GradientCheckpointingHorovod. We use gradient checkpointing technique and Horovod to jointly accelerate training large neural networks such as Bert, Roberta, and Xlnet.

github.com/jhsa26/GradientCheckpointingHorovod

Vaya's read on this project

Problem, audience, market, and the verdict — sign in to see it.

Updates

No recent activity.