xs)
ys = np.concatenate(ys)

return xs,ys

EDIT1：有关代码和流程的更多信息

X = data
y = X.pop('myclass')


#There is categorical and numerical attributes in my data set, so here I vectorize the categorical attributes
arrX = vectorize_attributes(X)

#Here I use some code to balance my class using SMOTE or "balanced_subsample" approach
X_train_balanced, y_train_balanced=mySMOTEfunc(arrX, y)
#X_train_balanced, y_train_balanced=balanced_subsample(arrX, y) 

#TRAIN/TEST SPLIT (STRATIFIED K_FOLD is implicit)
X_train,X_test,y_train,y_test = train_test_split(X_train_balanced,y_train_balanced,test_size=0.25)

#Estimator
clf=RandomForestClassifier(random_state=np.random.seed()) 
param_grid = { 'n_estimators': [10,50,100,200,300],'max_features': ['auto', 'sqrt', 'log2']}

#Grid search
score_func = metrics.f1_score
CV_clf = GridSearchCV(estimator=clf, param_grid=param_grid, cv=10)
start = time()
CV_clf.fit(X_train, y_train)

#FIT & PREDICTION
model = CV_clf.best_estimator_
y_pred = model.predict(X_test)

EDIT2：在这种情况下，我尝试使用Gradient Boosting Classifier（GBC）在3种情况下：1）GBC + SMOTE，2）GBC + SMOTE +特征选择，以及3）GBC + SMOTE +特征选择+标准化

X = data
y = X.pop('myclass')

#There is categorical and numerical attributes in my data set, so here I vectorize the categorical attributes
arrX = vectorize_attributes(X)

#FOR SCENARIO 3: Normalization
standardized_X = preprocessing.normalize(arrX)

#FOR SCENARIO 2 y 3: Removing all but the k highest scoring features
arrX_features_selected = SelectKBest(chi2, k=5).fit_transform(standardized_X , y)

#Here I use some code to balance my class using SMOTE or "balanced_subsample" approach
X_train_balanced, y_train_balanced=mySMOTEfunc(arrX_features_selected , y)
#X_train_balanced, y_train_balanced=balanced_subsample(arrX_features_selected , y) 

#TRAIN/TEST SPLIT (STRATIFIED K_FOLD is implicit)
X_train,X_test,y_train,y_test = train_test_split(X_train_balanced,y_train_balanced,test_size=0.25)

#Estimator
clf=RandomForestClassifier(random_state=np.random.seed()) 
param_grid = { 'n_estimators': [10,50,100,200,300],'max_features': ['auto', 'sqrt', 'log2']}

#Grid search
score_func = metrics.f1_score
CV_clf = GridSearchCV(estimator=clf, param_grid=param_grid, cv=10)
start = time()
CV_clf.fit(X_train, y_train)

#FIT & PREDICTION
model = CV_clf.best_estimator_
y_pred = model.predict(X_test)

3个拟议方案的学习曲线是：

情景1：

SCENARIO 2：GBC + SMOTE +功能选择

SCENARIO 3：GBC + SMOTE +特征选择+标准化

Answer 1

所以，你的第一条曲线是有道理的。当您增加训练点时，您希望测试错误降低。当你有一个没有最大深度和100％最大样本的随机森林树木时，你会期望接近0列车误差。你可能已经过度适应了，但你可能不会用RandomForests（或者根据数据集，其他任何东西）变得更好。

你的第二条曲线没有意义。你应该再次得到接近0的火车错误，除非发生一些完全不稳定的事情（就像一个真正破坏的输入集）。我看不出你的代码有什么问题，我跑了你的功能;似乎工作正常。如果没有用代码发布完整的工作示例，我无能为力。