Cannot plot PR curve with labels of type `np.str_`

[auguste-probabl]

from sklearn.datasets import load_iris
from sklearn.linear_model import LogisticRegression
from skore import train_test_split, ComparisonReport, CrossVal
idationReport
iris = load_iris(as_frame=True)
X, y = iris.data, iris.target
y = iris.target_names[y]
split_data = train_test_split(
    X=X, y=y, random_state=0, as_dict=True, shuffle=True
)
report_1 = CrossValidationReport(LogisticRegression(), X, y)
report_2 = CrossValidationReport(LogisticRegression(max_iter=500), X, y)
report = ComparisonReport([report_1, report_2])
display = report.metrics.precision_recall()
display.plot()

raises a strange-sounding error: UndefinedVariableError: name 'np' is not defined

The offending line is

skore/skore/src/skore/_sklearn/_plot/metrics/precision_recall_curve.py

Line 657 in b2ed6df

query = f"label == {label!r} & estimator_name == '{estimator_name}'"

The issue is that labels are np.str_ instances, so the !r in the f-string results in "label == np.str_('setosa') & ..." but it should be "label == 'setosa' & ...". We use !r to ensure that normal Python strings are printed with surrounding quotes:

f"label == {label}" -> "label == setosa"
while
f"label == {label!r}" -> "label == 'setosa'".

We need the surrounding quotes for Pandas queries to work.

Environment

Commit b2ed6df