November 2019
Intermediate to advanced
346 pages
9h 36m
English
In the following steps, we will load a dataset, train a classifier, and then tune a threshold to satisfy a false positive rate constraint:
import numpy as npfrom scipy import sparseimport scipyX_train = scipy.sparse.load_npz("training_data.npz")y_train = np.load("training_labels.npy")X_test = scipy.sparse.load_npz("test_data.npz")y_test = np.load("test_labels.npy")desired_FPR = 0.01
from sklearn.metrics import confusion_matrixdef FPR(y_true, y_pred): """Calculate the False Positive Rate.""" CM = confusion_matrix(y_true, y_pred) TN = CM[0][0] FP = CM[0][1] return FP / (FP + TN)def TPR(y_true, y_pred): """Calculate ...