# Learn PyTorch

> [!Important]
> Steps here to run PyTorch locally are specific to machines with Intel XPUs.

Experimenting with PyTorch using Intel architecture (i.e., Intel Core Ultra processor with iGPU).

After installing `ipex-llm` which is required to use Intel GPUs (see [Setup](#setup)), you will have access to `conda` and be able to `import torch` normally.

The XPU is an accelerator for working with Tensors.

## Setup

1. [Install IPEX-LLM on Intel GPU with PyTorch 2.6](https://git.ayo.run/ayo/ipex-llm/src/branch/main/docs/mddocs/Quickstart/install_pytorch26_gpu.md)

2. Clone repo

```bash
$ git clone https://git.ayo.run/ayo/learn-pytorch
```

3. Run `env.sh` to activate the conda environment and set

```bash
$ cd learn-python
$ . env.sh
```

4. (Optional) Confirm if XPU is detected

```bash
$ python # go intou the python shell

$ import torch
$ torch.xpu.is_available()
$ torch.xpu.get_device_name()

```

## Links

- [Get started with PyTorch locally](https://pytorch.org/get-started/locally/)
- [Getting started with Tensors](https://docs.pytorch.org/tutorials/beginner/basics/tensorqs_tutorial.html)
- [torch operations](https://docs.pytorch.org/docs/stable/torch.html)