-
**edit** See [this important issue](https://github.com/FluxML/ML-Coordination-Tracker/discussions/38)
The measures part of MLJBase could do with some TLC. It is not the shiniest part of the MLJ c…
-
**Is your feature request related to a problem? Please describe.**
Opportunity to improve model accuracy with ability to ensemble TabNet model (deep learning) with existing set of models.
**Descr…
-
Hi all, I've been playing around with the NGBRegressor, and came across a weird problem: when I put col_sample to below 1, shap_values plus the expected value don't add up to the prediction. I tried t…
-
Hello,
This related in part to #1025. The uncertainty coming from ngboost is aleatoric uncertainty; to get epistemic, a simple way is to bootstrap. Since this is a `post-processing' method, is ther…
-
### Problem Description
Release wheels for python 3.12.
Work-in-progress PR: #3414
Core dependencies:
- [x] numba: https://github.com/numba/numba/issues/9197
Test dependencies:
- [ ] cat…
-
#### Describe the workflow you want to enable / Describe your proposed solution
add an `base_estimator` argument to GradientBoostingRegressor/Classifier and HistGradientBoostingRegressor/Classifi…
-
Before I dive in and start using this codebase, I have the following question. In the paper, it suggests that, although we have freedom over the domain of the dependent variable, we have $x \in R^n$. …
-
I was wondering if a partial_fit method is possible for this algorithm similar to some of the scikit-learn API models or analogous to the update method of Light GBM.
-
### Describe the workflow you want to enable
Scikit-learn has a `predict` and `predict_proba` method for Classification classes but only a `predict` method for regression, with the option of quanti…
-
### Issue Description
The is a bug in shapley values for NGBoost when col_sample < 1.
It seems that shapley values excludes some features when col_sample