pub_soft/dgl

mirror of https://github.com/dmlc/dgl.git synced 2026-06-03 19:34:33 +08:00

Files

History

Muhammed Fatih BALIN ea940f72b5 [GraphBolt][CUDA] Destroy dist group at the end of script. (#7736 )

2024-08-23 15:43:12 -04:00

..

[GraphBolt][CUDA] Destroy dist group at the end of script. (#7736 )

2024-08-23 15:43:12 -04:00

node_classification_sage.py

[Misc] DGL multi-gpu example run error work around (#6999 )

2024-01-23 15:23:09 +08:00

README.md

[Benchmark] update node classification for multigpu benchmark (#5674 )

2023-05-12 20:15:55 +08:00

README.md

Multiple GPU Training

Requirements

pip install torchmetrics==0.11.4

How to run

Node classification

Run with following (available dataset: "ogbn-products", "ogbn-arxiv")

python3 node_classification_sage.py --dataset_name ogbn-products

Results with default arguments

* Test Accuracy of "ogbn-products": ~0.7716
* Test Accuracy of "ogbn-arxiv": ~0.6994