Hi
Since I run Wallacei X, “Evaluation Time Per Solution” is about 6 minute. As the generations pass during optimization the time increases so the “Estimate Time Remaining” is increased as well.
As my algorithm runtime is high and lasts for days, there is a big difference between the simulation time at the beginning and the time it actually takes.
I was just wondering if this time increasing is for the optimization process of my algorithm that it’s natural or there is a problem?
By the way I tested my algorithm on two different computers and the problem is still the same.
Thank you for your time.
Hi,
The Time per Solution is being calculated live as the optimisation progresses. And it is obvious that different solutions generated may require more time for calculation due to the complexity of the forms and geometries generated. So if the optimisation explores an area of the design space that is full of these complex solutions, the time will increase.
Thank you so much for the response. Just have one question. When I run one solution manually, it doesn’t take much time to calculate the objectives, but during the optimization process, the same solution takes much more time (not at the beginning but at the middle of the progress and so on). So apart from the complexity of the forms and design space, is this time increase due to the use of genetic algorithm for finding the best solutions?
when you manually compute the canvas for one single solution is not necessarily similar to the ones in the middle of the optimisation that may require more time.
Other factors such as the gene count may also have an impact on the time. If you share your file I can give a better insight.