Merge branch 'TopolOptimUpdate' into 'main'

Minor update of some of the optimisation functions

See merge request in-nova/in-nova!11

CHANGELOG.md

+### v0.4.1 ###
+** Optimisation - Memetic Algorithm**\
+\* Update `mutPop()` to use the parallel for-loop (`parfor`) whenever it is available.\
+\* Update `selParents()` to use the parallel for-loop (`parfor`) whenever it is available.
+
+-------------------------------------------
+
+
 ### v0.4 ###
 **Utilities - Generic**\
 \* Change the interface of `rotMat3d()`, so that the last argument is a boolean (`logical` in MATLAB), declaring whether the given angles are in degrees (if false, the angles are treated as radians). This is a **backwards incompatible** change.\

Optimisation/Memetic Algorithm/MATLAB/Functions/mutPop.m

@@ -3,7 +3,7 @@
 % Author: Achilles Kappis
 % e-mail: axilleaz@protonmail.com
 %
-% Date: 05/12/2023 (DD/MM/YYYY)
+% Date: 21/11/2024 (DD/MM/YYYY)
 %
 % Copyright: MIT
 % --------------------------------------------------
@@ -182,7 +182,7 @@ function muts = mutPop(pop, mutPerc, mutProb, mutVals, uniqueMuts)
     % ====================================================
     % Create probabilities for the variables to be mutated
     if isscalar(mutProb)
-        for mutIdx = nMuts:-1:1
+        parfor mutIdx = 1:nMuts
             if mutProb < 0
                 % Make sure the number of mutated variables will be exactly that asked
                 [~, varIdx] = maxk(rand(size(pop, 2), 1), abs(mutProb));

Optimisation/Memetic Algorithm/MATLAB/Functions/selParents.m

@@ -3,7 +3,7 @@
 % Author: Achilles Kappis
 % e-mail: axilleaz@protonmail.com
 %
-% Date: 09/02/2024 (DD/MM/YYYY)
+% Date: 21/11/2024 (DD/MM/YYYY)
 %
 % Copyright: MIT
 % --------------------------------------------------
@@ -188,7 +188,7 @@ function [parents, parentsVec] = selParents(selType, fitVals, pop, offsprPerc, n
         idx = randi(size(pop, 1), nOffspr, nParent);
 
         % Put the parents into the array
-        for ii = nParent:-1:1
+        parfor ii = 1:nParent
             parents(:, :, ii) = pop(idx(:, ii), :);
         end
 
@@ -214,7 +214,7 @@ function [parents, parentsVec] = selParents(selType, fitVals, pop, offsprPerc, n
     % Choose parents
     % ====================================================
     % Do it for each offspring
-    for offsprIdx = nOffspr:-1:1
+    parfor offsprIdx = 1:nOffspr
         % Get parent indices based on their probability
         parentIdx = pickValProb(1:size(pop, 1), fitVals, nParent);
 

README.md

@@ -87,7 +87,7 @@ The project is licensed under the ***MIT License***, which is a rather unrestric
 
 
 ## Versioning ##
-The project uses [semantic versioning](https://semver.org/) and the current version is **v0.4**. The project hasn't reached v1.0.0 yet so changes that break backwards compatibility may (and have been) introduced in any version update. For information about the changes in each version consult the [CHANGELOG](https://gitlab.com/in-nova/in-nova/-/blob/main/CHANGELOG.md?ref_type=heads).
+The project uses [semantic versioning](https://semver.org/) and the current version is **v0.4.1**. The project hasn't reached v1.0.0 yet so changes that break backwards compatibility may (and have been) introduced in any version update. For information about the changes in each version consult the [CHANGELOG](https://gitlab.com/in-nova/in-nova/-/blob/main/CHANGELOG.md?ref_type=heads).
 
 
 #### **Important**