相关疑难解决方法(0)

我需要specificity我的分类,定义为: TN/(TN+FP)

我正在写一个自定义得分手功能:

from sklearn.metrics import make_scorer
def specificity_loss_func(ground_truth, predictions):
    print predictions
    tp, tn, fn, fp = 0.0,0.0,0.0,0.0
    for l,m in enumerate(ground_truth):        
        if m==predictions[l] and m==1:
            tp+=1
        if m==predictions[l] and m==0:
            tn+=1
        if m!=predictions[l] and m==1:
            fn+=1
        if m!=predictions[l] and m==0:
            fp+=1
    `return tn/(tn+fp)

score = make_scorer(specificity_loss_func, greater_is_better=True)

然后,

from sklearn.dummy import DummyClassifier
clf_dummy = DummyClassifier(strategy='most_frequent', random_state=0)
ground_truth = [0,0,1,0,1,1,1,0,0,1,0,0,1]
p  = [0,0,0,1,0,1,1,1,1,0,0,1,0]
clf_dummy = clf_dummy.fit(ground_truth, p)
score(clf_dummy, ground_truth, p)

当我运行这些命令时,我p打印为:

[0 0 0 0 0 …

我正在使用此代码来比较多个模型的性能：

from sklearn import model_selection

X = input data
Y = binary labels

models = []
models.append(('LR', LogisticRegression()))
models.append(('LDA', LinearDiscriminantAnalysis()))
models.append(('KNN', KNeighborsClassifier()))
models.append(('CART', DecisionTreeClassifier()))

results = []
names = []
scoring = 'accuracy'

for name, model in models:
    kfold = model_selection.KFold(n_splits=10, random_state=7)
    cv_results = model_selection.cross_val_score(model, X, Y, cv=kfold,scoring=scoring)
    results.append(cv_results)
    names.append(name)
    msg = "%s: %.2f (%.2f)" % (name, cv_results.mean(), cv_results.std())
    print(msg)

我可以使用“准确性”和“回忆”作为评分，这些将提供准确性和敏感性。我怎样才能创建一个给我“特异性”的记分员

特异性= TN/(TN+FP)

其中 TN 和 FP 是混淆矩阵中的真负值和假正值

我试过这个

def tp(y_true, y_pred): 
error= confusion_matrix(y_true, y_pred)[0,0]/(confusion_matrix(y_true,y_pred)[0,0] + confusion_matrix(y_true, y_pred)[0,1])
return error …