1
0
mirror of https://github.com/msberends/AMR.git synced 2025-07-13 02:32:07 +02:00

0.4.0, just published on CRAN

This commit is contained in:
2018-10-01 19:36:42 +02:00
parent 6d761436f7
commit c7c57f4042
3 changed files with 5 additions and 5 deletions

View File

@ -472,7 +472,7 @@ microbenchmark(A = as.mo(repetive_results),
unit = "ms")
# Unit: milliseconds
# expr min lq mean median uq max neval
# A 14.61282 14.6372 14.70817 14.72597 14.76124 14.78498 1
# A 14.61282 14.6372 14.70817 14.72597 14.76124 14.78498 10
```
So transforming 25,000 times (!) `"Staphylococcus aureus"` only takes 4 ms (0.004 seconds) more than transforming it once. You only lose time on your unique input values.
@ -489,7 +489,7 @@ microbenchmark(A = mo_fullname("B_STPHY_AUR"),
# expr min lq mean median uq max neval
# A 13.548652 13.74588 13.8052969 13.813594 13.881165 14.090969 10
# B 15.079781 15.16785 15.3835842 15.374477 15.395115 16.072995 10
# C 0.171182 0.185639 0.2306307 0.2034135 0.224610 0.492312 10
# C 0.171182 0.18563 0.2306307 0.203413 0.224610 0.492312 10
```
So going from `mo_fullname("Staphylococcus aureus")` to `"Staphylococcus aureus"` takes 0.0002 seconds - it doesn't even start calculating *if the result would be the same as the expected resulting value*. That goes for all helper functions: