关于机器学习:机器学习实战-AutoML自动化机器学习建模

44次阅读

共计 20863 个字符，预计需要花费 53 分钟才能阅读完成。

作者：韩信子 @ShowMeAI
教程地址：http://www.showmeai.tech/tutorials/41
本文地址：http://www.showmeai.tech/article-detail/210
申明：版权所有，转载请分割平台与作者并注明出处
珍藏 ShowMeAI 查看更多精彩内容

1.AutoML 与自动化机器学习

在前序系列文章中大家跟着 ShowMeAI 一起学习了如何构建机器学习利用。咱们构建一个机器学习模型解决方案 baseline 很容易，但模型抉择和泛化性能优化是一项艰巨的工作。抉择适合的模型并是一个须要高计算成本、工夫和精力的过程。

针对上述问题就提出了 AutoML，AutoML(Automated machine learning) 是自动化构建端到端机器学习流程，解决理论场景问题的过程。

在本篇内容中咱们将介绍到微软开发的高效轻量级自动化机器学习框架 FLAML(A Fast and Lightweight AutoML Library)。

2.FLAML 介绍

2.1 FLAML 特点

官方网站对 FLAML 的特点总结如下：

  • 对于分类和回归等常见的机器学习工作，FLAML 能够在耗费尽量少的资源前提下，疾速找到高质量模型。它反对经典机器学习模型和深度神经网络。

  • 它很容易定制或扩大。用户能够有很灵便的调整与定制模式：

    • 最小定制(设定计算资源限度)
    • 中等定制(例如设定 scikit-learn 学习器、搜寻空间和度量规范)
    • 齐全定制(自定义训练和评估代码)。
  • 它反对疾速且低消耗的主动调优，可能解决大型搜寻空间。FLAML 由 Microsoft Research 创造的新的高效益超参数优化和学习器抉择办法撑持。

2.2 装置办法

咱们能够通过 pip 轻松装置上 FLAML

pip install flaml

有一些可选的装置选项，如下：

(1) Notebook 示例反对

如果大家要跑官网的 notebook 代码示例，装置时增加 [notebook] 选项:

pip install flaml[notebook]

(2) 更多模型学习器反对

  • 如果咱们心愿 flaml 反对 catboost 模型，装置时增加 [catboost] 选项
pip install flaml[catboost]
  • 如果咱们心愿 flaml 反对 vowpal wabbit，装置时增加 [vw] 选项
pip install flaml[vw]
  • 如果咱们心愿 flaml 反对工夫序列预估器 prophet 和 statsmodels，装置时能够增加[forecast]
pip install flaml[forecast]

(3) 分布式调优反对

  • ray
pip install flaml[ray]
  • nni
pip install flaml[nni]
  • blendsearch
pip install flaml[blendsearch]

3.FLAML 应用示例

3.1 全自动模式

上面咱们用一个场景数据案例 (二分类) 来演示 FLAML 工具库的全自动模式。(大家能够在 jupyter notebook 中运行下列的代码，对于 IDE 与环境配置大家能够参考 ShowMeAI 文章 图解 python | 装置与环境设置)。

!pip install flaml[notebook]

(1) 加载数据和预处理

咱们从 OpenML 下载航空公司数据集 Airlines dataset。这个数据集的建模工作是在给定预约腾飞信息的状况下预测给定航班是否会延误。

from flaml.data import load_openml_dataset
X_train, X_test, y_train, y_test = load_openml_dataset(dataset_id=1169, data_dir='./')

从运行后果能够看到训练集测试集及标签的维度信息。

(2) 运行 FLAML 全自动模式

上面咱们间接运行 FLAML automl 全自动模式，理论在运行配置中，咱们能够指定 工作类型、工夫估算、误差度量、学习者列表、是否下采样、重采样策略类型等。如果不作任何设定的话，所有这些参数都会应用默认值(例如，默认分类器是 [lgbm, xgboost, xgb_limitdepth, catboost, rf, extra_tree, lrl1])。

