Do you mean as a generic constructor params for all (binary) classifiers?

can I start to work on this issue?

@claramatos we first need to agree on what's need to be done. I am not sure what @amueller has in mind.

Initial post in #17565

I think that we should expose the pos_label as one of the parameters of plot_precision_recall_curve. I even think that we should issue a warning in case of class imbalance and that the positive class considered is the one with most samples. In most of cases, you are reporting the wrong part of your result but it might be what we give as a result with defaults.

import pandas as pd
from sklearn.datasets import fetch_openml
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import plot_precision_recall_curve

X, y = fetch_openml(
    name="blood-transfusion-service-center",
    as_frame=True, return_X_y=True,
)
# Make columns and classes more human-readable
X.columns = ["Recency", "Frequency", "Monetary", "Time"]
y = y.apply(
    lambda x: "donated" if x == "2" else "not donated"
).astype("category")

X_train, X_test, y_train, y_test = train_test_split(
    X, y, shuffle=True, random_state=0, test_size=0.5
)

classifier = LogisticRegression().fit(X_train, y_train)
plot_precision_recall_curve(classifier, X_test, y_test)

One would have expected the following instead:

@glemaitre I was thinking about implementing it as you are suggesting, exposing the pos_label (having 1 as default)
if it is ok with you (and @amueller) I'll start to work on that

@claramatos You can open a PR. IMO, pos_label is fine but it might be changing during the review process but I really think that we should solve this issue. In the meanwhile, I will address #17572.

I already opened #17569 before to find this issue. So you can focus on the plot_roc_curve function.
You can use a similar test than in #17569 to ensure that we plot the proper curve.

@claramatos did you have time to work on the issue related to the plot_roc_curve?

@glemaitre I'm planning to work on it over the weekend

I was looking into plot_precision_recall_curve and plot_roc_curve and there are some pieces that are common. Should I take the opportunity and move them to base.py?