fix: SC_apply() The selected columns were offset.

[shota109-eng]

Hello, I'm a university student using pyOMEA2 for my study and gratefull for your works!

I found a bug in pyoma2.functions.gen.SC_apply().

The following shows what my PR changes the current condition.

---

import numpy as np
import math

ordmin = 2
ordmax = 28
step = 3

#model orders
arr = np.arange(0, int(ordmax / step + 1)) * step

print(arr)
#[ 0 3 6 9 12 15 18 21 24 27]

#old
for oo in range(ordmin, ordmax + 1, step):
o = int(oo / step - 1)
print(arr[o])
#0 0 3 6 9 12 15 18 21

#new
for o in range(math.ceil(ordmin / step), math.floor(ordmax / step) + 1):
print(arr[o])
#3 6 9 12 15 18 21 24 27

---

import numpy as np
import math

ordmin = 0
ordmax = 5
step = 1

#model orders
arr = np.arange(0, int(ordmax / step + 1)) * step

print(arr)
#[0 1 2 3 4 5]

#old
for oo in range(ordmin, ordmax + 1, step):
o = int(oo / step - 1)
print(arr[o])
#5 0 1 2 3 4

#new
for o in range(math.ceil(ordmin / step), math.floor(ordmax / step) + 1):
print(arr[o])
#0 1 2 3 4 5

[Copilot]

Pull request overview

This PR fixes a critical bug in the SC_apply() function where array column indices were incorrectly calculated, causing the function to access wrong columns when processing model orders. The fix replaces the flawed index calculation with a mathematically correct approach that properly maps model orders to array column indices.

Key Changes:

• Corrected the loop iteration logic to properly calculate array column indices from model orders
• Added the math module import to support the new calculation using ceil() and floor() functions

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[dfm88]

@shota109-eng thank for your PR, makes sense to me, just asking a confirm to @dagghe for merging

[shota109-eng]

Thanks!

[dfm88]

great all tests have passed, think we can merge this!