# 导入工具库并初始化 AutoML 对象
from flaml import AutoML
automl = AutoML()
# 参数设定
settings = {"time_budget": 240,  # 总工夫下限(单位秒)
    "metric": 'accuracy',  # 候选能够是: 'r2', 'rmse', 'mae', 'mse', 'accuracy', 'roc_auc', 'roc_auc_ovr', 'roc_auc_ovo', 'log_loss', 'mape', 'f1', 'ap', 'ndcg', 'micro_f1', 'macro_f1'
    "task": 'classification',  # 工作类型
    "log_file_name": 'airlines_experiment.log',  # flaml 日志文件
    "seed": 7654321,    # 随机种子
}
# 运行自动化机器学习
automl.fit(X_train=X_train, y_train=y_train, **settings)

从上述运行后果能够看出，主动机器学习过程，对 [lgbm, xgboost, xgb_limitdepth, catboost, rf, extra_tree, lrl1] 这些候选模型进行了试验，并运行出了对应的后果。

(3) 最优模型与评估后果

print('Best ML leaner:', automl.best_estimator)
print('Best hyperparmeter config:', automl.best_config)
print('Best accuracy on validation data: {0:.4g}'.format(1-automl.best_loss))
print('Training duration of best run: {0:.4g} s'.format(automl.best_config_train_time))

运行后果如下

Best ML leaner: lgbm
Best hyperparmeter config: {'n_estimators': 1071, 'num_leaves': 25, 'min_child_samples': 36, 'learning_rate': 0.10320258241974468, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.08547376339713011, 'FLAML_sample_size': 364083}
Best accuracy on validation data: 0.6696
Training duration of best run: 9.274 s

能够通过运行结束的 automl 对象属性，取出对应的「最优模型」、「最佳模型配置」、「评估准则后果」等信息。在这里最优的模型是 1071 颗树构建而成的一个 LightGBM 模型。

更进一步，咱们能够通过上面的代码，取出最优模型，并用它对测试集进行预测。

# 最佳模型
automl.model.estimator

运行后果如下

LGBMClassifier(learning_rate=0.10320258241974468, max_bin=1023,
               min_child_samples=36, n_estimators=1071, num_leaves=25,
               reg_alpha=0.0009765625, reg_lambda=0.08547376339713011,
               verbose=-1)

(4) 模型存储与加载

# 模型存储与长久化
import pickle
with open('automl.pkl', 'wb') as f:
    pickle.dump(automl, f, pickle.HIGHEST_PROTOCOL)

# 模型加载
with open('automl.pkl', 'rb') as f:
    automl = pickle.load(f)
# 对测试集进行预估
y_pred = automl.predict(X_test)
print('Predicted labels', y_pred)
print('True labels', y_test)
y_pred_proba = automl.predict_proba(X_test)[:,1]

运行后果如下：

Predicted labels ['1' '0' '1' ... '1' '0' '0']
True labels 118331    0
328182    0
335454    0
520591    1
344651    0
         ..
367080    0
203510    1
254894    0
296512    1
362444    0
Name: Delay, Length: 134846, dtype: category
Categories (2, object): ['0' < '1']

能够看到，automl 失去的最佳模型，对测试集预估的形式，和本人建模失去的模型是一样的。

# 测试集成果评估
from flaml.ml import sklearn_metric_loss_score
print('accuracy', '=', 1 - sklearn_metric_loss_score('accuracy', y_pred, y_test))
print('roc_auc', '=', 1 - sklearn_metric_loss_score('roc_auc', y_pred_proba, y_test))
print('log_loss', '=', sklearn_metric_loss_score('log_loss', y_pred_proba, y_test))

评估后果如下：

accuracy = 0.6720332824110467
roc_auc = 0.7253276908529442
log_loss = 0.6034449031876942

