"New Neural" is an Open Source neural network engine project for the MetaTrader 5 platform. - page 28
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
1) Yes, many things have been invented. gpwr can (if he wants to) give a primer on the subject.
2) I realize that this question is pure provocation. But congratulations, I fell for it. More than a month ago I resigned from my main, fairly well-paid by today's standards, job, and engaged in auto-trading, using the accumulated knowledge and experimental experience in this field of knowledge. People who are close to me, my friends, respected forum users, are aware of the recent events in my life.
3) More than two years of experience in studying and mastering genetic algorithms, whose initial ideas in source codes are laid out in my article, allow me to say that building a robust system is not only possible but also successfully applied in real, everyday life. And I dare say that it is systems built on neural technology and self-organization/self-learning/adaptive principles that are capable of being robust in the long run in the marketplace.
i see. i just didn't understand how it all relates to the subject - that you quit and that you're sure about self-organization etc., but ok, forget it :)
Of course you have to build them, but you don't have to implement them.
First we need to understand what is a brick, then what is the ethane, and then mold a skyscraper of a lot of floors.
For now the discussion is more like how it would be nice to have a small multi-storey building, like 16-20 floors. Not that they have not learned how to build two floors, with technology and building materials have not figured out.
Two floors we can build, the materials will be agreed when the results of geological exploration, you need to know the soil on which to build.
Now in this thread has presented two concepts universal network + on the approach of sharing the project MQ, but we need to figure out exactly for all networks suitable this universalism. To do this, a request was made yesterday to post links to the most promising network algorithms.
And in parallel hear the views of users on the issue of "what should be the program" a common question, of course, but it's a brainstorming allowed any fantasy.
I see. just do not understand how it all relates to the sub-item - the fact that you have resigned and that you are sure about the self-organization, etc., but okay, forget it :)
Moreover, I do not understand the meaning of your presence here if you doubt the necessity and prospects of the project and the applicability of methods to the market, used in it.
Two floors we can build, the materials will be coordinated when the results of geological exploration, it is necessary to know the soil on which we build.
Now in this thread already presented two concepts of universal network + on the way to sharing the project MQ, but we need to find out exactly whether this universalism is suitable for all networks. To this end yesterday a request was made to post links to the most promising network algorithms.
And at the same time we listen to users' opinions on "what should be the program", a common question, of course, but it is a brainstorming session, any fantasies are allowed.
So we need to solve some simple problem based on these two concepts. For example, teach the NS to analyze and understand candlestick patterns consisting of one bar.
In general, it is possible to start from this "stovepipe", i.e. to think about what and how we can analyze in this situation, what results we can get, etc.
If new concepts will appear on the same task and to start.
But do not forget that you can analyze not just one candle, but a few (two, three, etc.). Also, it is worth to think that it is possible to analyze other things, such as indices (but the example of analysis of one candle is very simple and is a classic).
Correct construction - is on each new measurement (line shift) introduction of the corresponding function of a deviation, providing criterion of maximum efficiency of estimation of linear dependence of the analyzed sample, which(function) depends on the law of dispersion distribution, which(law of dispersion) also should be correctly estimated, that too a problem, etc. I haven't seen any implementation of it. This is a non-trivial task.
And what, then, prevents it from being realized.
Or are you one of those "critics". Who don't like everything and don't do anything about it.
Let's try it. I have nothing to steal, so I'm not afraid.
OK. All network functions are in the file EchoNet.mqh . I think I commented out a little bit.
Development of joint -- Joo and mine.
Yes, with "easy" I was a little too much, it's easy to use, but to redesign ... will have to work hard :)
Maybe it makes sense to see how this is implemented in the following software: NeuroShellDayTrader5.
It comes with Addons as options - ready-to-use working models.
Let's put it this way:
For specialists - post links to *promising topologies and learning algorithms.
For users - **create interfaces of the graphical editor of network creation, and send files of templates.
**Not promising ones can be expanded and connected (if possible) later, the main thing is to shake out the concept, so that the network creation possibilities do not contradict the best ones.
** in MT5 there is a possibility of adding buttons and input fields for arrows and similar objects, from this whole picture, a competent programmer will quickly grasp the essence of your idea. The template file will repeat completely all your buttons their location, text fields, color, etc.
Template files are stored in Profiles\Templates
Moreover, I do not understand the meaning of your presence here
Maybe it makes sense to see how this is implemented in the following software: NeuroShellDayTrader5.
It comes with Addons as options - ready-to-use working models.
Can I get the link?
By the way, it's for everyone, if you like something, post links.