What formula is used to calculate the "importance of features"?

Question

Machine learning

Using the code xgb.plot_importance (model) I got the plot shown in the example.
How is the importance of features calculated? It needs to be described mathematically. Is there a calculation formula? Very necessary!!!!!!!!!! Haven't found anything on the internet yet.

Answer 1

But why do you say that you received the schedule?
But in the same article where you took "your" example from:
https://machinelearningmastery.com/feature-importa...
there is a link to the book and the page in it, which describes "how feature importance is calculated in boosted decision trees ,"
Here is also a link to a publicly available book with a corresponding description:
https://forums.fast.ai/t/feature-importance-of-ran...
You can also read it here:
https://datascience.stackexchange .com/questions/87...
https://towardsdatascience.com/be-careful-when-int...
If you really "really need" - then master it. And yes, learn to search on the Internet. Come in handy.

Answer 2

I took this graph as an example, the idea is the same
, yes, but the code is described there. And I need a mathematical implementation. Probably met somewhere.
In any case, thanks for the help! I'll try to figure it out.

Answer 3

See here how you can find the importance of features.