-
-
I'm using LGBMRegressor with sklearn.model_selection.GridSearchCV, with cross-validation split based on sklearn.model_selection.GroupKFold. When I include early_stopping_rounds=5 in the estimator, I g…
-
I am using a python version of LightGBM on a Mac OSX system and I am having some issues with prediction functions. In short I am not able to reproduce the validation result using predict()/predict_pro…
-
Dear Nicolas Hug,
first of all i want to tell you that your library is interesting.
However, I just checked your library and see that the train test split is not on the level of the user or what…
-
##### System information (version)
- OpenCV => 4.1.2-dev
- Operating System / Platform => ArchLinux 5.3.11-arch1-1 x86_64 GNU/Linux (31.0 GiB)
- Compiler => Eclipse/Cmake
##### Detailed descri…
-
these degenerate cases are sometimes wanted for comparison
-
```julia
using DataFrames, Lasso, LassoPlot
X = rand(10, 20); y = rand(10);
path = fit(LassoPath, X, y);
plot(path)
```
results in
```julia
┌ Warning: `setindex!(df::DataFrame, v::AbstractVec…
jbrea updated
4 years ago
-
When adding loo to a model with a bym2 correlation structure if `reloo = FALSE` the output is normal and what is expected. However if `reloo = TRUE` the model is refit the right number of times but af…
-
**will you add cross_validation.KFold in turicreate or not?**
@igiloh @Kagandi @znation @srikris @hoytak @afranklin
_Originally posted by @hri-tikkumar in https://github.com/apple/turicreate/issue…
-
Hi,
Do you think you have time to create a docker image with everything installed?
It's very difficult to create the original working environment with conda as most the packages are updated.
Fo…