(5) 查看日志历史详情

咱们能够通过如下代码，查看 automl 对各个模型试验的后果具体数据。

from flaml.data import get_output_from_log
time_history, best_valid_loss_history, valid_loss_history, config_history, metric_history = \
    get_output_from_log(filename=settings['log_file_name'], time_budget=240)
for config in config_history:
    print(config)

后果如下

{'Current Learner': 'lgbm', 'Current Sample': 10000, 'Current Hyper-parameters': {'n_estimators': 4, 'num_leaves': 4, 'min_child_samples': 20, 'learning_rate': 0.09999999999999995, 'log_max_bin': 8, 'colsample_bytree': 1.0, 'reg_alpha': 0.0009765625, 'reg_lambda': 1.0, 'FLAML_sample_size': 10000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 4, 'num_leaves': 4, 'min_child_samples': 20, 'learning_rate': 0.09999999999999995, 'log_max_bin': 8, 'colsample_bytree': 1.0, 'reg_alpha': 0.0009765625, 'reg_lambda': 1.0, 'FLAML_sample_size': 10000}}
{'Current Learner': 'lgbm', 'Current Sample': 10000, 'Current Hyper-parameters': {'n_estimators': 4, 'num_leaves': 14, 'min_child_samples': 15, 'learning_rate': 0.22841390623808822, 'log_max_bin': 9, 'colsample_bytree': 1.0, 'reg_alpha': 0.0014700173967242716, 'reg_lambda': 7.624911621832711, 'FLAML_sample_size': 10000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 4, 'num_leaves': 14, 'min_child_samples': 15, 'learning_rate': 0.22841390623808822, 'log_max_bin': 9, 'colsample_bytree': 1.0, 'reg_alpha': 0.0014700173967242716, 'reg_lambda': 7.624911621832711, 'FLAML_sample_size': 10000}}
{'Current Learner': 'lgbm', 'Current Sample': 10000, 'Current Hyper-parameters': {'n_estimators': 4, 'num_leaves': 25, 'min_child_samples': 12, 'learning_rate': 0.5082200481556802, 'log_max_bin': 8, 'colsample_bytree': 0.9696263001275751, 'reg_alpha': 0.0028107036379524425, 'reg_lambda': 3.716898117989413, 'FLAML_sample_size': 10000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 4, 'num_leaves': 25, 'min_child_samples': 12, 'learning_rate': 0.5082200481556802, 'log_max_bin': 8, 'colsample_bytree': 0.9696263001275751, 'reg_alpha': 0.0028107036379524425, 'reg_lambda': 3.716898117989413, 'FLAML_sample_size': 10000}}
{'Current Learner': 'lgbm', 'Current Sample': 10000, 'Current Hyper-parameters': {'n_estimators': 23, 'num_leaves': 14, 'min_child_samples': 15, 'learning_rate': 0.22841390623808822, 'log_max_bin': 9, 'colsample_bytree': 1.0, 'reg_alpha': 0.0014700173967242718, 'reg_lambda': 7.624911621832699, 'FLAML_sample_size': 10000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 23, 'num_leaves': 14, 'min_child_samples': 15, 'learning_rate': 0.22841390623808822, 'log_max_bin': 9, 'colsample_bytree': 1.0, 'reg_alpha': 0.0014700173967242718, 'reg_lambda': 7.624911621832699, 'FLAML_sample_size': 10000}}
{'Current Learner': 'lgbm', 'Current Sample': 10000, 'Current Hyper-parameters': {'n_estimators': 101, 'num_leaves': 12, 'min_child_samples': 24, 'learning_rate': 0.07647794276357095, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.001749539645587163, 'reg_lambda': 4.373760956394571, 'FLAML_sample_size': 10000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 101, 'num_leaves': 12, 'min_child_samples': 24, 'learning_rate': 0.07647794276357095, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.001749539645587163, 'reg_lambda': 4.373760956394571, 'FLAML_sample_size': 10000}}
