画ROC曲线和计算AUC、准确率、召回率、F1等

#!/usr/bin/env python
# coding: utf-8

import sys
from sklearn.metrics import roc_curve, auc, precision_score, recall_score, f1_score
import matplotlib.pyplot as plt

yTrue = list()
yPred = list()
yPredBinary = list()

for line in open(sys.argv[1]):
    splits = line.strip().split("\t")
    if len(splits) < 2:
        continue
    label = [0, 1][int(splits[0])]
    yTrue.append(label)
    yPred.append(float(splits[1]))
    yPredBinary.append(float(splits[1]) > 0.5)

precision = precision_score(yTrue, yPredBinary)
recall = recall_score(yTrue, yPredBinary)
f1 = f1_score(yTrue, yPredBinary)
fpr, tpr, thresholds = roc_curve(yTrue, yPred)
auc = auc(fpr, tpr)

print("precision:\t" + str(precision))
print("recall:\t" + str(recall))
print("f1:\t" + str(f1))
print("auc:\t" + str(auc))

plt.plot(fpr, tpr, lw=1, label='ROC (area = %0.3f)' % auc)
plt.show()