Refactor export_llama_lib (#2030)
Summary: Pull Request resolved: https://github.com/pytorch/executorch/pull/2030 Separate out "receipe" code and actual "cooking" code. Introduces a new `LlamaBuilder` class that handles internal logic of exporting. Takes source transforms, quantizers, and partitioners. Reviewed By: mergennachin Differential Revision: D54027081 fbshipit-source-id: 8da52b7a538331389eb148bff1b143373fcefb18
M
Mengwei Liu committed
c4a1e95c691839e0e887bb0734dcaa004f794770
Parent: ca6995b
Committed by Facebook GitHub Bot <facebook-github-bot@users.noreply.github.com>
on 2/23/2024, 3:59:42 AM