Download MetaTrader 5
To add comments, please log in or register
Follow the news. View our company's official channel!
MetaQuotes Software Corp.
Moderator
177290
MetaQuotes Software Corp. 2016.04.26 12:27 

New article Deep neural network with Stacked RBM. Self-training, self-control has been published:

This article is a continuation of previous articles on deep neural network and predictor selection. Here we will cover features of a neural network initiated by Stacked RBM, and its implementation in the "darch" package.

Structure of a deep neural network initialized by Stacked RBM (DN_SRBM)

I recall that DN_SRBM consists of n-number of RBM that equals the number of hidden layers of neural network and, basically, the neural network itself. Training comprises two stages.

The first stage involves PRE-TRAINING. Every RBM is systematically trained without a supervisor on the input set (without target). After this weight of hidden layers, RBM are transferred to relevant hidden layers of neural network.

The second stage involves  FINE-TUNING, where neural network is trained with a supervisor. Detailed information about it was provided in the previous article, so we don't have to repeat ourselves here. I will simply mention that unlike the "deepnet" package that we have used in the previous article, the "darch" package helps us to implement wider opportunities in building and tuning the model. More details will be provided when creating the model. Fig. 1 shows the structure and the training process of DN_SRBM

Fig.1 Structure and the training process of DN_SRBM

Fig. 1. Structure of DN SRBM

Author: Vladimir Perervenko

Stephane Reynaud
2073
Stephane Reynaud 2016.05.20 19:01  

very very very good article

many thanks for your beautiful contribution

never try the R package before, i try the mql4 expert, at first everything ok, but 3 minutes after the launch (during teaching phase), rterm crash !

i will investigate why ... if you have some ideas ?

Stephane Reynaud
2073
Stephane Reynaud 2016.05.20 19:51  
it's seems as i have a problem with the dll (i see it with debugview)
yanoosh
11
yanoosh 2016.05.27 13:55  
For me it crashes OnTimer. Any hints?
Chengjie Feng
119
Chengjie Feng 2016.05.28 08:13  
Crashed... Did I make mistake?..
sfonti
54
sfonti 2016.07.02 13:49  

theoretically a very good article,

but practically useless because the application is not running, and advice is not to be expected...

sfonti
54
sfonti 2016.07.04 14:09  
sfonti:

theoretically a very good article,

but practically useless because the application is not running, and advice is not to be expected...


I would like to correct my opinion,

after downloading RStudio, a third-party interface for R, and donwloading all needed packages its running without Problems.

RStudio has the advantage that all dependencies are installed automatically.

Nice Work!
Vladimir Perervenko
2233
Vladimir Perervenko 2016.07.11 19:38  
sfonti:

I would like to correct my opinion,

after downloading RStudio, a third-party interface for R, and donwloading all needed packages its running without Problems.

RStudio has the advantage that all dependencies are installed automatically.

Nice Work!

Hi,

Rterm  falls mostly because of lack of necessary libraries and affiliated libraries.
There will be questions, ask.
Good luck

Edisonyk
5
Edisonyk 2016.07.11 22:39  
Great article, I would use the MT5. How do I run the MT5? Someone could share the source code?
Vladimir Perervenko
2233
Vladimir Perervenko 2016.07.14 20:15  
Edisonyk:
Great article, I would use the MT5. How do I run the MT5? Someone could share the source code?
With MT5 does not work. The developers have promised to do API to R. Let's wait.
B1-toan_49
9
B1-toan_49 2016.08.29 14:54  
Replacing OnTimer to OnTick for run strategy tester. The best article I read. Thank for share your idea.
/ /123
To add comments, please log in or register