eschibli
eschibli
This appears to be fixed in v1.1 for `max_model_size>=23555082` (the size of the hardcoded Resnet model?). Anything smaller and a `FailedTrialError` is raised and not handled.
Be cautious about running outside of 8-bit mode, I saw much worse results using the cpu-only version of bitsandbytes
I think it would be most helpful to maintain a leaderboard over a few standard datasets broken out by model (perhaps with/without additional feature engineering?) That could highlight how performance...
> Hi @eschibli, > ... Thanks @madtoinou. I was not able to get Gradle running on my machine and didn't realize ruff was that easy to set up so sorry...
Understood Dennis
Hi @madtoinou: Sorry I was very busy in December and January. I should have some time this week or next week to finish this. Also, in my own work I...
@madtoinou are you happy with these changes?
@madtoinou @dennisbader sorry to nag but could you please clarify if you are unhappy with this addition or if more changes are needed to approve this?
Hi @madtoinou, I would simply state "The (or a sequence of) target time series used to successively train and compute the historical forecasts. **(Regardless of whether a model is global,...
Hi @madtoinou - in my work I have seen significantly better performance in problems with both past and future covariates using an encoder-decoder model, projecting in the middle of the...