Machine learning in trading: theory, models, practice and algo-trading - page 1333
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
Maxim, I swapped the sample places - for training and validation, the test left - what will be the result based on scientific dogma? I myself do not know yet, the processing is not yet complete.
None, it won't do anything.
And what was expected in general?No, it will not do anything.
what was expected in general?I expect that at least the models will not be similar to those before the swap sample places.
And the percentage of accumulation of the best options will change. I think that content is more important than volume.I expect that at least the models will not be similar to those before the swap of the sampling places.
So they won't look alike, so what then? )
So they won't look alike, so what then? )
This will be good, because then you can use them in parallel.
This will be good, because they can be used in parallel then.
Yes, we are shooting in the sky, i.e. the optimal solution has not been found.
More seed change and a bunch of garbageit looks like we need to pump R instead of python, Renat wrote that soon there will be a direct link without crutches
i.e. catbust will be able to run in 1 line from mt5
I must have missed that message. Do you know when it was or the link?
Good luck
I must have missed that post. Can you tell me when it was or the link.
Good luck
Yeah, it was in passing right here.
https://www.mql5.com/ru/forum/302625/page8#comment_10568532
Well, yes, a finger in the sky, i.e. the optimal solution is not found
change the seed and make a pile of garbageWhat do you mean finger in the sky? Two models running in parallel is a good option, why not?
Seed, of course, will continue to change, until I hear sane arguments - what difference does it make who rotates it I or a random number generator - is not clear. Soon there will be more models with parameter"--bagging-temperature" - probably also garbage according to your classification, but I'm interested to see.
What do you mean by a finger in the sky? Two models running in parallel is a good option, why not?
Seed, of course, will continue to change, until I hear sane arguments - what difference does it make who rotates it I or a random number generator - is not clear. Soon there will be more models with parameter"--bagging-temperature" - probably also trash according to your classification, but I'm interested to see.
If they work, why not?
The point is that you can increase the efficiency many times over, applying very simple methods
I do not write about them openly, because I want you to read the book. And in general, there's a lot of buried, which can lead directly to good models. But just shhhh...
about the banging temperature is interesting to see, I don't know what you meanif they work, why not?
It's just that you can increase the efficiency many times over by applying very simple techniques.
I intentionally do not write about them openly, because I want you to read the book. And in general, there's a lot of buried, which can lead directly to good models. But just shhhh...
About the bagging temperature it's interesting to see, I don't know what do you meanI looked up MSUA, I don't know which book it refers to specifically, but it's not searchable with that title. As I understand it, this thing is used in CatBoost
--l2-leaf-reg.
l2-leaf-regularizer.
L2 regularization coefficient. Used for leaf value calculation.
Any positive values are allowed.
CPU and GPU
Or is it about something else? Also this method can be used for creating predictors themselves, for example to describe the patterns in certain areas.