Class or Function Names TPESampler APIs TPESampler(...) This is Optuna’s TPE sampler with support for categorical_distance_func. categorical_distance_func: A mapping from categorical parameter names to distance functions. Each function receives two choices and returns a non-negative distance. Example import optuna import optunahub module = optunahub.load_module(package="samplers/tpe_cat_distance") def hamming_distance(x: str, y: str) -> float: return float(x != y) sampler = module.TPESampler( categorical_distance_func={"optimizer": hamming_distance}, ) study = optuna.create_study(sampler=sampler) def objective(trial): optimizer = trial.suggest_categorical("optimizer", ["adam", "sgd"]) return float(optimizer !