Errors, bugs, questions - page 2685
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
Batch Optimisation.
The problem is that if the StepOver (F10) action is applied to a function call at least once, there is no way to debug this function later.
Steps for playback:
1) Run the code in debug mode;
2) After a breakpoint is triggered, execute StepOver (F10) twice;
That's it - now there is no way to "get" intoIncrement function, all manually set breakpoints do not work, and instead of operation StepInto (F11) is actually performed StepOver (F10).
A defect in ME debugger(build 2370) - after abreakpointis triggeredinstead of continuing code execution by pressing (F5), the samebreakpointis actually triggered againwithout executing any code.
Replay steps:
1) Manually setthe breakpoint for both lines " arr.Add()";
2) Start debugging (F5);
3) Use PCM to open the <Generic\ArrayList.mqh> nested file in a new tab.
4) Find the line "if(m_size==ArraySize(m_items))" in the function "bool CArrayList::Add(T item)" and setthe breakpoint manually.
5) While in the "ArrayList.mqh"tab, continue executing the code (F5).
This willtrigger the samebreakpoint as before without any actual code execution.
As a result, it is not possible to debug the codeof the static structure method at all:
Forum on trading, automated trading systems and strategy testing
Question to developers - Using all computational cores during optimization
Andrey Dik, 2020.03.26 23:31
You can set up to 100 mio steps so the optimizer will not switch to genetics.
I found this correspondence, and I think I read it when it happened - the result is a member ban, that's why I wanted to find the original source
I tested in full search mode by one parameter - the passes are numbered out of sequence, in general full search - works according to some logic of the developers, not the semantic concept
tested in full search mode by one parameter - passes are numbered out of sequence, in general full search - works according to some logic of developers, not the semantic concept
So, tasks are handed out in packs to agents.
1-100 to the first, 101-200 to the second, etc.
And in results appear as they come in (1, 101, 2, 102, etc.).
Easy to sort.
So tasks are handed out in packs to agents.
1-100 to the first, 101-200 to the second, etc.
And in the results appear as they come in (1, 101, 2, 102, etc.).
Easy to sort.
OK, that's a good answer, that's about right, pity the original source was never found,
it remains to find out at what total number of passes will be included GA, so far it seems like 100 000 000 passes, for my tasks is enough