DAMO-DI-ML / NeurIPS2023-One-Fits-All

The official code for "One Fits All: Power General Time Series Analysis by Pretrained LM (NeurIPS 2023 Spotlight)"
458 stars 64 forks source link

Multiple GPU using. #33

Open PiXianDouBanJiang opened 6 months ago

PiXianDouBanJiang commented 6 months ago

How can I use multiple gpus when doing Classification task?

PSacfc commented 6 months ago

You can modify the code with data DistributedDataParallel and DataParallel of torch, or refer to https://github.com/gzerveas/mvts_transformer