{'Current Learner': 'lgbm', 'Current Sample': 40000, 'Current Hyper-parameters': {'n_estimators': 101, 'num_leaves': 12, 'min_child_samples': 24, 'learning_rate': 0.07647794276357095, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.001749539645587163, 'reg_lambda': 4.373760956394571, 'FLAML_sample_size': 40000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 101, 'num_leaves': 12, 'min_child_samples': 24, 'learning_rate': 0.07647794276357095, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.001749539645587163, 'reg_lambda': 4.373760956394571, 'FLAML_sample_size': 40000}}
{'Current Learner': 'lgbm', 'Current Sample': 40000, 'Current Hyper-parameters': {'n_estimators': 361, 'num_leaves': 11, 'min_child_samples': 32, 'learning_rate': 0.13528717598813866, 'log_max_bin': 9, 'colsample_bytree': 0.9851977789068981, 'reg_alpha': 0.0038372002422749616, 'reg_lambda': 0.25113531892556773, 'FLAML_sample_size': 40000}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 361, 'num_leaves': 11, 'min_child_samples': 32, 'learning_rate': 0.13528717598813866, 'log_max_bin': 9, 'colsample_bytree': 0.9851977789068981, 'reg_alpha': 0.0038372002422749616, 'reg_lambda': 0.25113531892556773, 'FLAML_sample_size': 40000}}
{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 361, 'num_leaves': 11, 'min_child_samples': 32, 'learning_rate': 0.13528717598813866, 'log_max_bin': 9, 'colsample_bytree': 0.9851977789068981, 'reg_alpha': 0.0038372002422749616, 'reg_lambda': 0.25113531892556773, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 361, 'num_leaves': 11, 'min_child_samples': 32, 'learning_rate': 0.13528717598813866, 'log_max_bin': 9, 'colsample_bytree': 0.9851977789068981, 'reg_alpha': 0.0038372002422749616, 'reg_lambda': 0.25113531892556773, 'FLAML_sample_size': 364083}}
{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 547, 'num_leaves': 46, 'min_child_samples': 60, 'learning_rate': 0.281323306091088, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.001643352694266288, 'reg_lambda': 0.14719738747481906, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 547, 'num_leaves': 46, 'min_child_samples': 60, 'learning_rate': 0.281323306091088, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.001643352694266288, 'reg_lambda': 0.14719738747481906, 'FLAML_sample_size': 364083}}
{'Current Learner': 'lgbm', 'Current Sample': 364083, 'Current Hyper-parameters': {'n_estimators': 1071, 'num_leaves': 25, 'min_child_samples': 36, 'learning_rate': 0.10320258241974468, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.08547376339713011, 'FLAML_sample_size': 364083}, 'Best Learner': 'lgbm', 'Best Hyper-parameters': {'n_estimators': 1071, 'num_leaves': 25, 'min_child_samples': 36, 'learning_rate': 0.10320258241974468, 'log_max_bin': 10, 'colsample_bytree': 1.0, 'reg_alpha': 0.0009765625, 'reg_lambda': 0.08547376339713011, 'FLAML_sample_size': 364083}}

