TAP: Accelerating Large-Scale DNN Training Through Tensor Automatic Parallelisation

Add code
Feb 01, 2023

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: