Open Vaibhavs10 opened 7 months ago
I am starting this issue to do a more thorough benchmarking than the notebooks used in the repo.
What should we measure:
Hardware (this would give the best of both worlds IMO):
Tricks that we should measure:
scaled_dot_product_attention
Models that we should test:
Has this been finalized yet just out of curiosity?
I am starting this issue to do a more thorough benchmarking than the notebooks used in the repo.
What should we measure:
Hardware (this would give the best of both worlds IMO):
Tricks that we should measure:
scaled_dot_product_attention
via BetterTransformers API in Optimum.Models that we should test: