disable cuda warning

[the1gofer]

I'm running the en_core_web_trf on a Mac M1, and get the following:
.../venv/lib/python3.9/site-packages/torch/autocast_mode.py:162: UserWarning: User provided device_type of 'cuda', but CUDA is not available. Disablingt] warnings.warn('User provided device_type of \'cuda\', but CUDA is not available. Disabling')

I've seen some poeple suggest i use the lg model, but with some testing the trf seems far more accurate for my use. I don't really care if it's slow, but it seems to work other than the warning. Can I disable it? I have tried the following, and it doens't seem to work.

with warnings.catch_warnings():
    warnings.simplefilter("ignore")
    self.doc = self.nlp(obit_text)

[the1gofer]

Any thoughts?

[ljvmiranda921]

If you just wish to disable warnings, you can try:

import warnings
warnings.filterwarnings("ignore", category=UserWarning)

Although it's possible that the warning happens during import, not execution, that's why it still showed up in your initial approach.

[the1gofer]

Thanks, I'll see if that works.

[sushant1253]

This works, Thanks.

[the1gofer]

That seems to have done the trick. Thank you. Jason Crews

…

On Mon, Apr 4, 2022 at 5:18 PM Lj Miranda ***@***.***> wrote: If you just wish to disable warnings, you can try: import warnings warnings.filterwarnings("ignore", category=UserWarning) Although it's possible that the warning happens during import, not execution, that's why it still showed up in your initial approach. — Reply to this email directly, view it on GitHub <#10587 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAQKUNHJTJI3NHZ75C7OFL3VDOBGHANCNFSM5SCKHCQA> . You are receiving this because you authored the thread.Message ID: ***@***.***>