咱们能够绘制出验证集的学习曲线，如下：

import matplotlib.pyplot as plt
import numpy as np
plt.title('Learning Curve')
plt.xlabel('Wall Clock Time (s)')
plt.ylabel('Validation Accuracy')
plt.scatter(time_history, 1 - np.array(valid_loss_history))
plt.step(time_history, 1 - np.array(best_valid_loss_history), where='post')
plt.show()

(6) 比照默认 XGBoost/LightGBM 试验后果

咱们来比照一下全部应用默认参数的 XGBoost 模型在本数据集上的成果，代码如下

from xgboost import XGBClassifier
from lightgbm import LGBMClassifier

# 训练拟合
xgb = XGBClassifier()
cat_columns = X_train.select_dtypes(include=['category']).columns
X = X_train.copy()
X[cat_columns] = X[cat_columns].apply(lambda x: x.cat.codes)
xgb.fit(X, y_train)

lgbm = LGBMClassifier()
lgbm.fit(X_train, y_train)

# 测试集预估
X = X_test.copy()
X[cat_columns] = X[cat_columns].apply(lambda x: x.cat.codes)
y_pred_xgb = xgb.predict(X)

y_pred_lgbm = lgbm.predict(X_test)

# 评估成果
print('默认参数 xgboost accuracy', '=', 1 - sklearn_metric_loss_score('accuracy', y_pred_xgb, y_test))
print('默认参数 lgbm accuracy', '=', 1 - sklearn_metric_loss_score('accuracy', y_pred_lgbm, y_test))
print('flaml (4min) accuracy', '=', 1 - sklearn_metric_loss_score('accuracy', y_pred, y_test))

最终后果如下：

默认参数 xgboost accuracy = 0.6676060098186078
默认参数 lgbm accuracy = 0.6602346380315323
flaml (4min) accuracy = 0.6720332824110467

从比照后果中能够看出，flaml 主动机器学习调优的最佳模型，成果优于默认参数的 XGBoost 和 LightGBM 建模后果。

3.2 自定义学习器

除了齐全自动化模式应用 FLAML 工具库，咱们还能够对它的一些组件进行自定义，实现自定义调优。比方咱们能够对「模型」「参数搜寻空间」「候选学习器」「模型优化指标」等进行设置。

(1) 自定义模型

正则化贪婪森林 (RGF) 是一种机器学习办法，目前未蕴含在 FLAML 中。RGF 有许多调整参数，其中最要害的是：[max_leaf, n_iter, n_tree_search, opt_interval, min_samples_leaf]。要运行自定义 / 新学习器，用户须要提供以下信息：

  • 自定义 / 新学习器的实现
  • 超参数名称和类型的列表
  • 超参数的粗略范畴(即下限 / 上限)

在上面的示例代码中，RGF 信息被包装在一个名为 MyRegularizedGreedyForest 的 python 类中。

from flaml.model import SKLearnEstimator
from flaml import tune
from flaml.data import CLASSIFICATION

class MyRegularizedGreedyForest(SKLearnEstimator):
    def __init__(self, task='binary', **config):
        '''Constructor
        
        Args:
            task: A string of the task type, one of
                'binary', 'multi', 'regression'
            config: A dictionary containing the hyperparameter names
                and 'n_jobs' as keys. n_jobs is the number of parallel threads.
        '''super().__init__(task, **config)'''task=binary or multi for classification task'''
        if task in CLASSIFICATION:
            from rgf.sklearn import RGFClassifier

            self.estimator_class = RGFClassifier
        else:
            from rgf.sklearn import RGFRegressor
            
            self.estimator_class = RGFRegressor

    @classmethod
    def search_space(cls, data_size, task):
        '''[required method] search space
        Returns:
            A dictionary of the search space. 
            Each key is the name of a hyperparameter, and value is a dict with
                its domain (required) and low_cost_init_value, init_value,
                cat_hp_cost (if applicable).
                e.g.,
                {'domain': tune.randint(lower=1, upper=10), 'init_value': 1}.
        '''space = {'max_leaf': {'domain': tune.lograndint(lower=4, upper=data_size[0]),'init_value': 4,'low_cost_init_value': 4},'n_iter': {'domain': tune.lograndint(lower=1, upper=data_size[0]),'init_value': 1,'low_cost_init_value': 1},'n_tree_search': {'domain': tune.lograndint(lower=1, upper=32768),'init_value': 1,'low_cost_init_value': 1},'opt_interval': {'domain': tune.lograndint(lower=1, upper=10000),'init_value': 100},'learning_rate': {'domain': tune.loguniform(lower=0.01, upper=20.0)},'min_samples_leaf': {'domain': tune.lograndint(lower=1, upper=20),'init_value': 20},
        }
        return space

    @classmethod
    def size(cls, config):
        '''[optional method] memory size of the estimator in bytes
        
