Libraries: MultiTester - page 10
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
The test set has recently finished. With 50 iterations there were no double passes.
I put logging, mostly iterations 1, less often 0 and 2. But sometimes, rarely and unknown why, there are bursts up to 7-8. I will keep 50, with reserve and no negative effects from it.
In the MTTester.mqh file, the paths to the cache or configs are obtained via ::TerminalInfoString(TERMINAL_PATH)+.
But personally I have it installed (I set everything by default) in Programme Files, and the profile is separate. I corrected it to TERMINAL_DATA_PATH.
The test set has recently finished. With 50 iterations there were no double passes.
I put logging, mostly iterations 1, less often 0 and 2. But sometimes, rarely and unknown why, there are bursts up to 7-8. I will keep 50, with reserve and no negative effects from it.
I haven't had any negative effects. Thanks for the info, I'll put 50, it's not asking for food, but it will work for others.
In MTTester.mqh file paths to cache or configs are obtained via ::TerminalInfoString(TERMINAL_PATH)+
But personally I have it installed (I put everything by default) in Programme Files, and the profile is separately. I corrected it to TERMINAL_DATA_PATH.
I work only in portable, and there TERMINAL_PATH == TERMINAL_DATA_PATH. Thanks, I'll make universal mode work in all KB jobs.
It may be useful to get not only GetLastOptCache (and correspondingly GetLastOptCacheFileName) in principle the last one at all. And also the last on a particular character. I've added it for myself, but it may be useful in the general version.
When writing it, I am guided by the usage scenarios that I myself have encountered. I didn't consider the option with the last cache by symbol, because I still don't see where it can be useful.
It might be useful for comparing many characters against each other. I haven't come up with a useful script yet.
If you have your own implementations or usage scenarios, post them. It's even more constructive that way.
At the moment I'm just at the stage of finalising and fleshing out the tools. That's why I first ran optimisation on 90 symbols. And then I decided not to parse the results by hand, but by automat, and shake out interesting symbols from the whole pile to turn them by hand. And I failed with the basic functionality, so I needed filtering by symbol.
The corrections were minimal and compatible, I added an optional parameter const string SymbName = NULL, threw it through a couple of functions and slightly corrected GetLastOptCacheFileName
to
return(MTTESTER::GetFreshFileName(::TerminalInfoString(TERMINAL_DATA_PATH)+"\\\Tester\\\cache\\\",SymbName != NULL ? "*. "+SymbName+".*.opt" : "*.opt"));
At the moment I'm just at the stage of finalising and fleshing out the tools. That's why I first ran optimisation on 90 symbols. And then I decided not to parse the results by hand, but by automat, and shake out interesting symbols from the whole pile to turn them by hand. And I failed with the basic functionality, that's when I needed filtering by character.
Got it. Then it is logical to add the name of the EA as well.
The corrections were minimal and compatible, I added an optional parameter const string SymbName = NULL, threw it through a couple of functions and slightly corrected GetLastOptCacheFileName
to
return(MTTESTER::GetFreshFileName(::TerminalInfoString(TERMINAL_DATA_PATH)+"\\\Tester\\\cache\\\",SymbName != NULL ? "*. "+SymbName+".*.opt" : "*.opt")));
I'll fix the mask generation, thanks.
Something is wrong...
And there are only 3 .opt files...
How is that possible?
Ran a manual optimisation on one of the "lost" characters. Everything is fine....
Something happened...
Unfortunately, I don't always update them in KB when I find bugs or ways to improve my solutions. This is probably the case here.
Unfortunately, I don't always update my solutions in KB when I find bugs or ways to improve them. This is probably the case here.
So I haven't updated anything after adapting to 2190.
Let's write it off as a system glitch. At least until the reappearance. )))
So I haven't updated anything since adapting it to 2190.
I updated it considerably, because I found and eliminated pitfalls. I also refined it for other types of tasks.