Joel Schlosser ecc5e05854 Refactor NJT min / max seqlen handling for convenience (#138130) ... There's an annoying pattern emerging for pulling out the NJT min / max seqlen ints if they exist without computing / caching if they don't. This PR introduces private convenience functions to simplify handling this and avoiding redundant checks. Pull Request resolved: https://github.com/pytorch/pytorch/pull/138130 Approved by: https://github.com/soulitzer		2024-10-17 17:28:39 +00:00
..
_internal	Refactor NJT min / max seqlen handling for convenience (#138130)	2024-10-17 17:28:39 +00:00
__init__.py	Implement 2D version of masked_select for nestedtensors (#133889)	2024-08-20 21:46:32 +00:00