        Args:
            config - the dict of the hyperparameter config
        Returns:
            A float of the memory size required by the estimator to train the
            given config
        '''max_leaves = int(round(config['max_leaf']))
        n_estimators = int(round(config['n_iter']))
        return (max_leaves * 3 + (max_leaves - 1) * 4 + 1.0) * n_estimators * 8

    @classmethod
    def cost_relative2lgbm(cls):
        '''[optional method] relative cost compared to lightgbm'''
        return 1.0

(2) 运行 FLAML 自定义模型 automl

将 RGF 增加到学习器列表后，咱们通过调整 RGF 的超参数以及默认学习器来运行 automl。

automl = AutoML()
automl.add_learner(learner_name='RGF', learner_class=MyRegularizedGreedyForest)
# 增加配置
settings = {
    "time_budget": 10,  # total running time in seconds
    "metric": 'accuracy', 
    "estimator_list": ['RGF', 'lgbm', 'rf', 'xgboost'],  # list of ML learners
    "task": 'classification',  # task type    
    "log_file_name": 'airlines_experiment_custom_learner.log',  # flaml log file 
    "log_training_metric": True,  # whether to log training metric
}
automl.fit(X_train = X_train, y_train = y_train, **settings)

(3) 自定义优化指标

咱们能够为模型自定义优化指标。上面的示例代码中，咱们合并训练损失和验证损失作为自定义优化指标，并对其进行优化，心愿损失最小化。

def custom_metric(X_val, y_val, estimator, labels, X_train, y_train,
                  weight_val=None, weight_train=None, config=None,
                  groups_val=None, groups_train=None):
    from sklearn.metrics import log_loss
    import time
    start = time.time()
    y_pred = estimator.predict_proba(X_val)
    pred_time = (time.time() - start) / len(X_val)
    val_loss = log_loss(y_val, y_pred, labels=labels,
                         sample_weight=weight_val)
    y_pred = estimator.predict_proba(X_train)
    train_loss = log_loss(y_train, y_pred, labels=labels,
                          sample_weight=weight_train)
    alpha = 0.5
    return val_loss * (1 + alpha) - alpha * train_loss, {"val_loss": val_loss, "train_loss": train_loss, "pred_time": pred_time}
    # two elements are returned:
    # the first element is the metric to minimize as a float number,
    # the second element is a dictionary of the metrics to log
automl = AutoML()
settings = {
    "time_budget": 10,  # total running time in seconds
    "metric": custom_metric,  # pass the custom metric funtion here
    "task": 'classification',  # task type
    "log_file_name": 'airlines_experiment_custom_metric.log',  # flaml log file
}
automl.fit(X_train = X_train, y_train = y_train, **settings)

3.3 sklearn 流水线调优

FLAML 能够配合 sklearn pipeline 进行模型自动化调优，咱们这里仍旧以航空公司数据集 Airlines dataset 案例为场景，对其用法做一个解说。

(1) 加载数据集

# 数据集筹备
from flaml.data import load_openml_dataset
X_train, X_test, y_train, y_test = load_openml_dataset(dataset_id=1169, data_dir='./', random_state=1234, dataset_format='array')

(2) 构建建模流水线

import sklearn
from sklearn import set_config
from sklearn.pipeline import Pipeline
from sklearn.impute import SimpleImputer
from sklearn.preprocessing import StandardScaler
from flaml import AutoML
set_config(display='diagram')
imputer = SimpleImputer()
standardizer = StandardScaler()
automl = AutoML()
automl_pipeline = Pipeline([("imputuer",imputer),
    ("standardizer", standardizer),
    ("automl", automl)
])
automl_pipeline

输入后果如下

Pipeline(steps=[('imputuer', SimpleImputer()),
                ('standardizer', StandardScaler()),
                ('automl',)])
SimpleImputerSimpleImputer()
StandardScalerStandardScaler()
AutoML

