Machine learning in trading: theory, models, practice and algo-trading - page 2394
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
I wonder if anyone has tried to useML.NET? Well, unlike ONNX, MT5 already has support for .NET. I wonder if this is a workable option when running an EA on a VPS.
Microsoft products are not very popular, tensorflow, apache and others are used more often. + python
If some complex intolerable (not morally, but physically) models, then the best combination now is python + mt5 python api, in my opinion
Microsoft products are not very popular, more often use tensorflow, apache, etc. + python
If some complicated intolerable (not morally but physically) models, then the best combination now is python + mt5 python api, in my opinion.
I am interested in running such a combination on VPS. I want to do without compilation on server and installation of additional programs with their long configuration. Ideally - just copy a single .ex5 file to a server and run it as usual.
The question of running such a bundle on a VPS is interesting. I would like to do without compiling on the server and installing additional programs on it with their long configuration. Ideally, just copy a single .ex5 file to the server and run it as usual.
Well, install python + some packages and the bot itself in python, without .ex5. Minutes 15
maybe through a docker even fasterWell installing python + some packages and the bot itself in python, without .ex5. 15 minutes
Is it already possible to write EAs in python? Or is it a looped script?
Is it already possible to write EAs in Python? Or are we talking about a looped script?
looped, in fact, it does not differ from onTick or onTimer in EAs
In fact, it is no different from onTick or onTimer in Expert Advisors.
I'm afraid so far, I'm waiting when fxsaber gives his blessing to this approach)
I'm afraid for now, I'll wait until fxsaber blesses this approach)
hee... )
I am transferring my philosophical question about classification algorithms here from Aleksey Vyazmikin's thread about combining segments:
I am interested in the question, how to choose an algorithm and its training properly take into account that the classes can be a) clearly separated from each other, b) mixed, c) some mixture of points a) and b). For (a) you need a clear classification (answers are either 0 or 1), for (b) you need a fuzzy classification ( answers are numbers on a segment from 0 to 1) , and for (c) you need to mix them somehow, but not shuffle them.
I am transferring my philosophical question about classification algorithms here from Aleksey Vyazmikin's thread about combining segments:
I am interested in the question, how to choose an algorithm and its training properly take into account that the classes can be a) clearly separated from each other, b) mixed, c) some mixture of points a) and b). For (a) you need a clear classification (answers are either 0 or 1), for (b) you need a fuzzy classification ( answers are numbers on the 0 to 1 interval), and for (c) you need to mix them somehow, but not shake them up.
Well, this sounds like pointless tantrums with about this outcome
For clearly separable may sometimes be better suited SVM, but otherwise boosting or NSWell, it sounds like a pointless torment with about this outcome.
It makes sense) a philosophical question is a philosophical answer)
The thought was something like a composition of classification algorithms. For example, fuzzy classification is done by the K nearest neighbor method, but some neighbors are discarded by another algorithm with clear classification. If this makes any minimal sense, there must be some articles on this topic on the web)