How to train AI with indicators? - Articles, Library comments

Discussion of article "Neural networks made easy (Part 2): Network training and testing"

MetaQuotes · 2024-05-06T21:16:15.0000000

New article Neural networks made easy (Part 2): Network training and testing has been published: In this second article, we will continue to study neural networks and will consider an example of using our created CNet class in Expert Advisors. We will work with two neural network models, which show similar results both in terms of training time and prediction accuracy. The first epoch is strongly dependent on the weights of the neural network that were randomly selected at the initial stage. After 35 epochs of training, the difference in statistics increased slightly - the regression neural network model performed better: Value Regression neural network Classification neural network Root mean square error 0.68 0.78 Hit percentage 12.68% 11.22% Unrecognized fractals 20.22% 24.65% Testing results show that both neural network organization variants generate similar results in terms of training time and prediction accuracy. At the same time, the obtained results show that the neural network needs additional time and resources for training. If you wish to analyze the neural network learning dynamics, please check out the screenshots of each learning epoch in the attachment. Author: Dmitriy Gizlyk

Boris Egorov 2020.10.25 13:47 #11

Igor Makanu:

then you don't need an article section

for your requirements - a profitable state, the sections codobase and Market are suitable

yes, by the way, even your message does not fit your requirements, so it can be ignored too? )))

You need articles, you need them to properly test an EA working on the basis of neural network, it is highly probable that your state from the tester is not only from MT4, but you may not have divided the training sample into train/test/validation.

You need articles to learn how to write good structured and readable code - I think the author's code in the article is perfect for these requirements.

I don't understand about mt4, probably neural networks don't work on mt4 :-)) but I have mt5

about training sampling and further ... I don't get it either .... What is this about?

>write good structured and readable code - I think the author has a perfect code for these requirements in the article.

I don't agree, everything in this series of articles can be presented in a much simpler, shorter and clearer way, and this series of articles is complicated, incomprehensible, cumbersome and, most importantly, fruitless.

>articlesare necessary, necessary to properly test an Expert Advisor working on the basis of neural network.

cool .... I laughed for a long time at this ....

and another thing: my opinion is only my opinion.

I will not write any more comments, I tell the author for the second time that everything is complicated and useless, the second time I show him only the theoretical value and complete practical uselessness of such articles ..... God judge him ...

I agree in one thing, he has his own and original approach, it is normal and useful for those who read it.

Discussion of article "Deep Discussion of article "How Discussion of article "Comparative

Igor Makanu 2020.10.25 13:59 #12

Boris Egorov:

I don't understand about mt4, maybe neural networks don't work on mt4 :-)) but I have mt5.

You have a screen from MT4 tester.

if it is MT5, then teach me how to make a screen with the number of trades, not with test dates, I asked yesterday https://www.mql5.com/ru/forum/354075/page2#comment_18896390.

this is purely a practical question

I don't want to discuss everything else, it's your decision, everyone has their own interests and tasks.

Discussion of article "The Discussion of article "Evaluating Discussion of article "Practical

Boris Egorov 2020.10.25 14:18 #13

Igor Makanu:

you have a screenshot from the MT4 tester.

if it is MT5, then teach me how to make a screenshot with the number of trades, not with test dates, asked yesterday https://www.mql5.com/ru/forum/354075/page2#comment_18896390

this is purely a practical question

about everything else, I don't want to discuss, it's your decision, everyone has their own interests and tasks.

save a detailed report and you'll see

Igor Makanu 2020.10.25 14:26 #14

Boris Egorov:

save the detailed report and you will see

Yes, it works.

thank you!

Valeriy Yastremskiy 2020.10.26 19:39 #15

Boris Egorov:

I do not agree, everything in this series of articles can be presented much simpler, shorter and clearer, and this series of articles in general is complicated, incomprehensible, cumbersome and most importantly ineffective.

Write simpler, shorter and clearer and everything will fall into place. Remarks that you have it bad, I have it better, I need to redo it are not appropriate anywhere.

star-ik 2020.11.01 07:08 #16

A big request from a newbie. Could you please give me a few lines of code to make the Expert Advisor trade? I'm struggling for a long time, I can't get it right.

Scripts: sSyncScroll Discussion of article "How Discussion of article "Programming

star-ik 2020.11.01 07:12 #17

//+----------------------------------------------------------------------+

//| This is where we insert the algorithm for making a decision on the signals of the neural network |

//+----------------------------------------------------------------------+

Here, at least something simple for an example. Thanks in advance.

star-ik 2020.11.01 07:15 #18

And another question. How to forcefully stop the training. Thank you.

star-ik 2020.11.25 14:04 #19

I've been bombarded with answers. A heartfelt ...

Zhiqiang Zhu 2020.12.26 10:54 #20

12.68% hit rate? Does that have any real meaning? I'm getting about 50% probability on coin tosses too. With such a low probability, I wonder what the point of this thing is?

Discussion of article "Developing Discussion of article "14,000 Discussion of article "How

Discussion of article "Neural networks made easy (Part 2): Network training and testing" - page 2