A discussion on MLP and its use of the word - Articles, Library comments

Discussion of article "Neural Networks Cheap and Cheerful - Link NeuroPro with MetaTrader 5"

MetaQuotes · 2024-04-04T13:59:09.0000000Z

New article Neural Networks Cheap and Cheerful - Link NeuroPro with MetaTrader 5 has been published: If specific neural network programs for trading seem expensive and complex or, on the contrary, too simple, try NeuroPro. It is free and contains the optimal set of functionalities for amateurs. This article will tell you how to use it in conjunction with MetaTrader 5. The NeuroPro program was written in one of Russian institutes back in 1998 and is still relevant today. It efficiently runs on Windows XP, Vista and Windows 7. I cannot tell how it works on later versions of Windows as I have not tested it. Version 0.25 is free and can be found on many websites on the Internet. NeuroPro can create multilayer neural networks with the sigmoid activation function . If you have just started learning neural networks, you do not need more features at this stage. It should be kept in mind that the interface of NeuroPro is in Russian and has not been translated into any other languages. Neural network can be trained on one data array and then tested on another one. It is an essential feature for traders as it allows to understand quickly if the selected network structure is prone to overfitting and if it can consistently trade outside historical data, i.e. on a real account. Those who like to dig deeper have an opportunity to see neural network weights as well as which network inputs influence the result of network operation most of all. Beginners do not need that and they may skip this part of the program. This information is useful for experienced traders looking for the Grail because it lets them assume what pattern was identified by the neural network and see where they can continue their search. Other than that, there are no significant features in NeuroPro except various settings and useful utilities like minimizer of the network structure. These menu sections are not compulsory to use so novices do not have to complicate things and use only the default settings. From a trader's point of view, NeuroPro has only one disadvantage - absence of integration with MetaTrader 5. Actually, this article is mostly dedicated to loading market and indicator data from MetaTrader 5 to NeuroPro and then turn the received neural network into an Expert in MQL5. Advancing the topic, I can say that the neural network that we are going to create with NeuroPro will be converted with all the neuron weights into an MQL5 script (unlike the system of include DLL like in any other neural network program ). It will ensure fast work and minimum use of computer resources. That is a clear advantage of using NeuroPro. It can be used for creating any trading strategies, even scalping ones with their requirement for the Expert to make decisions nearly immediately. Trading Strategy In this article we are not going to consider scalping because the process of creating, training and testing scalping Experts has a lot of peculiarities and goes beyond this article. For educational purposes we shall create a simple Expert for the H1 timeframe and popular currency pair EURUSD. So, let our Expert analyze last 24 bars i.e. market behavior in the last day, forecast the direction of the price movement in the following hour and then trade based on that information. Fig. 22. Equity chart after the Expert Advisor has been tested in MetaTester Author: Andrew

СанСаныч Фоменко 2014.11.23 15:37 #11

vlad1949:

SanSanych

Here is a link to the article I was talking about. http://appliedpredictivemodeling. com/blog/2014/11/11/some-thoughts-on-do-we-need-hundreds-of-classifiers-to-solve-real-world-classification-problems

Also in the article there is a link to an earlier article by David Hand on an issue you and I have discussed before - poor results after training on real data. Very interesting thoughts. Maybe you could do a shortened translation?

Rummaging through the archives and found another article on the topic of comparing different machine learning algorithms.

http://www.cs.cornell.edu/~caruana/ctp/ct.papers/caruana.icml06.pdf 

Good luck

In this thread our posts are somewhat off-topic, and I have my own thread. Therefore, I suggest moving our discussion here, my post there is in response to yours.

Stanislav Korotky 2014.11.23 19:01 #12

faa1947:
In this thread our posts are somewhat off-topic, and I have my own thread. So I suggest that we move our discussion here, my post there is in response to yours.

And we don't know where this other discussion is, where to move to. ;-)

СанСаныч Фоменко 2014.11.23 19:23 #13

marketeer:
And we don't know where this other discussion is, where to move to. ;-)

This way

Vladimir Perervenko 2014.11.24 07:11 #14

faa1947:
In this thread our posts are somewhat off-topic, and I have my own thread. So I suggest we move our discussion here, my post there is in response to yours.

On the contrary. Many people who first heard the word "neural networks" for the first time will find it useful to know that science and practice have moved far ahead since the advent of MLP.

You may be right, though.

Gone.

Good luck to the author.

Neural Network indicator - Better NN EA 100$

Anton29 2014.11.24 21:12 #15

Da statya otlichnaya, tolko ne mogu ne mogu nikak razobratsya c camim nachalom, a imeno otkuda otkuda vzyzlsya script with ictoriei kotirovok #property script_show_inputs

//+------------------------------------------------------------------+
input string    Export_FileName  = "NeuroPro\\data.csv"; // Export file (in the "MQL5/Files" folder)
input int       Export_Bars_Skip = 0;                    // How many history bars to skip before exporting
input int       Export_Bars      = 5000;                 // Number of rows of data to be exported
//+------------------------------------------------------------------+
const int inputlen=24;   // How many past bars the trading strategy analyses
//+------------------------------------------------------------------+
void OnStart()
  {
  //--- create a file
   int file=FileOpen(Export_FileName,FILE_WRITE|FILE_CSV|FILE_ANSI,',');

   if(file!=INVALID_HANDLE)
     {
      //--- write the data header
      string row="date";
      for(int i=0; i<=inputlen; i++)
        {
         if(StringLen(row)) row+=",";
         //========================================================
         // Note! 
         // In the EA, we will replace underscores with [].
         // The field name in DBase format is up to 11 characters. Calc truncates to 10.
         // The maximum number of fields in DBase format is 128-512, depending on the version.
         //========================================================
         StringConcatenate(row,row,"Bar___",i,"__");
        }
      FileWrite(file,row);

