In working with Random, it seems to be biased toward generating values in the middle of the range rather than the extremes.
example: I’m asking Random to generate 12000 values within a range from 0-4 and there are about half as many 0’s and 4’s as there are 1,2,3
Thoughts on the most simple way to get random values that are more evenly distributed across a range?
The alternative that I am currently using is to generate a list of values with the desired distribution and using the shuffle component to randomize their location in the list.
I would still like to better understand the distribution issue that I am having with the random component.
I think that is because of rounding.
0-0.5 are counted as 0,
0.5-1.5 are counted as 1,
1.5-2.5 -> 2
2.5-3.5 -> 3
3.5-4 -> 4
maybe random from 0 to 5 with 2 decimal places and floor the rounding?
Although if you look at it, the 1500 changes between the different seeds.
Yeah it is indeed a rounding problem, but I couldn’t fix it without breaking existing files.
I’m afraid you’ll need to set the domain to -/+ half of the integer range.
