为什么sklearn中的sklearn.metrics.RocCurveDisplay可视化的图像中的AUC值的有效小数位数为两位？而不能自定义调节（floating point precision）

Posted 2022-10-11 Data+Science+Insight

为什么sklearn中的sklearn.metrics.RocCurveDisplay可视化的图像中的AUC值的有效小数位数为两位？而不能自定义调节（floating point precision）

目录

为什么sklearn中的sklearn.metrics.RocCurveDisplay可视化的图像中的AUC值的有效小数位数为两位？而不能自定义调节（floating point precision）

问题：

剖析：

---

问题：

我们需要使用sklearn.metrics.RocCurveDisplay的from_estimator可视化交叉验证的ROC曲线。

想自定义每一折的AUC值的小数位数。

import numpy as np

from sklearn import datasets

# Import some data to play with
iris = datasets.load_iris()
X = iris.data
y = iris.target
X, y = X[y != 2], y[y != 2]
n_samples, n_features = X.shape

# Add noisy features
random_state = np.random.RandomState(0)
X = np.c_[X, random_state.randn(n_samples, 200 * n_features)]



import matplotlib.pyplot as plt

from sklearn import svm
from sklearn.metrics import auc
from sklearn.metrics import RocCurveDisplay
from sklearn.model_selection import StratifiedKFold

# Run classifier with cross-validation and plot ROC curves
cv = StratifiedKFold(n_splits=6)
classifier = svm.SVC(kernel="linear", probability=True, random_state=random_state)

tprs = []
aucs = []
mean_fpr = np.linspace(0, 1, 100)

fig, ax = plt.subplots()
for i, (train, test) in enumerate(cv.split(X, y)):
    classifier.fit(X[train], y[train])
    viz = RocCurveDisplay.from_estimator(
        classifier,
        X[test],
        y[test],
        name="ROC fold ".format(i),
        alpha=0.3,
        lw=1,
        ax=ax,
    )
    interp_tpr = np.interp(mean_fpr, viz.fpr, viz.tpr)
    interp_tpr[0] = 0.0
    tprs.append(interp_tpr)
    aucs.append(viz.roc_auc)

ax.plot([0, 1], [0, 1], linestyle="--", lw=2, color="r", label="Chance", alpha=0.8)

mean_tpr = np.mean(tprs, axis=0)
mean_tpr[-1] = 1.0
mean_auc = auc(mean_fpr, mean_tpr)
std_auc = np.std(aucs)
ax.plot(
    mean_fpr,
    mean_tpr,
    color="b",
    label=r"Mean ROC (AUC = %0.2f $\\pm$ %0.2f)" % (mean_auc, std_auc),
    lw=2,
    alpha=0.8,
)

std_tpr = np.std(tprs, axis=0)
tprs_upper = np.minimum(mean_tpr + std_tpr, 1)
tprs_lower = np.maximum(mean_tpr - std_tpr, 0)
ax.fill_between(
    mean_fpr,
    tprs_lower,
    tprs_upper,
    color="grey",
    alpha=0.2,
    label=r"$\\pm$ 1 std. dev.",
)

ax.set(
    xlim=[-0.05, 1.05],
    ylim=[-0.05, 1.05],
    title="Receiver operating characteristic example",
)
ax.legend(loc="lower right")
plt.show()

剖析：

程序中是硬编码的，无能为力，如果需要其它位数需要说动编译或者自己写一个函数去处理。

根本就在于sklearn中的roc_curve函数底层使用了硬编码。

而sklearn.metrics.RocCurveDisplay的from_estimator可视化底层依赖的是roc_curve函数。

      if self.roc_auc is not None and name is not None:
            line_kwargs["label"] = f"name (AUC = self.roc_auc:0.2f)"
        elif self.roc_auc is not None:
            line_kwargs["label"] = f"AUC = self.roc_auc:0.2f"
        elif name is not None:
            line_kwargs["label"] = name

参考：Receiver Operating Characteristic (ROC) with cross validation — scikit-learn 1.1.2 documentation

参考:https://github.com/scikit-learn/scikit-learn/blob/7e1e6d09b/sklearn/metrics/_plot/roc_curve.py#L110

参考：sklearn.metrics.roc_curve — scikit-learn 1.1.2 documentation

参考：python - How to set floating point precision in Sklearn RocCurveDisplay? - Stack Overflow