Discussion of article "Connecting NeuroSolutions Neuronets" - page 4

[Вангелис]

Urain:

Change it

to

I did it, now it only fails on line 20: implicit conversion from 'number' to 'string' WeekPattern-Export.mq5 20 24

20 34, 20 43, 20 54

row += "Open"+i+",High"+i+",Low"+i+",Close"+i;

[[Deleted]]

Wangelys:

I did it, now it only fails on line 20: implicit conversion from 'number' to 'string' WeekPattern-Export.mq5 20 24

20 34, 20 43, 20 54

There is such a wonderful function as StringConcatenate.

In your case it will look like this (by the way, I solved similar problems in MT4 in this way)...

  for(int i=0; i<=5; i++)
  {
    if(StringLen(row)) row += ",";

  StringConcatenate(row,row,"Open",i,",High",i,",Low",i,",Close",i);
  }

[Вангелис]

Interesting:

There is such a wonderful function as StringConcatenate.

In your case it will look like this (by the way, I solved similar problems in MT4 in this way)...

Thanks, it helped.

[Mykola Demko]

Wangelys:

I did it, now it only fails on line 20: implicit conversion from 'number' to 'string' WeekPattern-Export.mq5 20 24

20 34, 20 43, 20 54

You can cite types, you summarise different types, namely string and int.

row += "Open"+(string)i+",High"+(string)i+",Low"+(string)i+",Close"+(string)i;

although yellow lines are not errors but warnings. If you do not edit anything, it will still work.

There are also functions from the "Data Conversion" section of the Reference Manual for type conversion.

[Вангелис]

Urain:

You can give types, you summarise different types, namely string and int.

although the yellow lines are warnings, not errors. If you don't change anything, it will still work.

There are also functions from the "Data Conversion" section of the Reference Manual for type conversion.

Thank you. Also an option.
But, judging by the description in the reference book, StringConcatenate gives a more elegant solution...

[Evgeniya]

URGENT!!!

Please help me to do a credit assignment in NeuroSolutions for a fee

Solve the classification problem for the data shown in the figure (figure in the attached file), where x - corresponds to instances of men, o - women.

ICQ: 367767066

[5Ro7j8t]

I am trying to connect a DLL to MT4 using this article, I have encountered this problem, the network connects successfully if it was created with NeuralExpert, but if I create it with NeuralBuilder, I get the error "Failed to load weights from file . . .". What can be the problem ? (I create a recurrent network).

[jcbrosse2]

Rosh:

New article Connecting NeuroSolutions Neuronets is published:

Author: Andrew

Rosh:

New article Connecting NeuroSolutions Neuronets is published:

Author: Andrew

Dear, 

I just read your article and found it very interesting. And I would like to ask if you can provide further information about 

 how to work this artificial intelligence  with which different ways of strategies to integrate ? 

well i suppose statistic ? and how about technical strategies Elliot waves & Fibonacci ? 

 

Best regards,

 

jean charles Brosse 

[Artem Ipatov]

Great article!!!

Using the NeuroSolutions package is a good idea, but there is only one BUT - the software is not free. Where to get the full version, maybe someone knows? ;)

EDITED: I couldn't find version 6, but version 5 of Developers Unity is available on the net ;).

[Artem Ipatov]

I noticed such a peculiarity - if you write data to a file using the script given by the author, some records in the file look like this:

0.0007200000000000001, 0.0006800000000000001

Or like this:

0.3e-005, 5e-005.

In the first case, we can get some slowdown in NS learning due to a bunch of extra digits. About exponential writing of numbers in the second case - I don't know if NS can correctly perceive such format.... Most likely yes, but I would like to hear experts' comments on this issue)

I think it is more correct to use the DoubleToString function instead of NormaliseDouble. It allows you to bring all numbers to a strict form.