Closed trivialfis closed 5 years ago
Pandas join benchmark is out of date. Initially, in 2016, this benchmark was testing mostly distributed tools + pandas, data.table, dplyr. If you are interested in those you may want to check this talk https://www.youtube.com/watch?v=5X7h1rZGVs0
Join scripts are left from those old times, except for the data.table, where I am drafting new join benchmark now. The old was very limited. Up to date info on join is in https://github.com/h2oai/db-benchmark/issues/18.
I am aware of gc
, it is the same in R, but for consistency is included in various tools.
Using wrapper functions in benchmark scripts was discussed in https://github.com/h2oai/db-benchmark/pull/41
Closing this, please discuss the feature/changes you want to push to avoid unnecessary work. Thanks for contributing!
@jangorecki Thanks for the explanation.