search

gussmith23 / glenside

A pure, low-level tensor program representation enabling tensor program optimization via program rewriting. See the web demo at https://gussmith23.github.io/glenside-web-demo/
68 stars 10 forks source link

Enable more linear layer matching on FlexASR #144

Closed gussmith23 closed 2 years ago

gussmith23 commented 2 years ago

We do this by rewriting matmuls to matmuls plus bias adds (with 0 biases).