DeepHypergraph
DeepHypergraph copied to clipboard
A pytorch library for graph and hypergraph computation.
中文文档有误
https://github.com/iMoonLab/DeepHypergraph/blob/main/docs/source/zh/start/low_order/bipartite_graph.rst#%E5%AE%9A%E4%B9%89 这个下面的介绍应该是“二分图”,写成了“有向图”。内容是正确的。
**Describe the bug** dhg\structure\hypergraphs\hypergraph.py 的 add_hyperedges() 函数 ```python // 判断e_weight 类型 if e_weight is None: e_weight = [1.0] * len(e_list) elif type(e_weight) in (int, float): e_weight = [e_weight] // 这里错误...
Collecting dhg Using cached dhg-0.9.3-py3-none-any.whl.metadata (25 kB) Using cached dhg-0.9.2-py3-none-any.whl.metadata (25 kB) Collecting sklearn (from dhg) Using cached sklearn-0.0.post12.tar.gz (2.6 kB) Preparing metadata (setup.py) ... error error: subprocess-exited-with-error × python...
No support for DHG.
Please add support for nested hypergraph for which a hyperedge can contain nested hyperedges, thus supporting multi-level hyperedge-based modeling. Thanks.
Could you please provide guidance on how to utilize batch size within the dhg.Hypergraph module?
When I try to run basic examples, I cannot successfully download datasets (e.g. cora, citeseer). It will report the following message: requests.exceptions.ConnectTimeout: HTTPSConnectionPool(host='download.moon-lab.tech', port=28501): Max retries exceeded with url: /datasets/cora/features.pkl...
The DHG requires torch version < 2 but I can't install torch version < 2.0 now