automl / Auto-PyTorch

Automatic architecture search and hyperparameter optimization for PyTorch
Apache License 2.0
2.38k stars 289 forks source link

Cannot run time-series example on GPU #483

Open LuciusMos opened 2 years ago

LuciusMos commented 2 years ago

NOTE: ISSUES ARE NOT FOR CODE HELP - Ask for Help at https://stackoverflow.com

Your issue may already be reported! Also, please search on the issue tracker before creating one.

Issue Description

Expected Behavior

run on GPU

Current Behavior

still run on CPU

Your Code

Auto-PyTorch/autoPyTorch/configs/default_pipeline_options.json
{
    "device": "cuda",
    .....
}

Error message

No error message

Your Local environment

dengdifan commented 2 years ago

Hi,

did you try other examples with cuda, did they have the same issue? You could also try to add api.set_pipeline_config(device="cuda") after you have initialized your TimeSeriesForecastingTask to check if the network still runs on cpus.