neural network and inputs - page 9
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 give up)))) I'm out ))))
I'm off to bed, nice talking to you)))
Yeah, you too)))
The beginning of the topic is interesting and then it all devolves into a clarification of points of view.
Figures and proof of one's rightness ended on page 1-2.
Is it realistic to feed?
I tried indices, quotes, cross pairs (with and without correlation), FFT, FHT. The only thing I have tried is to feed the weather....
I have made forecasts too, in various ways, from "tomorrow" to "the strongest wave".
The errors of the trained grids are higher than the practical application.
Further build-up of "power", I think is a dead end.
my opinion, neural nets are a 'thing', but you need to work on them forever. That is, if you decide to use them, you should know: "there will be no free money". It amounts to work for which you will receive your reward.
In my opinion, neural nets are a "thing", but it takes forever to work on them. That is, if you decide to use them, you should know: "There will be no freebies".
Is it real?
I tried indices, kotirs, cross pairs (with and without correlation), FFT, FHT. The only thing I have tried is to feed the weather....
I have made forecasts too, from "tomorrow" to the "strongest wave".
And what do you want to get in the output? It depends on the output and on the task which your network solves.
The NS is a complex.
1)Everyone "reaches" the inputs (some go through them to the gray balls),
2) Some decide on the output, decide on the networking task, its type and architecture,
3) only a few take a serious plunge into network training.
And there is nothing secondary in NS, hence the lack of result
I've tried to train the network many times, but it didn't want to go any further upwards.
So it did work on the feedback, didn't it?
Well, I was a little wrong here, and I hope everyone understands that the NS doesn't want to work on the feedback loop...
Well I was a little wrong here, I hope everyone understands me, that NS does not want to work on the OOS in any way...
And how did you form the inputs? I mean, a lot of people for some reason thinking that the examples laid out in kodobase are tunable skrdovalki, just feed whatever they want from Stel's reference book and run the built-in tester until they're blue in the face ;).
Can you tell me how the inputs were checked? On your fingers.
And then, if this is training with a teacher, what do you expect to get?
Or just: here's some shit to enter and let's go: NS, find me the maximal profit ;) ?
Guys, when you can prepare the data for the grid and select the right teacher, create the right network architecture. Then you won't need a grid anymore )))).
Because you will simply see the patterns with your eyes.
Quite possibly, but the highlighted is not quite right. ;) : What can be seen with the eyes is not always easy to turn into code. The web, on the other hand, makes it possible to work with nonverbalizable algorithms and "not to get bogged down" ;).
Quite possibly, but the highlighted is not quite right. ;) : What can be seen with the eyes is not always easy to turn into code. The net, on the other hand, makes it possible to work with nonverbalizable algorithms and "not to get bogged down" ;).
Well said! I'll have to work with them...