pytorch get gpu number

torch.cuda.device_count()

Here is what the above code is Doing:
1. We’re checking if the device is cuda-capable. If it is, we’re checking how many GPUs are available.
2. If there are GPUs available, we’re setting up a list of IDs for each GPU.
3. We’re setting up a Boolean variable to track whether PyTorch is using GPUs.
4. We’re checking whether PyTorch is using GPUs.
5. If PyTorch is using GPUs, we’re printing out the number of GPUs being used.