ClubberLang

Add PyStan 3 support

3

PyStan 3 benefits from an ISC license as opposed to the “viral” GPL license of PyStan 2. That alone is worth the upgrade.

Early Stopping in EBM

5

Correct me if I'm wrong but the native early stopping mechanism within EBM will just take a random slice of the data. In the case of (i) grouped observations (panel...

Verbose option during training

5

Training time can be quite long, particular with inner bagging is used. It would be helpful to have a verbose option to understand where the EBM model is while in...

Neural ODE based synthesis.

1

Interesting architecture for consideration for inclusion in this package: https://arxiv.org/pdf/2105.14969.pdf

Tracking metrics while training

4

Generator/Discriminator loss is not particularly interpretable. If one prints out these metrics via the verbose option, it's difficult to assess whether the GAN is improving or not. I've run scenarios...

Docker image vulnerabilties

2

My organisation has allowed us a test environment to try out various Azure tools. The container built as part of this example is shown as having over a dozen vulnerabilties...

LightGBM categorical variables

9

Is there any indication as to when this might be supported as a fit parameter for LGBMClassifier and LGBMRegressor?

Diffusion model-based synthetic data generation

2

Interesting use of diffusion models to generate synthetic data: https://github.com/rotot0/tab-ddpm/blob/main/ https://arxiv.org/abs/2209.15421

Best way to integrate into .NET

1

I'm looking for the best way to integrate a compiled model into .NET code. I'd rather not have to utilise an initial call to Python and rely on the underlying...

Python version

1

I read the excellent Arxiv paper. Any chance of a python version coming?