search

INK-USC / RE-Net

Recurrent Event Network: Autoregressive Structure Inference over Temporal Knowledge Graphs (EMNLP 2020)
http://inklab.usc.edu/renet/
436 stars 95 forks source link

Running time is too long #52

Closed zjwu0522 closed 2 years ago

zjwu0522 commented 3 years ago

I tried to run RE-Net on a Tesla V100, but it takes over 10 hours for one epoch.

Besides, GPU utilization has been very low, I guess it is the problem of dataloader.

I wanna know if anyone has successfully and efficiently implemented this algorithm?

woojeongjin commented 2 years ago

Hi! Which dataset/version of DGL did you use?

MrLiuCC commented 2 years ago

Hi! Which dataset/version of DGL did you use?

I meet the same problem. When i run train.py,it is too slow. I wanna know hou to run ecciciently.