Code for Quantitative Assessment of Factors Influencing Heat Vulnerability in Residential Areas using Machine Learning and UAV Data
收藏资源简介:
Author: Ja Woon Gu Email: umseakind2@kwater.or.kr Date: 2024-07-22 Version: 1.0 Description: This script compares multiple regression models using GridSearchCV and RepeatedKFold cross-validation. The script identifies the best performing model based on the mean cross-validation score (neg_mean_squared_error). It performs residual analysis including normality, homoscedasticity, and autocorrelation checks. Models included: - DecisionTreeRegressor - ExtraTreesRegressor - AdaBoostRegressor - XGBRegressor - LGBMRegressor - CatBoostRegressor - RandomForestRegressor - GradientBoostingRegressor The best model's feature importances are calculated and plotted. Residuals are analyzed using Shapiro-Wilk, Levene's, and Durbin-Watson tests.
作者:Ja Woon Gu 邮箱:umseakind2@kwater.or.kr 日期:2024-07-22 版本:1.0 描述: 本脚本通过网格搜索交叉验证(GridSearchCV)与重复K折交叉验证(RepeatedKFold)对多种回归模型进行对比。 本脚本基于平均交叉验证得分(负均方误差,neg_mean_squared_error)选出性能最优的模型。 该脚本将执行残差分析,涵盖正态性、同方差性与自相关性检验。 包含的模型如下: - 决策树回归器(DecisionTreeRegressor) - 极端随机树回归器(ExtraTreesRegressor) - AdaBoost回归器(AdaBoostRegressor) - XGB回归器(XGBRegressor) - LGBM回归器(LGBMRegressor) - CatBoost回归器(CatBoostRegressor) - 随机森林回归器(RandomForestRegressor) - 梯度提升回归器(GradientBoostingRegressor) 将计算并绘制最优模型的特征重要性。 本脚本将使用夏皮罗-威尔克检验(Shapiro-Wilk test)、莱文检验(Levene's test)与杜宾-瓦特森检验(Durbin-Watson test)对残差进行分析。