      //--- copy all the necessary data from the history
      MqlRates rates[],rate;
      int count=Export_Bars+inputlen;
      if(CopyRates(Symbol(),Period(),1+Export_Bars_Skip,count,rates)<count)
        {
         Print("Error! Insufficient history size to export required data.");
         return;
        }
      ArraySetAsSeries(rates,true);

      //--- write the data 
      for(int bar=0; bar<Export_Bars; bar++)
        {
         row="";
        //--- close price of the 1st bar is set as the zero level for normalisation of other data
         double zlevel=rates[bar+1].close; 
         for(int i=0; i<=inputlen; i++)
           {
            if(StringLen(row)) row+=",";
            rate=rates[bar+i];
            if(i==0) row+=TimeToString(rate.time,TIME_DATE || TIME_MINUTES)+",";
            row+=DoubleToString(rate.close-zlevel,Digits());
           }
         FileWrite(file,row);
        }
      FileClose(file);
      Print("Data export completed successfully.");
     }
   else Print("Error! Failed to create a file to export data. ",GetLastError());

}i gde etoo zhurnal expertov

date,Bar___0__,Bar___1__,Bar___2__,Bar___3__,Bar___4__,Bar___5__,Bar___6__,Bar___7__,Bar___8__,Bar___9__,Bar___10__,Bar___11__,Bar___12__,Bar___13__,Bar___14__,Bar___15__,Bar___16__,Bar___17__,Bar___18__,Bar___19__,Bar___20__,Bar___21__,Bar___22__,Bar___23__,Bar___24__
2014.09.25,-0.0008,0.0000,-0.0005,-0.0014,0.0007,0.0035,0.0035,0.0036,0.0047,0.0052,0.0050,0.0046,0.0046,0.0047,0.0049,0.0052,0.0049,0.0053,0.0055,0.0056,0.0067,0.0056,0.0097,0.0105,0.0113
2014.09.25,0.0005,0.0000,-0.0009,0.0012,0.0040,0.0040,0.0041,0.0052,0.0057,0.0055,0.0051,0.0051,0.0052,0.0054,0.0057,0.0054,0.0058,0.0060,0.0061,0.0072,0.0061,0.0102,0.0110,0.0118,0.0123
2014.09.25,0.0009,0.0000,0.0021,0.0049,0.0049,0.0050,0.0061,0.0066,0.0064,0.0060,0.0060,0.0061,0.0063,0.0066,0.0063,0.0067,0.0069,0.0070,0.0081,0.0070,0.0111,0.0119,0.0127,0.0132,0.0130

2014.09.25,-0.0021,0.0000,0.0028,0.0028,0.0029,0.0040,0.0045,0.0043,0.0039,0.0039,0.0040,0.0042,0.0045,0.0042,0.0046,0.0048,0.0049,0.0060,0.0049,0.0090,0.0098,0.0106,0.0111,0.0109,0.0122.

A imenno do momenta Convert from CSV to DBF, dalee vce uzhe proshel no ne c temi dannimi. Narod podckazhite kak cdelat' etot script c barami i ictoriei, pri tom chto export kotirovok ne preducmotren v MT5 !??? Cherez MQL5 to create an Expert Advisor template, custom indicator, script !?! ? HOW...

Neural Networks Cheap and Deep Neural Networks (Part Deep Neural Networks (Part

Eugeniy Lugovoy 2014.11.24 21:16 #16

I'd like to have a DLL of some kind to put the input data in and the coefficients back out. Does anyone know which neuro software has this...? or contact the author of NeuroPro about the sources and write the DLL yourself....

How do I use VPS not migrating MT4 String -> DLL

Maxym Kondratiuk 2014.11.24 22:37 #17

elugovoy:
I'd like to have a DLL of some kind to put the input data in and the coefficients back out. Does anyone know which neuro software has this...? or contact the author of NeuroPro about the sources and write the DLL yourself...

There is an old NeuroShell 2

Stanislav Korotky 2014.11.24 23:58 #18

elugovoy:
I'd like to have a DLL of some kind to put the input data in and the coefficients back out. Does anyone know which neuro software has this...? or contact the author of NeuroPro about the sources and write the DLL yourself....

In Deductor it is possible to run a prepared project with a bootnik.

Yury Reshetov 2014.11.25 07:03 #19

elugovoy:
I'd like to have a DLL of some kind to put the input data in there and the coefficients back out. Does anyone know which neuro software has this...? or contact the author of NeuroPro about the sources and write the DLL yourself....

If via dll, see https://www.mql5.com/en/articles/252.

Прогнозирование временных рядов в MetaTrader 5 при помощи библиотеки машинного обучения ENCOG

2011.04.25
investeo
www.mql5.com

Статья посвящена использованию функционала нейронных сетей библиотеки машинного обучения ENCOG в MetaTrader 5. В качестве примера приведена реализация простого нейросетевого индикатора на основе технических индикаторов и советника, торгующего по сигналам нейросетевого индикатора. Все исходные коды, скомпилированные библиотеки и примеры обученной сети прилагаются к статье.

СанСаныч Фоменко 2014.11.25 08:11 #20

Reshetov:
If via dll, see https://www.mql5.com/en/articles/252.

Funny name: ENCOG - machine learning... Well, it's hot.

The tools listed here are only a part on machine learning.

CRAN Task View: Machine Learning & Statistical Learning

cran.r-project.org

Several add-on packages implement ideas and methods developed at the borderline between computer science and statistics - this field of research is usually referred to as machine learning. The packages can be roughly structured into the following topics: Neural Networks : Single-hidden-layer neural network are implemented in package...

Discussion of article "Neural Networks Cheap and Cheerful - Link NeuroPro with MetaTrader 5" - page 2