biscuit
Veteran
How exactly would you do that? I kinda get the concept, but not quite.
Overland Park Open 2017
- Thread starter Tycubing
- Start date
How exactly would you do that? I kinda get the concept, but not quite.
AlphaSheep
Veteran
Most programming languages have libraries that can do it for you. Python is the one I'm most familiar with:
Code:
from numpy import mean, std
from random import gauss
times = [10.04, 12.53, 11.17, 10.07, 12.02] # Random times from someone's profile
newTime = gauss(mean(times), std(times)) # A randomly generated time
JustinTimeCuber
Legend
Problem with that is you could theoretically generate negative times. Is there a good way to stop that from happening? I know when I solve the histogram is slanted to the left, i.e. the mean is greater than the median.
I think my original plan will work decently because it will have a reasonable sample (for me, 45 solves) and it would have a slight random factor to make there be some smoothness. Not that it couldn't be better though, which it definitely could.
I think my original plan will work decently because it will have a reasonable sample (for me, 45 solves) and it would have a slight random factor to make there be some smoothness. Not that it couldn't be better though, which it definitely could.
Last edited:
AlphaSheep
Veteran
You get a truncated normal distribution to solve that problem... But there are advantages to keeping it simple too. It's only for fun anyway
biscuit
Veteran
You get a truncated normal distribution to solve that problem... But there are advantages to keeping it simple too. It's only for fun anyway
For now... If this works well I plan to steal the idea and start a gambling empire based on cubing! MWAHAHAHAHAHAHAHA
JustinTimeCuber
Legend
okay then I'm not letting you see the code, sorry
biscuit
Veteran
okay then I'm not letting you see the code, sorry
NOOOOOOO! Fine, I'll just have to make my own version from scratch (which I was already thinking about making a website version that would work for any competition.)
JustinTimeCuber
Legend
that would actually be really awesome
but yeah, I'm still posting the code, hopefully I can finish the main parts tonight
I am testing it on data from 5 people, see if you can guess the people or something. Hints: They are all registered for the competition, and exactly four of them have posted in this thread.
Last edited:
biscuit
Veteran
Oh, oh, is one me?(!)
JustinTimeCuber
Legend
aybuck37
Veteran
Before we do all the computer coding. We should fix the cubingusa psych sheet
JustinTimeCuber
Legend
CubingUSA should just stop doing registrations, the WCA does it way better
aybuck37
Veteran
Haha I think I'm in the simulation
JustinTimeCuber
Legend
Everyone will be in the final thing, but yeah, you're in my test data lolHaha I think I'm in the simulation
biscuit
Veteran
Here's my guess.
Christopher Olson
Justin Barker
Ashner Buckner
Tyler Specht
Me
Christopher Olson
Justin Barker
Ashner Buckner
Tyler Specht
Me
JustinTimeCuber
Legend
winner winner
now we can move on to the more important aspects of life, including who can manipulate a plastic cube so that all the sides have the same colored sticker the fastest.
biscuit
Veteran
And will Justin make an accurate simulation?
JustinTimeCuber
Legend
hopefully
note: it takes like 3 minutes per person to enter in the times.
biscuit
Veteran
JustinTimeCuber
Legend
@biscuit Alright, here's the code so far. It defines some stuff and does a bit of setup, eh, okay, just try to read this tangled mess (sorry)
https://pastebin.com/8XXU8Geh
Still need to get it to create the weight for each competition, that's my next order of business. For now, because I haven't finished all the input data (yeah I really need to learn how to do that better) it does what it's supposed to, outputs this
and closes.
e: BBcode decided to interpret all the [i ]s in the code as italics, so they are all gone.
https://pastebin.com/8XXU8Geh
Still need to get it to create the weight for each competition, that's my next order of business. For now, because I haven't finished all the input data (yeah I really need to learn how to do that better) it does what it's supposed to, outputs this
Code:
Empty file for 2015WEBS01
Empty file for 2015BUCK03
Empty file for 2015SPEC01e: BBcode decided to interpret all the [i ]s in the code as italics, so they are all gone.
Last edited:
Similar threads
