-
New test {{runit_XGBoost_bernoulli.R}} throws this error:
prostate.h2o@model$init_f not equal to `f0`. target is NULL, current is numeric
This is because XGBoostOutput doesn't contain the field …
-
Add pred_interactions to the predict method for XGBoost in H2O and SW.
-
I'm working on a script that is supposed to explain the results of an XGBoost ML-model. The model contains, among others, one feature with a large number of categorical levels (approx. 2000). I notice…
-
-
I'm wondering if you've considered allowing the user to pass in a separate training sets for the xgboost model vs the survival model?
For example, in XGBSEStackedWeibull, the current state is this…
-
I am getting this error while fitting Xgboost in Explain mode
![image](https://user-images.githubusercontent.com/61991627/177485571-600e7f2b-3044-4851-a530-261090d770d0.png)
-
## Background
We should also make sure that our documentation is kept up to date.
A scour through the open issues in this repo and also on [StackOverflow](https://stackoverflow.com/questions/tag…
-
Is there SHAP implementation for ranking models? specifically for xgboost model with ndcg ranking? 10x
-
Dear xgboost developers, I am "testing the waters" to see if some of my research is interesting and potentially could be implemented in xgboost either now or in the future.
**TLDR:**
Approximate t…
-
In the jvm-package, the NativeLibLoader extracts the library from the jar to the java.io.tmpdir. We prefer to be able to control that path for the xgboost library independently.
Thanks!