Machine learning in trading: theory, models, practice and algo-trading - page 1164
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 am struggling with catbust, I can not understand what metric is better to use to select a model from this list for classification?
For myself so far I see an interesting variant of the formula %Regular from all*%Of target 1, but I do not see something similar there.
Accuracy
Accuracy
This does not work if the number of ones and zeros is unequal, as I have for example....
This does not work if the number of ones and zeros is unequal, as I have for example....
BalancedAccuracy
I don't know how... somehow :) well, there are criteria for selecting models - at least this minimum, MSUA, again, I recommend to read
I.e. in your language - choose the best one among a lot of indicators... already something :)I haven't got to MSUA yet, I'm reading a lot, but I'm still in the thick of mathematics again
about the indicators - that's a great idea! but you have to look at it from another angle... I need to visualize your examples, or rather visualize the learning process
BalancedAccuracy
I don't quite understand what are the settings there, what are the coefficients, how do I apply them correctly?
I haven't gotten to MSUA yet, I'm reading a lot, but I've gone back to the maze of mathematics.
about the indicators - that's a great idea! but you need to look at it from another angle... You need to visualize your examples, or rather visualize the learning process
Roffild did it here, I think... he visualized forest as an indicator
I do not quite understand what are the settings there, what are the coefficients, how to apply them correctly?
I dunno, I did not use )) maybe something simpler? AdaBoost or LightGBM... They're all the same thing, in essence. As for boosting, so is boosting
I don't know, I haven't used it)) maybe something simpler? AdaBoost or LightGBM... It's all the same thing, basically.
As far as I know, they can't be run from the command line, you need python or R, which is difficult for me... And we're talking about metrics, I thought it's a standard thing and doesn't depend on model creation package...
As far as I know, they do not have the ability to run from the command line, you have to put python or R, which is difficult for me... And we're talking about the metric, I thought it was a standard thing and does not depend on the model creation package...
well the metric number of correctly guessed zeros divided by the total number of zeros. the same for units, then the total multiplied by 0.5
weighted metric it turns out
well, the metric is the number of correctly guessed zeros divided by the total number of zeros. the same for units multiplied by 0.5
the weighted metric comes out.
And the coefficient for shifting the weights? As in, if the most important 1, their weight is greater, or what?