dmlc/dgl

The ALG2 for cusparsespmm under performed when feature size is small

Open

#4090 aperta il 7 giu 2022

Vedi su GitHub
 (2 commenti) (0 reazioni) (0 assegnatari)Python (2928 fork)batch import
help wanted

Metriche repository

Star
 (12.665 star)
Metriche merge PR
 (Nessuna PR mergiata in 30 g)

Descrizione

🐛 Bug

@yzh119 report the ALG2 seems performed badly for cusparsespmm when the feature size is small.

To Reproduce

Steps to reproduce the behavior:

Expected behavior

Environment

  • DGL Version (e.g., 1.0):
  • Backend Library & Version (e.g., PyTorch 0.4.1, MXNet/Gluon 1.3):
  • OS (e.g., Linux):
  • How you installed DGL (conda, pip, source):
  • Build command you used (if compiling from source):
  • Python version:
  • CUDA/cuDNN version (if applicable):
  • GPU models and configuration (e.g. V100):
  • Any other relevant information:

Additional context

Guida contributor