Can the infinitely simple pythongo be backtested?
In the evaluation of binary classification results, ROC curve is a general standard, which is established by TPR and FPR. Related knowledge recommendation blog posts let you thoroughly understand the accuracy rate, accuracy rate, recall rate, truth rate, false positive rate and ROC/AUC. The calculation of TPR and FPR is based on a series of selected/(/(thresholds). The purpose of this paper is to find the optimal threshold and judge whether ROC is enough to distinguish between the good and bad classification model and two kinds of objects with false positive rate.