[scikit-learn]混淆矩阵（confusion_mat

from sklearn.metrics import confusion_matrix
y_true = ["dog", "dog", "dog", "cat", "cat", "cat", "cat"]
y_pred = ["cat", "cat", "dog", "cat", "cat", "cat", "cat"]
C2 = confusion_matrix(y_actual, y_predict, labels=["dog","cat"])
tn, fp, fn, tp =C2.ravel()
print(C2)
print(C2.ravel())
#对于两类 的情况ctrl/case   #labels 中顺序是 [FALSE,TRUE]  [negative,positive]
##labels  可以时字符串也可以时数字，对应 标签 y_actual的值
#因为两类多是  患病或者对照样品这样的参考，如果写反了就错了

tn,fp,fn,tp 并不是 比率，而是数量

image.png

下面的图和混淆矩阵内部位置不太一样 ，帮助理解

v2-ebfeb4f43468f0434b19dd9587d64782_720w.jpg

参考：
https://blog.csdn.net/wsljqian/article/details/99435808
sklearn.metricsconfusion_matrix