Discussion of article "Experiments with neural networks (Part 1): Revisiting geometry" - page 2

New comment
 
Excellent article!
 
Enrique Enguix #:
Excellent article!

Thank you!

 

Hi Roman, thank you for your idea and sharing, may I ask for a query about the Perception4:

double perceptront4() 
  {
   double w1 = x1 - 100.0;
   double w2 = x2 - 100.0;
   double w3 = x3 - 100.0;
   double w4 = x4 - 100.0;
   
   double a1 = (ind_In1[1]-ind_In1[10])/10;
   double a2 = (ind_In2[1]-ind_In1[10])/10;
   double a3 = (ind_In1[1]-ind_In1[10])/10;
   double a4 = (ind_In2[1]-ind_In2[10])/10;
   
   return (w1 * a1 + w2 * a2 + w3 * a3 + w4 * a4);
  }

according to your code, it seems a3 mis-typed equal to a1 = (ind_In1[1]-ind_In1[10])/10;

(I found the source code is the same as be)

Refer to the figure (as attached), (I think) it should be 

   double a3 = (ind_In1[1]-ind_In2[10])/10;


I'm a newer to MT4/5 and neural network, it is an very interesting to study and discover your AI method article, is there any chance to change your results and conclusion? 

hope to see your next article.


Best regards

WK

Files:
11.10.2022_18.15.55_REC.png  104 kb
 
День семьи ФФДень10 # :

Привет, Роман, спасибо за идею и то, что поделились, могу я задать вопрос о Perception4:

согласно вашему коду, кажется, что a3 неправильно выбрано, равно a1 = (ind_In1[1]-ind_In1[10])/10;

(Я наблюдал, что исходный код такой же, как и был)

Обратитесь к рисунку (как прилагается), (я думаю) он должен быть


Я новичок в MT4/5 и  нейронной сети , очень интересно узнать и открыть для себя вашу статью по методу ИИ, есть ли шанс изменить ваши результаты и заключение? 

надеюсь увидеть вашу статью.


с настройками

ВК

Hello. Perhaps there is a typo. Try to change. Thanks for your feedback.

 

Good afternoon, the topic is certainly interesting. Thank you for your labour. But it seems that we need to dig much deeper here.

No offence, but this is a classic testor grail. To be more specific: at the size of the tp of 60 pips (and on the old ones it is only 6 pips) when testing on H1 at the opening prices, the deal is closed and opened on one candle, which gives the grail. Even 200 new pips on a five-digit trade fit into one candle. As a result - testing error.

It is understandable. To catch a fish in this highly effective market, one technique is not enough for an enthusiast, unfortunately....

 
vinnipyx opening prices, the deal is closed and opened on one candle, which gives the grail. Even 200 new pips on a five-digit trade fit into one candle. As a result - testing error.

It is understandable. To catch a fish in this highly effective market, one technique is not enough for an enthusiast, unfortunately....

Good afternoon. Thank you for your feedback.

 

Good afternoon. I like your idea. I decided to expand it a bit.

Made some changes to your main file.

The idea is as follows. With a small period of optimisation "very few" deals it turns out to be a good indicator of the Maximum of complex criterion.

I made your Expert Advisor as a tester, then we save the results of optimisation in xml file. And in MT5 itself the table of results should be ordered exactly by this indicator.

Then I run my own parser written in c#.

Exe accepts 4 parameters - the path to the results file, the index of the complex criterion which is interesting for example 95, the minimum number of trades, the path to the file with the final sample.

Then my Expert Advisor comes into play - the logic is the same as in your file, but it takes a csv file as input.

And it trades a huge number of sets.

And if several sets give one direction, the input is only one.

Optimisation by opening prices, testing on real ticks.

It is better to run exe from cmd file for example

ParsingPerceprtonOpti.exe "E:/\MetaTrader 5\MQL5\Profiles\\Tester\\\ReportOptimizer-123321.xml". 99.0 70 "e:\MetaTrader 5\MQL5\Files\\#AllGoodTrade.csv."

If anything, the exe project is also in the trailer (VS 2019 C#).

Password for the archive opti

The name of the result file is only #AllGoodTrade.csv (I need it for my EA version to work properly in the tester) - such a feature #property tester_file "#AllGoodTrade.csv".


Files:
opti.zip  241 kb
 
Alexey Klenov we save the results of optimisation in xml file. And in MT5 itself the table of results should be ordered exactly by this indicator.

Then I run my own parser written in c#.

Exe takes 4 parameters - the path to the results file, the index of the complex criterion which is interesting for example 95, the minimum number of trades, the path to the file with the final sample.

After that, my Expert Advisor comes into play - the logic is basic from your file, but takes a csv file as input.

And it is already trading a huge number of sets

And if several sets give one direction, the input is only one.

Optimisation by opening prices, testing on real ticks.

It is better to run exe from cmd file for example

ParsingPerceprtonOpti.exe "E:\MetaTrader 5\MQL5\Profiles\Tester\\ReportOptimiser-123321.xml". 99.0 70 "e:\MetaTrader 5\MQL5\Files\\#AllGoodTrade.csv"

If anything, the exe project is also in the trailer (VS 2019 C#).

Password for the archive opti

The name of the result file is only #AllGoodTrade.csv (it is necessary for my version of the EA to work properly in the tester) - such a feature #property tester_file "#AllGoodTrade.csv".


Good day. Thank you for your feedback. The number of transactions as a filter is the right solution. Read part 3 there is similar to yours.

 

I didn't understand from the article, what percentage of passes to the forward yielded positive results?

 
Aleksey Vyazmikin #:

I didn't understand from the article, what percentage of passes to the forward yielded positive results?

Good afternoon. The percentage was not counted. Read part 2 and 3.

123
New comment