Cannot identify xpu on Intel Xe GPUs

[tnnandi]

Bug description

The following code does not detect xpu when ran on an Intel Xe GPU and throws an error ValueError: You selected an invalid accelerator name: accelerator='xpu'. Available names are: auto, cpu, cuda, mps, tpu. Could you please tell if lightning is currently compatible with xpu (based on my understanding there was some work on xpu compatibility here #19443)?

If accelerator is set to "auto", it uses the cpu.

Could you please suggest what needs to be fixed in the code below to utilize the GPUs?

What version are you seeing the problem on?

v2.5

Reproduced in studio

No response

How to reproduce the bug

import torch
import torch.nn.functional as F
from torch.utils.data import DataLoader, TensorDataset
from lightning import LightningModule, Trainer
import intel_extension_for_pytorch as ipex

class SimpleNet(LightningModule):
    def __init__(self):
        super().__init__()
        self.l1 = torch.nn.Linear(28 * 28, 10)

    def forward(self, x):
        return torch.relu(self.l1(x.view(x.size(0), -1)))

    def training_step(self, batch, batch_idx):
        x, y = batch
        logits = self(x)
        loss = F.cross_entropy(logits, y)
        self.log("train_loss", loss)
        return loss

    def configure_optimizers(self):
        return torch.optim.Adam(self.parameters(), lr=0.02)

def get_dataloader():
    # 1000 samples of 28x28 images, 10 classes
    X = torch.randn(1000, 1, 28, 28)
    y = torch.randint(0, 10, (1000,))
    dataset = TensorDataset(X, y)
    return DataLoader(dataset, batch_size=64, shuffle=True)

train_loader = get_dataloader()

model = SimpleNet().to("xpu")

optimizer = torch.optim.Adam(model.parameters(), lr=0.02)
model, optimizer = ipex.optimize(model, optimizer=optimizer, dtype=torch.float32)

trainer = Trainer(
    accelerator="xpu",  # Lightning should detect "xpu" but throws an error "ValueError: You selected an invalid accelerator name: `accelerator='xpu'`. Available names are: auto, cpu, cuda, mps, tpu."
    devices=1,
    max_epochs=3000,
    log_every_n_steps=10,
)


trainer.fit(model, train_dataloaders=train_loader)

Error messages and logs

ValueError: You selected an invalid accelerator name: `accelerator='xpu'`. Available names are: auto, cpu, cuda, mps, tpu.

Environment

Current environment

#- PyTorch Lightning Version (e.g., 2.5.0):
#- PyTorch Version (e.g., 2.5):
#- Python version (e.g., 3.12):
#- OS (e.g., Linux):
#- CUDA/cuDNN version:
#- GPU models and configuration:
#- How you installed Lightning(`conda`, `pip`, source):

More info

No response