AutoML- ray tune

原创

已于 2023-03-27 21:50:55 修改 · 357 阅读

2 ·

CC 4.0 BY-SA版权

文章标签：

#python #pytorch #开发语言 #autoML

于 2023-03-27 21:50:35 首次发布

一、介绍

支持深度学习调参，优化器算法支持全面，且支持分布式计算，本地可以有效利用多核计算，文档及维护更新较多，安装使用快捷方便。

评分较好
在这里插入图片描述

安装:

pip install "ray[tune]" torch torchvision

二、相关用法示例

2.1 demo 示例

定义参数搜索空间 search_space

定义模型评分计算函数 objective

from ray import tune

# 1. Define an objective function.
def objective(config):
    score = config["a"] ** 2 + config["b"]
    return {
   
   "score": score}


# 2. Define a search space.
search_space = {
   
   
    "a": tune.grid_search([0.001, 0.01, 0.1, 1.0]),
    "b": tune.choice([1, 2, 3]),
}

# 3. Start a Tune run and print the best result.
tuner = tune.Tuner(objective, param_space=search_space)
results = tuner.fit()
print(results.get_best_result(metric="score", mode="min").config)

结果:

在这里插入图片描述

三、资源参数示例

3.1 机器资源参数设定

tune.with_resources(trainable, {“cpu”: 2})

示例:

# If you have 4 CPUs on your machine, this will run 2 concurrent trials at a time.
trainable_with_resources = tune.with_resources(trainable, {
   
   "cpu": 2})
tuner = tune.Tuner(
    trainable_with_resources,
    tune_config=tune.TuneConfig(num_samples=10)
)
results = tuner.fit()

# If you have 4 CPUs on your machine, this will run 1 trial at a time.
trainable_with_resources = tune.with_resources(trainable, {
   
   "cpu": 4})
tuner = tune.Tuner(
    trainable_with_resources,
    tune_config=tune.TuneConfig(num_samples=10)
)
results = tuner.fit()

# Fractional values are also supported, (i.e., {"cpu": 0.5}).
# If you have 4 CPUs on your machine, this will run 8 concurrent trials at a time.
trainable_with_resources = tune.with_resources(trainable, {