(3) 参数设定与 automl 拟合

# 设定
settings = {
    "time_budget": 60,  # 总时长束缚
    "metric": 'accuracy',  # 可选: ['accuracy','roc_auc', 'roc_auc_ovr', 'roc_auc_ovo', 'f1','log_loss','mae','mse','r2']
    "task": 'classification',  # 工作类型
    "estimator_list":['xgboost','catboost','lgbm'],
    "log_file_name": 'airlines_experiment.log',  # flaml 日志文件
}

# 拟合
automl_pipeline.fit(X_train, y_train, 
                        automl__time_budget=settings['time_budget'],
                        automl__metric=settings['metric'],
                        automl__estimator_list=settings['estimator_list'],
                        automl__log_training_metric=True)

(4) 取出最优模型

# Get the automl object from the pipeline
automl = automl_pipeline.steps[2][1]
# Get the best config and best learner
print('Best ML leaner:', automl.best_estimator)
print('Best hyperparmeter config:', automl.best_config)
print('Best accuracy on validation data: {0:.4g}'.format(1-automl.best_loss))
print('Training duration of best run: {0:.4g} s'.format(automl.best_config_train_time))
automl.model

运行后果如下：

Best ML leaner: xgboost
Best hyperparmeter config: {'n_estimators': 63, 'max_leaves': 1797, 'min_child_weight': 0.07275175679381725, 'learning_rate': 0.06234183309508761, 'subsample': 0.9814772488195874, 'colsample_bylevel': 0.810466508891351, 'colsample_bytree': 0.8005378817953572, 'reg_alpha': 0.5768305704485758, 'reg_lambda': 6.867180836557797, 'FLAML_sample_size': 364083}
Best accuracy on validation data: 0.6721
Training duration of best run: 15.45 s
<flaml.model.XGBoostSklearnEstimator at 0x7f03a5eada00>

(5) 测试集评估与模型存储

import pickle
with open('automl.pkl', 'wb') as f:
    pickle.dump(automl, f, pickle.HIGHEST_PROTOCOL)
# 测试集预估与成果评估
y_pred = automl_pipeline.predict(X_test)
print('Predicted labels', y_pred)
print('True labels', y_test)
y_pred_proba = automl_pipeline.predict_proba(X_test)[:,1]
print('Predicted probas',y_pred_proba[:5])

运行后果如下

Predicted labels [0 1 1 ... 0 1 0]
True labels [0 0 0 ... 1 0 1]
Predicted probas  [0.3764987  0.6126277  0.699604   0.27359942 0.25294745]

3.4 XGBoost 主动调优

这里咱们简略给大家讲一下如何应用 FLAML 调优最常见的模型之一 XGBoost。

(1) 工具库导入与根本设定

# 导入工具库
from flaml import AutoML
automl = AutoML()
# 参数设定
settings = {
    "time_budget": 120,  # total running time in seconds
    "metric": 'r2',  # primary metrics for regression can be chosen from: ['mae','mse','r2','rmse','mape']
    "estimator_list": ['xgboost'],  # list of ML learners; we tune xgboost in this example
    "task": 'regression',  # task type    
    "log_file_name": 'houses_experiment.log',  # flaml log file
}

(2) 自动化机器学习拟合

automl.fit(X_train=X_train, y_train=y_train, **settings)

(3) 最优模型与评估

咱们能够输入最优模型配置及详细信息

# 最优模型
print('Best hyperparmeter config:', automl.best_config)
print('Best r2 on validation data: {0:.4g}'.format(1 - automl.best_loss))
print('Training duration of best run: {0:.4g} s'.format(automl.best_config_train_time))

运行后果：

Best hyperparmeter config: {'n_estimators': 776, 'max_leaves': 160, 'min_child_weight': 32.57408640781376, 'learning_rate': 0.034786853332414935, 'subsample': 0.9152991332236934, 'colsample_bylevel': 0.5656764254642628, 'colsample_bytree': 0.7313266091895249, 'reg_alpha': 0.005771390107656191, 'reg_lambda': 1.49126672786588}
Best r2 on validation data: 0.834
Training duration of best run: 9.471 s

