salbalkus / CausalTables.jl

A new type of Table to store and simulate data for causal inference in Julia.
https://salbalkus.github.io/CausalTables.jl/
MIT License
7 stars 0 forks source link

Initial Code Review #14

Closed salbalkus closed 6 months ago

salbalkus commented 9 months ago

@nhejazi

The code is ready for your initial review. Three aspects need review:

  1. src: The code of the package. It can be divided into three major parts: Wrapping causal data for input into other models (i.e a modified treatment policy), drawing randomly from a data generating process, and computing ground truth conditional densities. It would be great if you could suggest any potential changes to the API that might make it easier to use. Are there places where I could improve the code itself? Also, are there any places I could improve code readability or add new features or function parameters?
  2. test: I've written unit tests to evaluate the code. I do plan on adding more to improve coverage, so here I'm mostly looking for stylistic edits. Where can major changes be made to support readability? Are there places where it might be good to simplify the testing process?
  3. docs: Probably the most important area. I would love an extra eye on the documentation to evaluate, from the perspective of someone who's never used the package, where improvements to clarity could be made. Are there places where the explanation is difficult to understand? Are additional vignettes necessary to answer questions that new users might have? Should I change the organizational structure or layout anywhere to be clearer?

Thank you!