Roffild's library - page 7
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
The porting of Alglib to MQL5 is done by MQ themselves.
Alglib in the free version runs in a single thread. From the description, the fixes are mostly for the paid multi-threaded version.
I stopped scaffolding in Alglib because of limitations of the free version. Apache Spark handles scaffolding much faster if the input data doesn't fit in RAM.
Now I'm going back to using neural networks, because I've encountered the insurmountable problem of scaffolding: you can't just tweak the forest without fully generating a new one.
The porting of Alglib to MQL5 is done by MQ themselves.
Alglib in the free version runs in a single thread. From the description, the fixes are mostly for the paid multi-threaded version.
I stopped scaffolding in Alglib because of limitations of the free version. Apache Spark handles scaffolding much faster if the input data doesn't fit in RAM.
Now I'm going back to using neural networks, because I've encountered an insurmountable problem with scaffolding: you can't just tweak a forest without fully generating a new one.
yeah, maybe it's for multithreading... not really clear. Serious libraries yes, no one's disputing that, it's just that it's sometimes more convenient and faster without leaving the cash register (mt5)
In the library added work with MLPDataFile from Python.
buildall_and_tests.py - Script to build all library components and run tests.
Who has already tried running Agents on amazon?
Please clarify the situation. The developers said that under Linux you can't run agents as they are essentially Windows services, how did you solve this problem?
And, is it possible to have a GUI to work with so many cores? I'm aiming to run a script on R, it looks reasonable for the money, but it's not clear how to organise it and make it work.
Please clarify the situation. The developers said that under Linux you can't run agents because they are essentially Windows services, how did you solve this problem?
And, is it possible to have a GUI to work with so many cores? I have a goal to run a script on R, it looks reasonable for the money, but it's not clear how to organize it and make it work.
It is possible to run agents under Linux, but it is unlikely they will be paid for providing them to the MQL5 Cloud.
About GUI and R I don't understand at all, because MT5 and its "Agents" tab are used.
It is possible to run agents under Linux, but it is unlikely to get money for them when providing them to the MQL5 Cloud.
I don't understand about GUI and R at all, because MT5 and its "Agents" tab are used.
I see, Wine is used? I don't know if it's the same for this type of server or if it's possible to use graphical interface to manage the server. Actually I am interested in using these servers for R calculations, so I have more questions, as someone who has understood everything and actually uses this service.
I see, is Wine being used? Just further somehow you have everything configured without a graphical shell, so the question arises, is this the type of service for this type of server, or can you somehow see a normal "Desktop" and use a graphical shell to manage the server. Actually I am interested in using this server for calculations on R, so I have more questions, as a person who has understood everything and actually uses this service.
I have noticed that wine restricts my work with market and some other services related to terminal-MQL connection (may be it's an IE issue). Everything else works fine. It's even possible to set up requests within a local network, or via a remote URL/IP.
The use of wine has a limitation in the work with the market and some other services related to the work of connecting the MQL account to the terminal (this is my observation, most likely a problem with IE). Everything else works fine. It's even possible to set up requests within a local network, or via a remote URL/IP.
That's good, so I was wrong.
That's good, so I was wrong.
Wine is quite flexible. But not everyone knows how to cook it. It's easier to use a virtual machine