怎么做跳转网站,内蒙古 网站建设,央美老师做的家具网站,海口省建设厅网站1. Optuna库的优势 对比bayes_opt和hyperoptOptuna不仅可以衔接到PyTorch等深度学习框架上#xff0c;还可以与sklearn-optimize结合使用#xff0c;这也是我最喜欢的地方#xff0c;Optuna因此特性可以被使用于各种各样的优化场景。 2. 导入必要的库及加载数据 用的是sklea…1. Optuna库的优势 对比bayes_opt和hyperoptOptuna不仅可以衔接到PyTorch等深度学习框架上还可以与sklearn-optimize结合使用这也是我最喜欢的地方Optuna因此特性可以被使用于各种各样的优化场景。 2. 导入必要的库及加载数据 用的是sklearn自带的房价数据只是我把它保存下来了。
import optuna
import pandas as pd
import numpy as np
from sklearn.model_selection import KFold,cross_validate
print(optuna.__version__)
from sklearn.ensemble import RandomForestRegressor as RFR
data pd.read_csv(rD:\2暂存文件\Sth with Py\贝叶斯优化\data.csv)
X data.iloc[:,0:8]
y data.iloc[:,8]
3. 定义目标函数与参数空间 Optuna相对于其他库不需要单独输入参数或参数空间只需要直接在目标函数中定义参数空间即可。这里以负均方误差为损失函数。
def optuna_objective(trial) :# 定义参数空间n_estimators trial.suggest_int(n_estimators,10,100,1)max_depth trial.suggest_int(max_depth,10,50,1)max_features trial.suggest_int(max_features,10,30,1)min_impurtity_decrease trial.suggest_float(min_impurity_decrease,0.0, 5.0, step0.1)# 定义评估器reg RFR(n_estimatorsn_estimators,max_depthmax_depth,max_featuresmax_features,min_impurity_decreasemin_impurtity_decrease,random_state1412,verboseFalse,n_jobs-1)# 定义交叉过程输出负均方误差cv KFold(n_splits5,shuffleTrue,random_state1412)validation_loss cross_validate(reg,X,y,scoringneg_mean_squared_error,cvcv,verboseTrue,n_jobs-1,error_scoreraise)return np.mean(validation_loss[test_score])
4. 定义优化目标函数 在Optuna中我们可以调用sampler模块进行选用想要的优化算法比如TPE、GP等等。
def optimizer_optuna(n_trials,algo):# 定义使用TPE或GPif algo TPE:algo optuna.samplers.TPESampler(n_startup_trials20,n_ei_candidates30)elif algo GP:from optuna.integration import SkoptSamplerimport skoptalgo SkoptSampler(skopt_kwargs{base_estimator:GP,n_initial_points:10,acq_func:EI})study optuna.create_study(sampleralgo,directionmaximize)study.optimize(optuna_objective,n_trialsn_trials,show_progress_barTrue)print(best_params:,study.best_trial.params,best_score:,study.best_trial.values,\n)return study.best_trial.params, study.best_trial.values
5. 执行部分
import warnings
warnings.filterwarnings(ignore,messageThe objective has been evaluated at this point before trails)
optuna.logging.set_verbosity(optuna.logging.ERROR)
best_params, best_score optimizer_optuna(200,TPE)
6. 完整代码
import optuna
import pandas as pd
import numpy as np
from sklearn.model_selection import KFold,cross_validate
print(optuna.__version__)
from sklearn.ensemble import RandomForestRegressor as RFRdata pd.read_csv(rD:\2暂存文件\Sth with Py\贝叶斯优化\data.csv)
X data.iloc[:,0:8]
y data.iloc[:,8]def optuna_objective(trial) :# 定义参数空间n_estimators trial.suggest_int(n_estimators,10,100,1)max_depth trial.suggest_int(max_depth,10,50,1)max_features trial.suggest_int(max_features,10,30,1)min_impurtity_decrease trial.suggest_float(min_impurity_decrease,0.0, 5.0, step0.1)# 定义评估器reg RFR(n_estimatorsn_estimators,max_depthmax_depth,max_featuresmax_features,min_impurity_decreasemin_impurtity_decrease,random_state1412,verboseFalse,n_jobs-1)# 定义交叉过程输出负均方误差cv KFold(n_splits5,shuffleTrue,random_state1412)validation_loss cross_validate(reg,X,y,scoringneg_mean_squared_error,cvcv,verboseTrue,n_jobs-1,error_scoreraise)return np.mean(validation_loss[test_score])def optimizer_optuna(n_trials,algo):# 定义使用TPE或GPif algo TPE:algo optuna.samplers.TPESampler(n_startup_trials20,n_ei_candidates30)elif algo GP:from optuna.integration import SkoptSamplerimport skoptalgo SkoptSampler(skopt_kwargs{base_estimator:GP,n_initial_points:10,acq_func:EI})study optuna.create_study(sampleralgo,directionmaximize)study.optimize(optuna_objective,n_trialsn_trials,show_progress_barTrue)print(best_params:,study.best_trial.params,best_score:,study.best_trial.values,\n)return study.best_trial.params, study.best_trial.valuesimport warnings
warnings.filterwarnings(ignore,messageThe objective has been evaluated at this point before trails)
optuna.logging.set_verbosity(optuna.logging.ERROR)
best_params, best_score optimizer_optuna(200,TPE)
