在Python中创建一个阈值编码的ROC图

R's ROCR package provides options for ROC curve plotting that will color code and label threshold values along the curve:

R的ROCR软件包提供了ROC曲线绘制的选项，它将沿着曲线绘制颜色代码和标签阈值:

The closest I can get with Python is something like

与Python最接近的是类似的东西

from sklearn.metrics import roc_curve
fpr, tpr, thresholds = roc_curve(qualityTrain.PoorCare, qualityTrain.Pred1)
plt.plot(fpr, tpr, label='ROC curve', color='b')
plt.axes().set_aspect('equal')
plt.xlim([-0.05, 1.05])
plt.ylim([-0.05, 1.05])

which gives

这给了

Are there packages that provide functionality equivalent to R's ability to label (using print.cutoffs.at) and color code (using colorize) thresholds? Presumably this information is in thresholds, returned by sklearn.metrics.roc_curve, but I can't figure out how to use it to color code and label the figure.

是否有提供与R标记(使用print.cutoff .at)和颜色代码(使用着色)阈值功能相同的功能的包?假定该信息是阈值，由sklearn.metrics返回。roc_curve，但我不知道如何用它来给图形上色和贴标签。

1 个解决方案

#1

Look at this gist:

看看这个要点:

https://gist.github.com/podshumok/c1d1c9394335d86255b8

https://gist.github.com/podshumok/c1d1c9394335d86255b8

roc_data = sklearn.metrics.roc_curve(...)
plot_roc(*roc_data, label_every=5)