David Hou aebaab011f faster wino compile by catting consts across data expand dim (#3293) ...

* PoC faster wino compile by catting consts across data expand dim

* fix fusions

* faster + golf it

* noqa 501

* implicit broadcast

* Revert "implicit broadcast"

This reverts commit 5915a9083d045ec1e6be84dcb492333325d48666.

* shorter

* shorter

* oops

* 216 upcasts is probably fine

* wino kernel count test

* test winograd number of sts

* specify device for apply_matrix mat elements

2024-02-02 03:47:45 -05:00

27 KiB

Raw Blame History

View Raw