What should be added for additional support of universal mathematical calculations in MQL5 and MQL5 Cloud Network? - page 11
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
Yes, someone's mind is boiling with indignation :) the sorts are hilarious.
Reshetov, "wash your eyes", I asked not to show your single development, but a project where at least a couple of other programmers take part.
Now, I'll drop everything and start introducing all sorts of demagogues and other inquisitive dodgers to the projects.
and go easy on the choice of words, you're being aggressive.
I'm not swearing too much and I try to choose words according to the context, i.e. I try to call things by their proper names.
If you've hurt my feelings, excuse me, Mr. Commander, but no one forced you to get into a discussion and criticize my research.
Your personal opinion on how to run projects is already known and I am not interested in it, i.e. ignored, and therefore further discussion with you I do not see any sense. I do not need advisers and other well-wishers.
By no means do I want to have an argument.
I always welcome constructiveness and even less of a waste of other people's time, as I respect it.
It's fun here.
There's "New Neural" - an Open Source neural network engine project for the MetaTrader 5 platform. at least there's something to read, but this thread starts right off with a rant.
Renat:
Hence the question - what other functions need to be included to improve the capability of the calculation network?
A refinement is needed. A genetic algorithm often stops far from an extremum, which is unacceptable for many mathematical calculations.
On my homemade genetics I did the following.
After the GA, worked it out:
1. I set the flag to false.
2. take the input parameters (arguments of the investigated function) individually
3. I memorize the value of the selected input parameter and run a dichotomy on it (search for an extremum).
4. If the value of the input parameter has changed after the dichotomy, then the flag is true.
5. If there are any more input parameters, par. 2
6. If the flag is true, then item 1. 1
The results are often quite different from those generated by the genetics. Therefore, if extrema are stable, then the finetuning is the only way to detect them with accuracy given by the discreteness of the input parameters.