Olivier Sprangers comments

Results 191 comments of


                                            Olivier Sprangers

Auto models get ray.exceptions.ActorDiedError when run on multi-GPU node

Thanks for reporting, great that this works now. The issue with Ray still remains, but it is then likely a harder issue for us to solve. I'd have to read...

Feature: Coherence Diagnostics Report (post-reconcile())

Thanks - can you provide an example of the report, including code to produce it (e.g. based on reconciliation results obtained from our reconcilers)?

Feature: Coherence Diagnostics Report (post-reconcile())

I've made a diagnostics report in #424. @ankitlade12 Let me know if that helps you.

Strictly hierarchical assertions preceding fit method

First is in place. Second we could add a class attribute `IS_STRICTLY_HIERARCHICAL` that can be checked in reconcile, but it's a minor thing imho (in time).

[FEAT] Improved CI test

Could include this in the CI, CI is now fast enough to just run it on the fly

Change reconcile input name from S to S_df

Agree - with the note that there will be a separate S for temporal aggregation, so that needs to be included in the naming too.

TimeGPT Nano?

Just wanted to state that we take note of the feature request :)

Add tweedie on loss file

Thanks for your contribution! Two things quickly come to mind (reviewing on my phone) 1. Run utils.ipynb, but don't clear it. It's used for documentation in this repo. 2. The...

Add tweedie on loss file

Thanks, great work @janrth! Couple of things: - The Polars loss doesn't return the same value for power=2 when y=0. Pandas will give an inf, Polars doesn't. - In general...

Add tweedie on loss file

> @elephaint very true with the apply, I should have realised this! The issue with power=2 I didn't see at all. Thanks for taking care of it. I refactored the...