Alexander März

Kindly asking for an update on this. Are you planning to release the code any time soon?

Hi @Cattes, thanks again for your effort, really appreciate your PR. I am planning to release a revised version of both XGBoostLSS and LightGBMLSS soonish, that is more tailored to...

Congratulations on the [paper ](https://proceedings.neurips.cc/paper/2020/hash/c6b8c8d762da15fa8dbbdfb6baf9e260-Abstract.html)! Kindly asking to make your code publicly available. Many Thanks!

@hihihihiwsf Many Thanks for sharing the code, very much appreciated! In your [paper](https://proceedings.neurips.cc/paper/2020/hash/c6b8c8d762da15fa8dbbdfb6baf9e260-Abstract.html), you also mention that you have a DeepAR extension equipped with adversarial training. Would it be possible...

@hihihihiwsf I would greatly appreciate if you would be so kind to share the Adverserial DeepAR code, since this would facilitate reproducibility of your aproach. Would love to test the...

IBM has released its version of the [PatchTSMixer on HuggingFace](https://github.com/huggingface/transformers/blob/v4.36.1/src/transformers/models/patchtsmixer/configuration_patchtsmixer.py). Maybe this helps to have it available in darts soon

Very interesting discussions. Even though there is no quantile regression available yet, you might find my repo [LightGBMLSS](https://github.com/StatMixedML/LightGBMLSS) that models and predicts the full conditional distribution of a univariate target...

@mat-ej Sorry for the late reply. Yes I have a working version of non-crossing quantiles, where all quantiles are jointly estimated in one model. I haven't included it yet as...

@RektPunk The approach of [Cannon (2018)](https://link.springer.com/article/10.1007/s00477-018-1573-6) is extremely nice and using the monotonicity constraint on the quantile-column "guarantees" they are non-crossing. Extremely happy you brought this to our attention! I...

I used the above implementation and compared it with the base quantile regression available in LightGBM. Both models have the same set of hyper-parameters and number of iterations, and the...