从 explain_tidymodels 更改颜色图

Question

创建以下 explain_tidymodels，以显示部分依赖图。

explainer <- explain_tidymodels(rf_vi_fit, data = Data_train, y = Data_train$Lead_week)

现在我正在通过执行以下操作创建地块：

model_profile(explainer, variables = c("Month", "AC")) %>% plot()

它会自动给我蓝色的图。但是，如何将绘图颜色更改为红色？我已经尝试过 %>% plot(color = "red") 和 %>% plot(col = "red") 之类的东西，但都不起作用。

有什么建议可以将部分依赖图更改为首选颜色吗？提前致谢！

Answer 1

我建议您访问基础数据并使用 ggplot2 制作您自己的图，就像我在这里展示的那样：

library(tidymodels)
#> Registered S3 method overwritten by 'tune':
#>   method                   from   
#>   required_pkgs.model_spec parsnip
library(DALEXtra)
#> Loading required package: DALEX
#> Welcome to DALEX (version: 2.2.0).
#> Find examples and detailed introduction at: http://ema.drwhy.ai/
#> Additional features will be available after installation of: ggpubr.
#> Use 'install_dependencies()' to get all suggested dependencies
#> 
#> Attaching package: 'DALEX'
#> The following object is masked from 'package:dplyr':
#> 
#>     explain

data(ames)
ames_train <- ames %>%
    transmute(Sale_Price = log10(Sale_Price),
              Gr_Liv_Area = as.numeric(Gr_Liv_Area), 
              Year_Built, Bldg_Type)

rf_model <- 
    rand_forest(trees = 1000) %>% 
    set_engine("ranger") %>% 
    set_mode("regression")

rf_wflow <- 
    workflow() %>% 
    add_formula(
        Sale_Price ~ Gr_Liv_Area + Year_Built + Bldg_Type) %>% 
    add_model(rf_model) 

rf_fit <- rf_wflow %>% fit(data = ames_train)
explainer_rf <- explain_tidymodels(
    rf_fit, 
    data = dplyr::select(ames_train, -Sale_Price), 
    y = ames_train$Sale_Price,
    label = "random forest"
)
#> Preparation of a new explainer is initiated
#>   -> model label       :  random forest 
#>   -> data              :  2930  rows  3  cols 
#>   -> data              :  tibble converted into a data.frame 
#>   -> target variable   :  2930  values 
#>   -> predict function  :  yhat.workflow  will be used ( [33m default [39m )
#>   -> predicted values  :  No value for predict function target column. ( [33m default [39m )
#>   -> model_info        :  package tidymodels , ver. 0.1.3 , task regression ( [33m default [39m ) 
#>   -> predicted values  :  numerical, min =  4.896787 , mean =  5.220582 , max =  5.51655  
#>   -> residual function :  difference between y and yhat ( [33m default [39m )
#>   -> residuals         :  numerical, min =  -0.8021289 , mean =  5.872977e-05 , max =  0.3613971  
#>  [32m A new explainer has been created! [39m

pdp_rf <- model_profile(explainer_rf, N = NULL, 
                        variables = "Gr_Liv_Area", groups = "Bldg_Type")

as_tibble(pdp_rf$agr_profiles) %>%
    mutate(`_label_` = stringr::str_remove(`_label_`, "random forest_")) %>%
    ggplot(aes(`_x_`, `_yhat_`, color = `_label_`)) +
    geom_line(size = 1.2, alpha = 0.8) +
    labs(x = "Gross living area", 
         y = "Sale Price (log)", 
         color = NULL,
         title = "Partial dependence profile for Ames housing sales",
         subtitle = "Predictions from a random forest model")

^{由 reprex package (v2.0.0)}

于 2021-05-27 创建

您还可以访问pdp_rf$cp_profiles。

从 explain_tidymodels 更改颜色图

Change colour plot from explain_tidymodels

r

plot

tidymodels