Timeline for Finding the period of an array of integers
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 19, 2015 at 3:13 | comment | added | VividD | Thanks, performance comparison is beautiful, and also valuable info. | |
| Apr 18, 2015 at 18:36 | comment | added | 2012rcampion |
@rasher Also, the performance difference is exactly zero for lists that do have a repeat (excepting a difference of one clock cycle for the /2/>>1), and the difference is around 2x for (random) lists with no repeats.
|
|
| Apr 18, 2015 at 18:27 | comment | added | 2012rcampion |
@rasher I would, if not for the comments on Pickett's answer: "{1, 2, 7, 3, 5, 1} should be counted as one pattern of length six, instead of a pattern of length five" "it is not so critical to make such distinction or not, but I would think your way"
|
|
| Apr 18, 2015 at 18:17 | comment | added | ciao | Please note that in your answer so readers can't miss it, and update benchmark to show the significant performance impact this change has. | |
| Apr 18, 2015 at 18:04 | comment | added | 2012rcampion |
@rasher change Quotient[n,2] to n to get that behavior.
|
|
| Apr 18, 2015 at 17:54 | comment | added | ciao |
Fast is nice, when it's fast and correct. This fails, e.g. {1,2,3,1,2} should return 3 per OP, this returns 5.
|
|
| Apr 18, 2015 at 16:43 | history | answered | 2012rcampion | CC BY-SA 3.0 |