And in this form MLP can be more interesting than a single indicator or a bunch of several indicators, because it can create some additional signals non-linearly from one of them or from the same bunch.
Yes, the strategy search in this case is continued manually (something intelligent to collect manually), and MLP will be like a new super-indicator-filter
Optimisation can take two forms:
1) Parameters are optimised first, then MLP is switched on and weights are optimised
2) All together are optimised simultaneously
So let it live, it can still be twisted and turned.
Do not engage in nonsense) it has already reached the point that the data even on harmonics parsed as sound waves, trying to catch frequencies.No neural networks do not work here.In fact, if you look deep into their logic, it will become clear why.They work only for scammers like the author of quantum empiror, who in earnest claims that his owls are not scam and Neuronki he has working, such clowns only laugh at their mammoths, which bring him the money
Ivan Butko #: You see, it's a question of application: what and how.
After all, a price chart is not an earthquake chart, it's a coded language.
Well, as if the earthquake graph is analysed into harmonics and the frequencies with coordinates and the cause of the earthquake are detected with an accuracy of more than 90%.On forex we also tried this +Neurons were connected, but alas it did not work)))) No mathematics works here and especially neural networks.
And in this form MLP can be more interesting than a single indicator or a bunch of several indicators, because it can create some additional signals non-linearly from one of them or from the same bunch.
Yes, the strategy search in this case is continued manually (something intelligent to collect manually), and MLP will be like a new super-indicator-filter
Optimisation can take two forms:
1) Parameters are optimised first, then MLP is switched on and weights are optimised
2) All together are optimised simultaneously
So let it live, it can still be twisted and turned.
Но перед запуском скриптов действительно пришлось провести некоторую работу. Для тех кто не может запустить скрипт установите catboost. 3 скачайте все скрипты пайтон и положите их в одну папку. From tester пропишите путь к файлу в скрипте Tester Пожалуйста
And in this form MLP can be more interesting than a single indicator or a bunch of several indicators, because it can create some additional signals non-linearly from one of them or from the same bunch.
Yes, the strategy search in this case is continued manually (something intelligent to collect manually), and MLP will be like a new super-indicator-filter
Optimisation can take two forms:
1) Parameters are optimised first, then MLP is switched on and weights are optimised
2) All together are optimised simultaneously
So let it live, it can still be twisted and turned.
It will just adjust the condition to the ideal environment.On short range it will barely work.I've done similar.Same optimiser from a tester no more than that
There is an even more complex approach through meta models, it requires a bit more knowledge and experience. All this was written in the MO topic and articles. For the first time it was invented and voiced by me.
I don't even understand what you have written there)))
I'm just looking at the pictures.
If there was a manual for dummies "How to start this and what I do", it would be good. But most of the time you have to put everything together piece by piece.
Maxim Dmitrievsky #: There is an even more complex approach through meta models, it requires a bit more knowledge and experience. All this was written in the MO topic and articles. It was first invented and voiced by me.
Ivan Butko #: I don't even understand what you have written there.)
I'm just looking at the pictures.
If there was a manual for dummies "How to start this and what I do", it would be good. But most of the time you have to put everything together piece by piece.
Because everyone writes in different languages :) so I write only briefly my thoughts.
Maxim Dmitrievsky #: There is an even more complex approach through meta models, it requires a bit more knowledge and experience. All this was written in the MO topic and articles. It was first invented and voiced by me.
What is modesty? How to deal with it and whether it should be done?
Allow the use of cookies to log in to the MQL5.com website.
Please enable the necessary setting in your browser, otherwise you will not be able to log in.
You are signed in. Two-factor authentication has been disabled. Set it up again and scan a new QR code, or continue with two-factor authentication disabled.
And in this form MLP can be more interesting than a single indicator or a bunch of several indicators, because it can create some additional signals non-linearly from one of them or from the same bunch.
Yes, the strategy search in this case is continued manually (something intelligent to collect manually), and MLP will be like a new super-indicator-filter
Optimisation can take two forms:
1) Parameters are optimised first, then MLP is switched on and weights are optimised
2) All together are optimised simultaneously
So let it live, it can still be twisted and turned.
You see, it's a question of application: what and how.
And in this form MLP can be more interesting than a single indicator or a bunch of several indicators, because it can create some additional signals non-linearly from one of them or from the same bunch.
Yes, the strategy search in this case is continued manually (something intelligent to collect manually), and MLP will be like a new super-indicator-filter
Optimisation can take two forms:
1) Parameters are optimised first, then MLP is switched on and weights are optimised
2) All together are optimised simultaneously
So let it live, it can still be twisted and turned.
And in this form MLP can be more interesting than a single indicator or a bunch of several indicators, because it can create some additional signals non-linearly from one of them or from the same bunch.
Yes, the strategy search in this case is continued manually (something intelligent to collect manually), and MLP will be like a new super-indicator-filter
Optimisation can take two forms:
1) Parameters are optimised first, then MLP is switched on and weights are optimised
2) All together are optimised simultaneously
So let it live, it can still be twisted and turned.
You've twisted the original approach again
There is an even more complex approach through meta models, it requires a bit more knowledge and experience. All this was written in the MO topic and articles. It was first invented and voiced by me.
I don't even understand what you have written there.)
There is an even more complex approach through meta models, it requires a bit more knowledge and experience. All this was written in the MO topic and articles. It was first invented and voiced by me.
What is modesty? How to deal with it and whether it should be done?