Machine learning in trading: theory, models, practice and algo-trading - page 1887
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
In the early 80's with six employees of the research institute decided to break (or destroy) the backbone of Sportloto (as Yusuf said about forex :).
One of us was a doctor of technical sciences, as well as Ph.D., mathematicians, professionals of different profiles in the field of development of computing complexes.
. . .
*** If >=three participants are interested in how it all ended, then I will continue.
In the early 80's with six employees of the research institute decided to break (or destroy) the backbone of Sportloto (as Yusuf said about forex :).
One of us was a doctor of technical sciences, as well as Ph.D., mathematicians, professionals of different profiles in the field of development of computing complexes.
. . .
*** If >=the three participants are interested in how it all ended, then I will continue.
interesting
In the early 80's with six employees of the research institute decided to break (or destroy) the backbone of Sportloto (as Yusuf said about forex :).
One of us was a doctor of technical sciences, as well as Ph.D., mathematicians, professionals of different profiles in the field of development of computing complexes.
. . .
*** If >=the three participants are interested in how it all ended, then I'll continue.
In another thread, please, there are enough insane people here
interesting
+1
I added sigmoid. In diapason +-1 all activation functions have a better result. Although, it seems, by logic, sigmoid and relu should do better with data in the range 0-1. Perhaps this is due to the fact that increments are fed to the input, I should try to feed non-differentiated prices.
I just want to help the lost developers of Machine Learning. All this has long been passed, tested on real trading and there is no use of MO for forex.
It's a waste of time.
I just want to help the lost developers of Machine Learning. All this has long been passed, tested on real trading and there is no use of MO for forex.
This is a waste of time.
You may help me if you want to, why are you ranting and saying terms?
I just want to help the lost developers of Machine Learning. All this has long been passed, tested on real trading and there is no use of MO for forex.
This is a waste of time.
I added sigmoid. In diapason +-1 all activation fi ions have a better result. Although, it seems, by logic, sigmoid and relu should do better with data in the range 0-1. Perhaps this is due to the fact that the input are fed increments, I should try to feed non-differentiated prices.
13:36 18.07.20 Agent Mulder, end of entry.
The sigmoid should be on the output layer, it's kind of a must if you have classification.
I tried doubling the number of neurons, it always used to be by the number of fics and the quality went up by 3 percent, surprise ))