Machine learning in trading: theory, models, practice and algo-trading - page 925
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
How not to retrain the model, who has any guesses?
early stop, regularization, dropouts... everything works. In fact, I use all 3.
Completed at the top. Tried a lot of things. Regularization, something I read I'll have to look into. If it is not difficult what is it?
Confused dropouts with cross validation (cross validation shows the same thing)
Completed at the top. Tried a lot of things. Dropouts show the same thing not much different. Regularization, something I read I'll have to look into. If it's not difficult what is it?
Idon't know whatf1_score is.
In neural networks, it's a decrease in the degree of the polynomial that interpolates the f-scores. I think this is what it sounds like scientifically.
What kind of NS do you have, where are you from? or the forest
I don't know whatf1_score is
In neural networks, it is a decrease in the degree of the polynomial by which the f-value is interpolated. I think that's what it sounds like scientifically
What is your NS, where is it from?
https://msdn.microsoft.com/ru-ru/magazine/dn904675.aspxhttps://en.wikipedia.org/wiki/F1_score
I don't know whatf1_score is
In neural networks, it is a decrease in the degree of the polynomial by which the f-value is interpolated. I think that's what it sounds like scientifically.
What's your NS, where is it from? Or the forest.
https://msdn.microsoft.com/ru-ru/magazine/dn904675.aspxLes.
https://en.wikipedia.org/wiki/F1_score
Les.
in general, the closer the values on the train and the test the better
Mm-hmm. But, I know that and it's not hard to guess)
Uh-huh. But, I know about it and it's not hard to guess)
So you can see that you have retrained
So far I can not say exactly, because I played a lot of parameters improvements were about 0.1. Maybe there are some techniques, which I do not know, that's why I asked.