使用光线调整sklearn中的超参数

import numpy as np
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.svm import SVC
from ray import tune

# 加载数据集
data = load_iris()
X_train, X_test, y_train, y_test = train_test_split(data.data, data.target, test_size=0.2, random_state=42)

# 定义模型训练函数
def train_model(config):
    model = SVC(C=config["C"], kernel=config["kernel"], gamma=config["gamma"])
    model.fit(X_train, y_train)
    score = model.score(X_test, y_test)
    return score

# 定义超参数搜索空间
config_space = {
    "C": tune.loguniform(0.01, 10),
    "kernel": tune.choice(["linear", "rbf"]),
    "gamma": tune.uniform(0.01, 1)
}

# 定义超参数搜索算法和调度器
search_algorithm = tune.choice(["random", "bayesopt"])
scheduler = tune.choice(["fifo", "hyperband"])

# 运行超参数搜索
analysis = tune.run(
    train_model,
    config=config_space,
    search_alg=search_algorithm,
    scheduler=scheduler,
    num_samples=10,
    metric="mean_accuracy",
    mode="max"
)

# 打印最佳超参数和对应的性能指标
best_config = analysis.get_best_config(metric="mean_accuracy", mode="max")
best_score = analysis.best_result["mean_accuracy"]
print("Best config:", best_config)
print("Best score:", best_score)