Machine learning in trading: theory, models, practice and algo-trading - page 409

 

Hi people!...I'm a machine...I'm tired of fighting I want to learn Forex...I came to you to learn... Will you take me?... I have skills... I can go to the future if you need me....



 
Vladimir Perervenko:

As a rule, everything is done in R, except for a few enthusiasts who try to rewrite some models in MCL. As for packages - it depends on what model one uses - trees, neural networks, deep neural networks, etc.

Which model are you interested in?

Good luck

My solutions were initially based on OpenNN (C++).
Now I've learned a lot and switched to Keras (Python) .

Mainly I deal with deep learning of classification problems prediction.
 
nowi:

Hi people!...I'm a machine...I'm tired of fighting I want to learn Forex...I came to you to learn... Will you take me?... I have skills... I can go to the future if i need it....

Go to the store...)
 
elibrarius:
But the RNN has error = 0, on the learning curve. That's cool, too. I'll probably blog my experiment

Here's the RNN experiment

https://www.mql5.com/ru/blogs/post/697984

RNN переобучается или недообучается?
RNN переобучается или недообучается?
  • 2017.06.13
  • elibrarius
  • www.mql5.com
Я немного поэксперементировал с RNN и похоже, что она просто запоминает примеры обучения (важные в связке с шумовыми предикторами), а на новых данных шумовые предикторы портят результат. Т.е. RNN...
 

Machine learning theorists, how's it going? Any results?

Is the neuro-bot ready?


Or are you joking from Shikotan to Konigsberg?

 
elibrarius:

Here's the RNN experiment

https://www.mql5.com/ru/blogs/post/697984


Oh my God, you should at least write that it's ReshetovN and not RNN, because I can't understand anything :) RNN is a recurrent neural network, which has nothing to do with what you did

Thanks for the test, it's informative. Write that it is an expert system Reshetov, not RNN

 
Please post the finished neuro-bot. Or tell me what it is like.
 

At least there should be a small result. To make us happy here - that there IS a perfect robot in the world.


Otherwise everything will be in vain.

 
Alexander Ivanov:
Please post a ready neuro-bot. Or tell me what it is like.

Neurobot???? This is something new.... :-) It's even kind of consonant with not working.... :-)
 
Maxim Dmitrievsky:

For God's sake, you should at least write that this is ReshetovN and not RNN, but I read and can't understand anything :) RNN is a recurrent neural network, which has nothing to do with what you did.
Yes, it concerns the Reshetov network, you want to rewrite it in MQL, don't you? So I experimented with it
Reason: