search

time-series-foundation-models / lag-llama

Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting
Apache License 2.0
1.09k stars 121 forks source link

Update requirements.txt #2

Closed ShaneOss closed 3 months ago

ShaneOss commented 4 months ago

bump gluonts version. Previous version use old version of pydantic.

kashif commented 4 months ago

@ShaneOss thanks mate! I was thinking of upgrading to lighting instead of pytorch-lightning... what do you think?

ShaneOss commented 4 months ago

Hey @kashif , good idea. It works, no issues. Also fixes a few legacy dependency problems.

kashif commented 4 months ago

no worries, this week is choka for me but will have a look soon

ashok-arjun commented 4 months ago

Thanks @ShaneOss. I'd have to check if this causes any problems when training, since we found our version of GluonTS to be the most stable one. Will check it soon.

Regarding lightning, @kashif, that'll require a good amount of refactoring if you want to upgrade: are there any benefits?

kashif commented 3 months ago

@ShaneOss so we realized that if we require gluonts[torch] as the top level requirement, it pulls in the working versions of lighting and pytorch-lightning so closing this PR for now