For each of the following topics, we'll spend some time reading and running the code independently, then come back together and discuss.
- Readability
- What makes this code easy or hard to understand?
- How could this code be more readable?
- Effectiveness:
- How would you go about testing that this code does what it's supposed to?
- How could your reorganize the code to make that testing more straight forward?
- Efficiency:
- What resource is the bottleneck in this code? (CPU? Memory? Something else?)
- Which parts of the code should we focus on to improve efficiency of that resource?
- What strategies could we use to make those parts run faster?
The script getting the code review was 08_comoving_analysis_AA\ rewrite2.R, subset.R was used to generate a subset of the data suitable for testing, all other scripts are possible rewrites and tests for different portions generated during the workshop.
There is also a branch of this repo perf-improvement that has some suggestions for the code.