Gathering a team to develop an IO (decision tree/forest) in relation to trend strategies - page 16

New comment
 
Looked at your signals ... I have a different approach to the market ..."as little as possible to lose on a mistake" - there's only a stop or dances with locking or hedging . But the idea of making more profit is interesting! But it also has many pitfalls. The market is crazy in its moods! Why did I look into the subject ... I am launching my project based on my ATS. I do not use signals from it. The project is interesting and promising and here investors from the crypto industry - would like to see a team, but to collect the team is really hard)))
 
Aleksey Vyazmikin:

Great! How does this information relate to this thread?

don't ask)

Sorry, I'm rambling.

 
Fast528:

don't ask).

Sorry, I'm rambling.

Alexei can set up a team of amateurs in riddlcal?

 
Aleksey Vyazmikin:

Thanks for the suggestions!

Is the forum site a server or something else?

Is the engine to bind Python and R together, or with MQL5?

Please tell me a bit more about your generous offers!

I have new versions of the site and programs, want to run the forum and debug the programs.

Primarily interested in revitalizing the resource, testing and refining freeware engine, so direct participation in your project is not planned, but if it happens, I will focus on its components and commercial plugins, which I also do based on the same engine.

In MQL the engine can be used as NamedPipe server, which receives and plays source code blocks, with the ability to visualize errors and results.

Here is a primitive example of MQL script which runs Python and teaches the neural network multiplication table by 2

//+------------------------------------------------------------------+
//|                                                        hl_py.mq4 |
//|                        Copyright 2013, MetaQuotes Software Corp. |
//|                                              http://www.mql5.com |
//+------------------------------------------------------------------+
#property copyright "Copyright 2013, MetaQuotes Software Corp."
#property link      "http://www.mql5.com"
#property version   "1.00"

static string python =
"import os\n"
"import sys\n"
"#print('Python',sys.version)\n"
"x = [[2,1],[2,2],[2,3],[2,4],[2,5],[2,6],[2,7],[2,8],[2,9]]\n"
"y = [[2],[4],[6],[8],[10],[12],[14],[16],[18]]\n"
"f = 'c:/tmp/.pnet.nnb'\n"
"pnet=PyPnet()\n"
"if (pnet):\n"
"  if (os.path.isfile(f)):\n"
"    pnet.LoadFromFile(f)\n"
"  else:\n"
"    pnet.fit(x,y)\n"
"    pnet.SaveToFile(f)\n"
"  for ix in x: print(' {:2.2f}*{:2.2f}={:2.2f} '.format(ix[0],ix[1],pnet.Predict(ix)[0]))\n"
"  del pnet";

//+------------------------------------------------------------------+
//| MQL Script program start function                                |
//+------------------------------------------------------------------+
void OnStart()
  {
   int num=0, res=0, handle=FileOpen("\\\\.\\pipe\\Hlaim.Application", FILE_READ|FILE_WRITE|FILE_BIN);
   if(handle!=0) {
     FileWriteString(handle, python, StringLen(python));
     FileFlush(handle);
     while(res<=0 && num<100) {
       Sleep(1000);
       res=(int)FileSize(handle);
       if(res>0) {
         Print(FileReadString(handle, res));
         break;
       }
       num++;
     }
   }   
  }

screenshot of the results

The example uses a native P-net neural network, its library is embedded in the engine itself, all external libraries are installed in the usual way in the Python environment which is referred to by a link in the settings.

 
Oleg Mamchenko:
I`ve looked at your signals ... I`ve got a different approach to the market ..."as little as possible to lose on the error" - I`ve only got a stop loss or dances with locks or hedging. But the idea of making more profit is interesting! But it also has many pitfalls. The market is crazy in its moods! Why did I look into the subject ... I am launching my project based on my ATS. I am not trading signals from it. The project is interesting and promising and here investors from the crypto industry - would like to see a team, but it's really hard to assemble a team)))

My signals are from a different approach, not related to MO in any way.

To lose as little as possible on an error is just pulling up a stop in the loss zone, and also moving to breakeven on certain patterns.

I wish you success with your project!

However, I still do not understand your intentions concerning this branch.

 
"Still don't understand your intentions regarding this thread though." - looking for my people on the team )) and good luck to you!
 
Ivan Negreshniy:

I have new versions of the website and programs, I want to run the forum and debug the programs.

I see, do you have some kind of website and forum engine? Maybe you can present it then? So far it is not clear what we are talking about.


Ivan Negreshniy:

In MQL, the engine may be used as a NamedPipe server that receives and plays blocks of source code, with the ability to visualize errors and results.

Here is a primitive example of MQL script that runs Python and teaches the neural network multiplication table by 2

screenshot of results

The example uses a native P-net neural network, its library is built into the engine itself, all external libraries are normally installed in the Python environment which is referenced in settings.

To put it more primitively, you have functionality capable of passing commands from MT5 to Python for further execution there, am I correct?

However, how do you plan to use this? Automatic retraining of the model after a certain time, or something else?

I'm interested only in one theoretical possibility - the use of remote agents, but I guess this possibility is not feasible in your architecture?

 
Fast528:

Alexei can set up a team of amateurs in riddlcal?

I looked at their website, but somehow I didn't get the impression at all - I didn't understand their concept.

Can you tell me in your own words what's so remarkable there?

 
Oleg Mamchenko:
"However, never understood your intentions regarding this branch." - looking for my people on the team )) and good luck to you!

How are you looking for where your offer of cooperation is?

 
"How are you looking, where is your offer of cooperation? "I have already written once ... they made a fuss. so if you're interested, please contact me in person.
1...91011121314151617181920
New comment