Add documentation

Signed-off-by: Ashwin Vaidya <[email protected]>

README.md

@@ -180,6 +180,34 @@ anomalib predict --model anomalib.models.Patchcore \
 
 > 📘 **Note:** For advanced inference options including Gradio and OpenVINO, check our [Inference Documentation](https://anomalib.readthedocs.io).
 
+# Training on Intel GPUs
+
+> [!Note]
+> Currently, only single GPU training is supported on Intel GPUs.
+> These commands were tested on Arc 750 and Arc 770.
+
+Ensure that you have PyTorch with XPU support installed. For more information, please refer to the [PyTorch XPU documentation](https://pytorch.org/docs/stable/notes/get_start_xpu.html)
+
+## 🔌 API
+
+```python
+from anomalib.data import MVTec
+from anomalib.engine import Engine, SingleXPUStrategy, XPUAccelerator
+from anomalib.models import Stfpm
+
+engine = Engine(
+    strategy=SingleXPUStrategy(),
+    accelerator=XPUAccelerator(),
+)
+engine.train(Stfpm(), datamodule=MVTec())
+```
+
+## ⌨️ CLI
+
+```bash
+anomalib train --model Padim --data MVTec --trainer.accelerator xpu --trainer.strategy xpu_single
+```
+
 # ⚙️ Hyperparameter Optimization
 
 Anomalib supports hyperparameter optimization (HPO) using [Weights & Biases](https://wandb.ai/) and [Comet.ml](https://www.comet.com/).

docs/source/markdown/guides/how_to/index.md

@@ -72,6 +72,13 @@ Learn more about anomalib's deployment capabilities
 Learn more about anomalib hpo, sweep and benchmarking pipelines
 :::
 
+:::{grid-item-card} {octicon}`cpu` Training on Intel GPUs
+:link: ./training_on_intel_gpus/index
+:link-type: doc
+
+Learn more about training on Intel GPUs
+:::
+
 ::::
 
 ```{toctree}
@@ -83,4 +90,5 @@ Learn more about anomalib hpo, sweep and benchmarking pipelines
 ./models/index
 ./pipelines/index
 ./visualization/index
+./training_on_intel_gpus/index
 ```

docs/source/markdown/guides/how_to/training_on_intel_gpus/index.md

@@ -0,0 +1,52 @@
+# Training on Intel GPUs
+
+This tutorial demonstrates how to train a model on Intel GPUs using anomalib.
+Anomalib comes with XPU accelerator and strategy for PyTorch Lightning. This allows you to train your models on Intel GPUs.
+
+> [!Note]
+> Currently, only single GPU training is supported on Intel GPUs.
+> These commands were tested on Arc 750 and Arc 770.
+
+## Installing Drivers
+
+First, check if you have the correct drivers installed. If you are on Ubuntu, you can refer to the [following guide](https://dgpu-docs.intel.com/driver/client/overview.html).
+
+Another recommended tool is `xpu-smi` which can be installed from the [releases](https://github.com/intel/xpumanager) page.
+
+If everything is installed correctly, you should be able to see your card using the following command:
+
+```bash
+xpu-smi discovery
+```
+
+## Installing PyTorch
+
+Then, ensure that you have PyTorch with XPU support installed. For more information, please refer to the [PyTorch XPU documentation](https://pytorch.org/docs/stable/notes/get_start_xpu.html)
+
+To ensure that your PyTorch installation supports XPU, you can run the following command:
+
+```bash
+python -c "import torch; print(torch.xpu.is_available())"
+```
+
+If the command returns `True`, then your PyTorch installation supports XPU.
+
+## 🔌 API
+
+```python
+from anomalib.data import MVTec
+from anomalib.engine import Engine, SingleXPUStrategy, XPUAccelerator
+from anomalib.models import Stfpm
+
+engine = Engine(
+    strategy=SingleXPUStrategy(),
+    accelerator=XPUAccelerator(),
+)
+engine.train(Stfpm(), datamodule=MVTec())
+```
+
+## ⌨️ CLI
+
+```bash
+anomalib train --model Padim --data MVTec --trainer.accelerator xpu --trainer.strategy xpu_single
+```