What is AUC?
AUC is the area under the ROC curve and the coordinate axis.
The ROC curve is generally above the straight line y=x, so the value range of AUC is between 0.5 and 1.
The closer the AUC is to 1, the higher the authenticity of the detection method; when it is equal to 0.5, the authenticity is the lowest and has no application value. Because for the 0-1 classification problem, the probability of guessing right is 0.5, and if the AUC is less than or equal to 0.5, it means that the accuracy is similar to guessing.
What are ROCs?
The ROC curve is a curve to evaluate the classification ability of the classifier; the ROC curve is determined by the values of FPR and TPR;
The x-axis is FPR, and the y-axis is TPR;
FPR (x-axis): False positive rate, the smaller the better; the prediction is wrong; it is actually a negative example, but the prediction is a positive example, and the prediction result is wrong;
TPR (y-axis): true positive rate, the bigger the better; it is actually a positive case, and the prediction is also a positive case, and the prediction result is correct;
The smaller the x-axis value and the larger the y-axis value, the better the classification ability of the classifier. So the closer the ROC curve is to the y-axis, the better the classifier is.
Replenish:
If AUC is used as the scoring standard, the result should be represented by a probability value, and you need to use predict_proba() to generate a probability value; and if you use predict(), if the result of the value calculation is 0, 1, the score will be much lower .