Skip to content

Latest commit

 

History

History
42 lines (23 loc) · 2.17 KB

File metadata and controls

42 lines (23 loc) · 2.17 KB

Classification performance standard

Description

001 Classification-Performance-Standard.ipynb Classification performance standards such as accuracy, precision, recall, confusion matrix, F1, ROC, AUC. Take MNIST as example.

002 Classification-Performance-Standard.ipynb The framework to evaluate multi-classification classifier automatically.

Standard

Confusion matrix

Predict Positive Predict Negative
Actual Positive TP(True Positive) FN(False Negative)
Actual Negative FP(False Positive) TN(True Negative)

Confusion matrix

The vertical axis of the ROC(Receiver Operating Characteristic) curve is TPR(True Positive Rate) and the horizontal axis is FPR(False Positive Rate).

AUC is the area under ROC curve.

ROC

ROC(each class)