Discussion of article "Multilayer perceptron and backpropagation algorithm (Part II): Implementation in Python and integration with MQL5"
I haven't read the whole thing (yet?) ;)
But this one I can't help but respond to.....
"To work with TensorFlow you need to install a version above 3.3 and below 3.8; I'm using version3.7."
What the hell? ???
TensorFlow works better for me on python version 3.9....
On lower versions you can't do without tambourines!!!!
How many tambourines were broken, let's not count.... OK?
Artigo sobre Python e MQL e nem uma única linha coda.
A versão em russo está quebrada.....

- www.mql5.com
I haven't read the whole thing (yet?) ;)
But this one I can't help but respond to.....
"To work with TensorFlow you need to install a version above 3.3 and below 3.8; I'm using version3.7."
What the hell? ???
TensorFlow works better for me on python version 3.9....
On lower versions you can't do without tambourines!!!!
How many tambourines were broken, let's not count.... OK?
nenhum problema com a utilização da versão 3.7
Ate o momento em que o artigo foi escrito as versões oficiais no site eram 3.3/3.8, hoje estamos em 3.6/3.9
I wanted to clarify this thought of the author -" use the built-in functions in Python that open and close positions, but for this scenario we will not have many possibilities that MQL offers us".
What MQL features are we talking about?
And the second question - do you work under Linux?
What MQL features are we talking about?
And the second question - do you work under Linux?
1 - For example, in Python we don't have Tick and Book events, which forces us to use an infinite loop in the code, which I find particularly horrible. Other functions such as OnTradeTransaction are also not available, which forces you to put more effort to find out if a position has been fully or partially filled, on our stock exchange (B3) there are many partial fill situations on exit.
2- I don't work in Linux, I use Windows. I only use Linux in my work.
New article Multilayer perceptron and backpropagation algorithm (Part II): Implementation in Python and integration with MQL5 has been published:
Author: Jonathan Pereira

- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
New article Multilayer perceptron and backpropagation algorithm (Part II): Implementation in Python and integration with MQL5 has been published:
There is a Python package available for developing integrations with MQL, which enables a plethora of opportunities such as data exploration, creation and use of machine learning models. The built in Python integration in MQL5 enables the creation of various solutions, from simple linear regression to deep learning models. Let's take a look at how to set up and prepare a development environment and how to use use some of the machine learning libraries.
First, you should download Python from the official website www.python.org/downloads/
To work with TensorFlow, you should install a version between 3.3 and 3.8 (I personally use 3.7).
After downloading and starting the installation process, check the option "Add Python 3.7 to PATH". This will ensure that some things will work without additional configuration later.
A Python script can then be easily run directly from the MetaTrader 5 terminal.
Open MetaEditor and go to Tools \ Options.
Specify here the path at which the Python executable is locates. Note that after installation it should have the default Python path. If not, enter the full path to the executable file manually. This will allow you to run scripts directly from your MetaTrader 5 terminal.
I personally use a completely separate library environment called virtual environment. This is a way to get "clean" installation and to collect only those libraries which are required for the product.
Author: Jonathan Pereira