Does n_jobs =-1 work on SPE?

[IncubatorShokuhou]

import imbens

from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split

RANDOM_STATE = 42
X, y = make_classification(
    n_classes=3,
    n_samples=20_000,
    class_sep=2,
    weights=[0.01, 0.29, 0.6],
    n_informative=3,
    n_redundant=1,
    flip_y=0,
    n_features=400,
    n_clusters_per_class=2,
    random_state=RANDOM_STATE,
)
X_train, X_valid, y_train, y_valid = train_test_split(
    X, y, test_size=0.5, stratify=y, random_state=RANDOM_STATE
)
clf = imbens.ensemble.SelfPacedEnsembleClassifier(random_state=RANDOM_STATE,n_jobs=-1)
clf.fit(X_train, y_train)

Here is my code. I set n_jobs=-1 in SPEClassfier, but when I looked at bpytop, I can only see one or two cores running. Is there something wrong with my code, or SPE is not designed for parallelization?