- Apr 29, ‘24:
1) Reviewed PR 6102, found a solution to the error with pkg.edit.fun not being detected (out of scope).
2) Discussed with Michael how to fix the git2r::revparse issue (with atime + GHA) for an invalid commit SHA. (6110)
- Apr 30, ‘24:
1) Discussed with Toby the tasks to work on and changes to be made.
2) Created the data.table.threads repository and an issue with my current benchmarking code for finding the optimal thread count and times for the different functions in data.table that are parallelizable.
3) Changed the format of this work (b)log to a weekly basis for readability.
4) Trying to get the combined speedup plot to work.
- May 1, ‘24:
1) Finished the speedup plot generating code, see #2 (not the most ideal speedup as you can see the deviation from the dotted line I added to those subplots too, but that’s how it is for repeated runs too). Refactored my old code twice and used a much more simpler approach. Also researched and learned how to create multiple plots in a better way. Will update codeowners and check on issues tomorrow.
- May 2, ‘24:
1) Onto creating an R package for data.table.threads.
2) Went back and commented/closed some old issues.
3) Updated CODEOWNERS (GHA-based performance testing).
- Mat 3, ‘24:
1) Made Toby’s suggestions to the speedup plot gen code.
2) Currently debugging and trying to fix the issues when building my package.
3) Also gotta make changes to my presentation for next week. I’ll have to go through it over the weekend.