Skip to content

Commit 247a9a5

Browse files
EddyLXJmeta-codesync[bot]
authored andcommitted
Fix changed name (#3549)
Summary: X-link: pytorch/FBGEMM#5138 Pull Request resolved: #3549 X-link: https://github.com/facebookresearch/FBGEMM/pull/2140 is_st_publish is changed to load_ckpt_without_opt in diff D85830053 Reviewed By: emlin Differential Revision: D87135958 fbshipit-source-id: a693e766fe2c6cea58ad2ca1b4c6b80f54373564
1 parent e312596 commit 247a9a5

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

torchrec/distributed/batched_embedding_kernel.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -480,7 +480,7 @@ def _populate_zero_collision_tbe_params(
480480

481481
optimizer_type_for_st: Optional[str] = None
482482
optimizer_state_dtypes_for_st: Optional[FrozenSet[Tuple[str, int]]] = None
483-
if kvzch_tbe_config and kvzch_tbe_config.is_st_publish:
483+
if kvzch_tbe_config and kvzch_tbe_config.load_ckpt_without_opt:
484484
optimizer_type_for_st = kvzch_tbe_config.optimizer_type_for_st
485485
optimizer_state_dtypes_for_st = kvzch_tbe_config.optimizer_state_dtypes_for_st
486486

0 commit comments

Comments
 (0)