-
Notifications
You must be signed in to change notification settings - Fork 175
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add SHAP calculation to GBT regression #1399
Conversation
336cb94
to
4a6a507
Compare
aca4498
to
ca14608
Compare
i would also add example and documentation explaining new API |
Added |
|
@napetrov done - updated the RST |
Should we consider adding MB examples to separate directory? at same level as scikit/daal4py? @razdoburdin - might be you can share thoughts as well The problem with keeping in daal4py examples is that it's hard to find, while in case of modelbuilders separate folder we can put all MB related examples in same place. I think that majority of users are not overlapping between MB use and daal4py use |
That is a good idea. In my mind the structure of examples should reproduce the structure of the python imports. Currently we are not close to this structure. But if we want to move in this direction, we should place all the model builders related examples in examples/daal4py/mb folder. |
@napetrov @razdoburdin I have created the |
Yes, they are all in place. |
i would keep at same level as daal4py/scikit . Although technically this is part of daal4py but from usage scenario it ether of those. The only question is this should be model_builders but ether "Model builders" or MB? The point - there is no places there we have model_builders format used. |
Don't have a strong opinion here @napetrov. Do you want me to rename the directory to |
I'm not sure on corrects one - but version with underline doesn't match other usages. mb would be consistent with module name |
Co-authored-by: Alexandra <alexandra.epanchinzeva@intel.com>
f818a77
to
c8b2a69
Compare
Co-authored-by: Nikolay Petrov <nikolay.a.petrov@intel.com>
Pendant to oneapi-src/oneDAL#2460. This PR enables to new SHAP value functionality in scikit-learn-intelex.
The API is going to be similar to XGBoosts predict() method