Metadata-Version: 2.1
Name: dgl-ascend
Version: 0.0.2
Summary: Deep Graph Library with Ascend NPU support
Home-page: https://gitcode.com/AI4Science/dgl-ascend
Author: dgl-ascend-team
Author-email: lihepeng1@huawei.com
Maintainer: dgl-ascend Team
License: APACHE
Classifier: Development Status :: 3 - Alpha
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: Apache Software License
Description-Content-Type: text/markdown
Requires-Dist: torch (==2.8.0)
Requires-Dist: torch-npu (==2.8.0.post2)
Requires-Dist: networkx (>=2.1)
Requires-Dist: numpy (>=1.14.0)
Requires-Dist: packaging
Requires-Dist: pandas
Requires-Dist: psutil (>=5.8.0)
Requires-Dist: pydantic (>=2.0)
Requires-Dist: pyyaml
Requires-Dist: requests (>=2.19.0)
Requires-Dist: scipy (>=1.1.0)
Requires-Dist: tqdm

# dgl-ascend

Deep Graph Library (DGL) with Ascend NPU support, developed by the BUPT-GAMMA group.

## Installation

```bash
pip install dgl-ascend
```

Requirements: Ascend 910B NPU, CANN >= 8.5, aarch64 Linux, Python 3.11.

## Quick Start

```python
import torch
import torch_npu
import dgl

device = torch.device("npu:0")
g = dgl.graph(([0, 1, 2], [1, 2, 3])).to(device)
g.update_all(dgl.function.copy_u("x", "m"), dgl.function.sum("m", "h"))
```

## License

Apache License 2.0.
