Describe the issue:

Please take a look at here:

• GPSampler suddenly slows down once n_trials hits exactly 100 optuna/optuna#6230 (comment)

Reproduce the code example:

import optuna

def objective(trial: optuna.Trial) -> float:
    x = trial.suggest_float("x", -5, 5)
    y = trial.suggest_float("y", -5, 5)
    return (x - 2.0) ** 2 + (y - 3.0) ** 2

sampler = optuna.samplers.GPSampler(seed=0)
study = optuna.create_study(sampler=sampler)
study.optimize(objective, n_trials=200)

Error message:

Python and NumPy Versions:

Python: 3.11
OS: Ubuntu 22.04
NumPy: 2.0.0+ (this problem seems to go away for v1.X)

Runtime Environment:

No response

Context for the issue:

Are there any thresholdings to switch which implementations to use in some operations?
Tbh, the sudden slowdown appears very strange to me without such a manual algorithm switch.