torch.backends.mps.is_available() is the new torch.cuda.is_available()

[geg00]

Feature requests should first be proposed on the forum.

Link to forum discussion.

Please place URL here.
https://forums.fast.ai/t/fastai-on-apple-m1/86059/28?u=gerardo

Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
When you use learner there is no way to select the device to be used

learn = vision_learner(dls, resnet18, metrics=error_rate)

Describe the solution you'd like
A clear and concise description of what you want to happen.
It just goes to use CPU not the MPS

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.
Availability to select a device

Additional context
Add any other context or code examples about the feature request here.
https://forums.fast.ai/t/how-to-specify-gpu-like-pytorch-device-cuda-2-in-fastai/54379/2?u=gerardo

THIS IS FOR FASTAI ON APPLE M1 MAC

[jph00]

mac is not supported for fastai

[geg00]

Well that's disappointing