咱们能够取出最优模型

automl.model.estimator

后果如下：

XGBRegressor(base_score=0.5, booster='gbtree',
             colsample_bylevel=0.5656764254642628, colsample_bynode=1,
             colsample_bytree=0.7313266091895249, gamma=0, gpu_id=-1,
             grow_policy='lossguide', importance_type='gain',
             interaction_constraints='', learning_rate=0.034786853332414935,
             max_delta_step=0, max_depth=0, max_leaves=160,
             min_child_weight=32.57408640781376, missing=nan,
             monotone_constraints='()', n_estimators=776, n_jobs=-1,
             num_parallel_tree=1, random_state=0,
             reg_alpha=0.005771390107656191, reg_lambda=1.49126672786588,
             scale_pos_weight=1, subsample=0.9152991332236934,
             tree_method='hist', use_label_encoder=False, validate_parameters=1,
             verbosity=0)

同样能够对 XGBoost 模型绘制特色重要度

import matplotlib.pyplot as plt
plt.barh(X_train.columns, automl.model.estimator.feature_importances_)

(4) 模型存储

import pickle
with open('automl.pkl', 'wb') as f:
    pickle.dump(automl, f, pickle.HIGHEST_PROTOCOL)

(5) 测试集预估及模型评估

# 测试集预估
y_pred = automl.predict(X_test)
print('Predicted labels', y_pred)
print('True labels', y_test)

# 测试集评估
from flaml.ml import sklearn_metric_loss_score
print('r2', '=', 1 - sklearn_metric_loss_score('r2', y_pred, y_test))
print('mse', '=', sklearn_metric_loss_score('mse', y_pred, y_test))
print('mae', '=', sklearn_metric_loss_score('mae', y_pred, y_test))

3.5 LightGBM 主动调优

LightGBM 调优的过程和 XGBoost 十分相似，仅仅在参数配置的局部指定模型须要做一点调整，其余局部是统一的，如下：

# 导入工具库
from flaml import AutoML
automl = AutoML()

# 参数配置
settings = {
    "time_budget": 240,  # total running time in seconds
    "metric": 'r2',  # primary metrics for regression can be chosen from: ['mae','mse','r2','rmse','mape']
    "estimator_list": ['lgbm'],  # list of ML learners; we tune lightgbm in this example
    "task": 'regression',  # task type    
    "log_file_name": 'houses_experiment.log',  # flaml log file
    "seed": 7654321,    # random seed
}

# 自动化机器学习拟合调优
automl.fit(X_train=X_train, y_train=y_train, **settings)

参考资料

  • 图解机器学习算法 | 从入门到精通系列
  • Airlines dataset
  • notebook 代码示例

ShowMeAI 系列教程举荐

  • 图解 Python 编程：从入门到精通系列教程
  • 图解数据分析：从入门到精通系列教程
  • 图解 AI 数学根底：从入门到精通系列教程
  • 图解大数据技术：从入门到精通系列教程
  • 图解机器学习算法：从入门到精通系列教程
  • 机器学习实战：手把手教你玩转机器学习系列

相干文章举荐

  • Python 机器学习算法利用实际
  • SKLearn 入门与简略利用案例
  • SKLearn 最全利用指南
  • XGBoost 建模利用详解
  • LightGBM 建模利用详解
  • Python 机器学习综合我的项目 - 电商销量预估
  • Python 机器学习综合我的项目 - 电商销量预估 < 进阶计划 >
  • 机器学习特色工程最全解读
  • 自动化特色工程工具 Featuretools 利用
  • AutoML 自动化机器学习建模

正文完
机器学习
发表至： 机器学习
2022-03-22
 0
关于机器学习:系统召回太慢上-Milvus-×-PaddleRec-双剑合璧大法
关于机器学习:GID旷视提出全方位的检测模型知识蒸馏-CVPR-2021
关于机器学习:瑞利商Rayleigh-quotient与广义瑞利商genralized-Rayleigh-quotient
关于机器学习:吴恩达机器学习逻辑回归的代价函数06