在Optuna中有没有等同于hyperopts lognormal的东西？

Question

我正在尝试使用Optuna对我的模型进行超参数调整。

我被困在一个地方，我想定义一个具有对数正态/正态分布的搜索空间。这在使用hp.lognormal的hyperopt中是可能的。可以使用Optuna的现有suggest_ api的组合来定义这样的空间吗？

Answer 1

您也许可以利用suggest_float(..., 0, 1)的逆转换(即U(0，1))，因为Optuna目前不直接为这两个发行版提供suggest_变体。此示例可能是一个起点https://gist.github.com/hvy/4ef02ee2945fe50718c71953e1d6381d请查找以下代码

import matplotlib.pyplot as plt
import numpy as np
from scipy.stats import norm
from scipy.special import erfcinv

import optuna


def objective(trial):
    # Suggest from U(0, 1) with Optuna.
    x = trial.suggest_float("x", 0, 1)

    # Inverse transform into normal.
    y0 = norm.ppf(x, loc=0, scale=1)

    # Inverse transform into lognormal.
    y1 = np.exp(-np.sqrt(2) * erfcinv(2 * x))

    return y0, y1


if __name__ == "__main__":
    n_objectives = 2  # Normal and lognormal.

    study = optuna.create_study(
        sampler=optuna.samplers.RandomSampler(),
        # Could be "maximize". Does not matter for this demonstration.
        directions=["minimize"] * n_objectives,
    )
    study.optimize(objective, n_trials=10000)

    fig, axs = plt.subplots(n_objectives)
    for i in range(n_objectives):
        axs[i].hist(list(t.values[i] for t in study.trials), bins=100)
    plt.show()