OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity

[ez] fix grammar mistakes in StatefulSymbolicContext comment (#152598)

Browse Source 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/152598
Approved by: https://github.com/malfet
ghstack dependencies: #151407
This commit is contained in:
bobrenjc93 2025-05-01 07:02:07 -07:00 committed by PyTorch MergeBot
parent 36e5ff6bc4
commit 1f898657e6
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
torch/fx/experimental/symbolic_shapes.py
View File

@ -1996,12 +1996,12 @@ class StatefulSymbolicContext(StatelessSymbolicContext):
other values - dynamic_sizes and constraint_sizes will not be read if we cache
hit.
It is the cache owners responsibility to maintain the lifecycle of the cache
w/r/t different shape_envs, clearing, etc.
It is the cache owner's responsibility to maintain the lifecycle of the cache
with respect to different shape_envs, clearing, etc.
"""
tensor_source: Source = None # type: ignore[assignment]
# Why is this keyd on int first?
# Why is this keyed on int first?
# That integer is actually the id of the shape_env. This cache short-circuits symbol
# creation, and we must store it per shape env. Now, while tracing invariants are a single
# shape env per tracing context, and every new frame gets a new shape_env. So where would we have