Discussing the article: "Population optimization algorithms: Micro Artificial immune system (Micro-AIS)" - page 5
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Any idea why it's hanging? Have you reported in the bugs and errors thread?
I wrote both on the forum and in the support. A long time ago. No response.
For a long time I tried to drag algorithms of this series into the optimiser (I wanted to parallelise them), but it glitches epically - https://www.mql5.com/en/forum/454524/page2#comment_50233782.
Apparently, you have encountered it. If you leave six ticks in that example, it will work. Therefore, you can adapt the algorithms of this series of articles now, just use simpler examples.
Apparently, you've encountered this. If you leave the six ticks in that example, it will work. Therefore, you can adapt the algorithms of this series of articles now, just use simpler examples.
Well, that would be an unrealistic simplification. There is a limit on the number of runs, as far as I understand, not on the number of parameters.
I took parameters from the settings of test fitness functions to have more or less real size of the optimisation space when comparing algorithms. ;-).
Well, that would be an unrealistic simplification. There is a limit on the number of runs, as far as I understand, not on the number of parameters.
I took the parameters from the settings of test fitness functions to have more or less real size of the optimisation space when comparing algorithms. ;-).
Honestly, I don't understand why MQ takes this Step literally. It's the level of minimum discretisation. And it has almost zero relation to the optimisation algorithm.
Frankly, I don't understand why MQ takes this Step literally. It is the level of minimum discreteness. It has almost nothing to do with the optimisation algorithm.
In the real representation of features - yes, and not almost, but zero relation.
For binary GA in this respect things are a bit more complicated, there are nuances.
I said earlier that you cannot compare the algo from the articles with the standard GA head-on, it is incorrect. A standard GA is a complex, which takes into account many nuances that would provide work on most user PCs: speed of work, uniqueness of new solutions, memory saving.
As far as I understood, custom optimisation is done only on the terminal graph on one core, and I was talking about multithreaded optimisation in the tester (for the particle swarm algorithm I described in the article, for most other algorithms it should also be possible by analogy, as there is usually a principle of dividing tasks into groups of agents). But the tester hangs on the most primitive example (I gave the test above), which nipped the idea in the bud.
Share the link. This?
Share the link. This?
Yes.