The number of rank is larger than the total number of sample, after running the rank_results function. Here is how to reproduce the bug:
library(benchmarkme)
res_cpu <- benchmark_std(runs = 3)
rank_results(res_cpu)
# You are ranked 70 out of 45 machines. (the number may vary across different machines)
A closer look at the rank_results function reveals that the bug lies in the last few lines:
if (verbose)
message("You are ranked ", ben_rank, " out of ", nrow(results),
" machines.")
where nrow(results) should be nrow(pas_res).
Another question I want to confirm is that does the past performance dataset benchmarkmeData::past_results change/accumulate as time passes by? If so, then perhaps I may yield different benchmarks for the same machine?
The number of rank is larger than the total number of sample, after running the
rank_results
function. Here is how to reproduce the bug:A closer look at the
rank_results
function reveals that the bug lies in the last few lines:where
nrow(results)
should benrow(pas_res)
.Another question I want to confirm is that does the past performance dataset
benchmarkmeData::past_results
change/accumulate as time passes by? If so, then perhaps I may yield different benchmarks for the same machine?Thank you. Yu