glum

glum copied to clipboard

Daily run failure: Unit tests

5

The daily unit tests failed. See https://github.com/Quantco/glum/actions/runs/8578861406 for details.

github-actions[bot]

How to set `max_alpha` in `GeneralizedLinearRegressorCV`?

3

The class does not have an argument `max_alpha`. The `min_alpha` description states > Minimum alpha to estimate the model with. The grid will then be created over `[max_alpha, min_alpha]` If...

mlondschien

glum with ray

3

Using glum and joblib with ray, I ran multiple models and found that threads could use 1 core, and if I set n_jobs=1, I could only use 50% of all...

cfkstat

I.e. a link to http://statmills.com/2023-11-20-Penalized_Splines_Using_glum/.

MatthiasSchmidtblaicherQC

Change base levels for categorical fields?

3

When modeling with `glum` using a dataset containing both categorical and numeric features, I want to manually set base levels for the categorical fields. This can be done in `statsmodels`...

enriquecardenas24

This is set [here](https://github.com/Quantco/glum/blob/2649d5b4abc8f9ae56a840975086f65689c8cad0/src/glum/_solvers.py#L424). Reducing this to a more reasonable number such as 1000 results in much faster convergence and fewer warnings ``` /cluster/customapps/biomed/grlab/users/lmalte/mambaforge/envs/icufm/lib/python3.10/site-packages/glum/_solvers.py:58: ConvergenceWarning: Coordinate descent did not converge....

mlondschien

Use dtype dependent precision

6

xref https://github.com/Quantco/glum/issues/843

mlondschien

Force dtype `np.float64` for optimization while data dtype is `np.float32`

1

We're having quite a few problems with optimization in `float32`. In small batches, these go away if we `.astype(np.float64)` our data before calling `glum.GeneralizedLinearRegressor.fit`. This also makes the algorithm much...

mlondschien

`glum.GeneralizedLinearRegressor(verbose=1, alpha_search=True).fit(X, y)` prints ``` Iteration 0: | | 0/? [s/it, gradient norm=7.508788257837296e-09] Iteration 1: 75%|████████████████████████████████████████████▎ | 1.5/2.0 [0.13s/it, gradient norm=0.0003172657161485404] Iteration 1: 51%|█████████████████████████████▌ | 1.02/2.0 [0.13s/it, gradient norm=0.0009597403695806861] Iteration...

mlondschien

Support for linux arm64? `<jemalloc>: Unsupported system page size` and `Segmentation fault`

9

I am using `glum` on an `arm64` machine. Both after installing from `conda` and `pip` (NB: why are there no arm64 linux wheels? Outputs below are from `conda` installation) and...

mlondschien