在 Python 中获取数组的 True Positives、TN、FP 和 FN

Question

my data set result look like this

yval
Out[59]: 
array([[0, 0, 0, ..., 0, 0, 0],
       [0, 0, 0, ..., 0, 0, 0],
       [0, 0, 0, ..., 0, 0, 0],
       ...,
       [0, 0, 0, ..., 0, 0, 0],
       [0, 0, 0, ..., 0, 0, 0],
       [1, 0, 0, ..., 0, 0, 0]])

and predicted results look like this

y_pred
Out[60]: 
array([[0, 0, 0, ..., 0, 0, 0],
       [0, 0, 0, ..., 0, 0, 0],
       [0, 0, 0, ..., 0, 0, 0],
       ...,
       [0, 0, 0, ..., 0, 0, 0],
       [0, 0, 0, ..., 0, 0, 0],
       [0, 0, 0, ..., 0, 0, 0]])

I want to find the TP, TN, FP, and FP

I tried this

cm=confusion_matrix(yval, y_pred)

which gave this error

ValueError: multilabel-indicator is not supported

tried this

cm=confusion_matrix(yval.argmax(axis=1), y_pred.argmax(axis=1))
TN = cm[0][0]
FN = cm[1][0]
TP = cm[1][1]
FP = cm[0][1]

gave zeros for all values TN=0, FN=0, TP=0 and FP=0

how can I get these values for a predicted array?

Answer 1

You can use from sklearn.metrics import multilabel_confusion_matrix to import multilabel confusion matrix. After that the drill is:

cm = multilabel_confusion_matrix(yval, y_pred)