Describe the issue linked to the documentation

I have observed that the plotted feature receive the color of the expected label, i.e., train, test, and outlier.

Instead, I would expect to plot the predicted labels, i.e., the color of the plotted features according to the prediction of the method. In the published example, the variables defined below are not used for the plot.

Suggest a potential alternative/fix

The expected plot would be something like this:

b1 = plt.scatter(X_train[:, 0], X_train[:, 1], c=['white' if x > 0 else 'red' for x in y_pred_train],
                 s=20, edgecolor='k')
b2 = plt.scatter(X_test[:, 0], X_test[:, 1], c=['green' if x > 0 else 'red' for x in y_pred_test],
                 s=20, edgecolor='k')
c = plt.scatter(X_outliers[:, 0], X_outliers[:, 1], c=['yellow' if x > 0 else 'red' for x in y_pred_outliers],
                s=20, edgecolor='k')

And the result in this case would vary:

Link to the online example: https://scikit-learn.org/stable/auto_examples/ensemble/plot_isolation_forest.html