Machine learning in trading: theory, models, practice and algo-trading - page 1198
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 need to communicate - videos are common, I do not argue.
Here I still could not solve the problem of splitting calculations into clusters - R has a library, but how to activate it without much code modification, and I have not found someone who would do it...
http://ods.ai/
what about the group in discord, by the way, is there any activity?
Does anyone have any idea how to describe such curves? It's all about creating predictors....
In general, looking at these curves want to make a semi-automatic machine that will achieve improvement in all important indicators - slow learning and run from a stopping point with a random selection of predictors to build a tree, and so iterations of 1000 and from them choose the best option, because this approach of learning by logloss does not seem right, looking at the graph of changes in other indicators.
http://ods.ai/
In general, I do not understand what it is - a legalized group of persons ...
In general, I do not understand what it is - a legalized group of persons ...
It is difficult with you)
It's hard with you )
It's true :)
I can't finish a little bit for half a year - my interest has dropped.
So without me, at least for now.
Illusions. And you understand that. That's why interest has fallen.
Does anyone have any idea how to describe such curves? It's all about creating predictors....
What do you mean, how do you describe them? Should they go up, should they go down? What do you want them to do?
Illusions. And you understand that. That's why interest has fallen.
What do you mean, how do you describe it? Should it go up, should it go down? What do you want from it?
+Some indicators go up and some go down... While the idea of controlling the maximum drawdown and counting the number of points within 2\3 of growth comes to mind - this will all be indicative of the end/degradation of the learning curve. I'll add points of extremum as a percentage of time and maybe a slope factor...what else is possible?
Some indicators go up and some go down... While the idea of controlling the maximum drawdown and counting the number of points within 2\3 of growth comes to mind - this will all indicate the end/ deterioration of the learning curve. I'll add points of extremum as a percentage of time and maybe a slope factor...what else is possible?
I don't know if it would help, but! You could try correlation. Imagine the ideal result that you want in the form of the curve you want, then in the process of finding the best model, compare the current result with the ideal (your ideal curve) by calculating the correlation. The model with the closest correlation to the ideal will be as close to the model you want
Great blog!!!
I advise everyone to browse throughhttps://smart-lab.ru/my/egenui/blog/all/