ideas

[ShixiangWang]

link to https://github.com/ShixiangWang/ezcox/issues/21

定义Model和Batcher两个对象

model分为数据，公式，focal term，covars，strates等

To be rethink

表格展示

• https://github.com/davidgohel/flextable

R6

• By default, R6 objects have a clone() method, which is a fairly large function. If you do not need this feature, you can save some memory by using cloneable=FALSE.

• By default, a class attribute is added to R6 objects. This attribute adds a slight performance penalty because R will attempt to use S3 dispatch when using $ on the object.

可视化

• https://github.com/pbreheny/visreg
• https://github.com/IndrajeetPatil/ggstatsplot#ggcoefstats

[ShixiangWang]

https://strengejacke.github.io/ggeffects/articles/introduction_marginal_effects.html

[ShixiangWang]

https://github.com/adayim/forestploter

[ShixiangWang]

> ggeffects::ggpredict(mm$model, "x")
Error in names(datlist) <- names(focal_terms) : 
  'names' attribute [1] must be the same length as the vector [0]
In addition: Warning message:
Could not get model data. 
> ggeffects::ggpredict(mm$model, terms = "x")
Error in names(datlist) <- names(focal_terms) : 
  'names' attribute [1] must be the same length as the vector [0]
In addition: Warning message:
Could not get model data. 

好像不兼容。模型本身的数据访问可能存在问题，这跟函数通过代码自动构建有关系

[ShixiangWang]

交互项可视化还不支持：

library(regport)
library(survival)

lung = survival::lung
lung$sex = factor(lung$sex)

model = REGModel$new(
  lung,
  recipe = Surv(time, status) ~ age * sex
)

model

model$get_forest_data()
model$plot_forest()

[ShixiangWang]

• [x] $build()并行支持