Machine learning in trading: theory, models, practice and algo-trading - page 2941
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
And I think the Market is the main thing here.
Everything else is solvable.
That's what I wrote above
Between them they communicate perfectly well via sockets, via files, via SqLite ....
Now everything comes down to a running MT5 + a running script with TensorFlow and ALL!
They communicate perfectly well with each other via sockets, via files, via SqLite ....
For a long time, using the MetaTrader5 package for Python, you can run *.py scripts directly in the terminal with direct access to data.
It has long been possible to run *.py scripts directly in the terminal with direct access to data using the MetaTrader5 package for Python
UPDATE, I found it, I'm getting into it...
I think the Market is the main thing here.
I wrote only about myself, I am not interested in the Market.
Everything else is solvable.
Solvable, of course, but by very crutch methods, which are not universal, inconvenient and collapse even with a not too large number of TCs and frequency of their change.
It seems so obvious to me the benefit of testing and trading a TS assembled in one single file and without third-party software that I am not even ready to seriously defend this approach. This is the case when "if you need to explain, you don't need to explain".
It has long been possible to run *.py scripts directly in the terminal with direct access to data using the MetaTrader5 package for Python
here we are talking about "receiving stock exchange information via interprocess communication directly from MetaTrader terminal "
it's not Python in the terminal
I think the Market is the main thing here.
Everything else is solvable. If only for trading, MO can be connected to MT5 without any problems, but you can't get to the Market that way.
That's why on the scales there is dancing with tambourine on one bowl and the Market on the other.
And what will you sell? An expert? A trained model? How long will it run? How do you retrain it? How to maintain it? Although there are options here too... You don't need to install Python. You don't need to learn Python and everything related to MO.
This integration enables developers to test the model in a quality way.
It's not a fact that you need to go the way the developers have shown-all in the meta-editor. But it's who can do what, it's not important.
What is important is that research, development, debugging and optimisation of the model are separated into a separate stage. The result of which will be the best model in a universal format ready for implementation on any platform, not only on the one on which it was developed. Besides, you don't need Python to work.
You need to wait for the final solution and try it.
About models in R. At least these packages can have a converter: Catboost, torch, H2O, XGboost. It is necessary to ask on Githab.
here we are talking about "receiving stock exchange information via interprocess communication directly from MetaTrader terminal "
is not Python in the terminal.
I don't understand the thought. Can you elaborate on this - getting stock exchange information through interprocess interaction directly from MetaTrader terminal. Just curious ....
The proposed integration has one big flaw. It does not know how to "talk" with MKL5 programmes. Only with the terminal. Therefore, there are variants of solving this issue.
I don't understand the idea. Can you elaborate on this - receiving stock exchange information via inter-process interaction directly from the MetaTrader terminal? Just curious .....
The proposed integration has one big flaw. It does not know how to "talk" with MKL5 programmes. Only with the terminal. Therefore, there are variants of solving this issue.
This is a quote from the documentation on working with Python
The MetaTrader for Python package is designed for convenient and fast retrieval of stock exchange information via interprocess communication directly from the MetaTrader 5 terminal.
And it was said that you can run it in the terminal itself, I didn't realise that.
here we are talking about "receiving stock exchange information via interprocess communication directly from MetaTrader terminal "
is not Python in the terminal.
Python scripts can be run directly in the terminal (not in the tester) for a long time:
Information to broaden everyone's horizons: