frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

[ez] fix typo (#165282)

Browse Source 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165282
Approved by: https://github.com/ezyang, https://github.com/mlazos
This commit is contained in:
Bob Ren
2025-10-12 20:49:17 -07:00
committed by PyTorch MergeBot
parent 8e510e1095
commit 0c14f55de6
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
torch/fx/passes/runtime_assert.py
View File

@ -61,7 +61,7 @@ def insert_deferred_runtime_asserts(
"""
During tracing, we may have discovered that some data-dependent values
had runtime assert on them; e.g., torch.empty(x.item()) induces a runtime
that x.item() >= 0. This asserts can happen unpredictably during fake
that x.item() >= 0. These asserts can happen unpredictably during fake
tensor propagation, so we cannot conveniently insert them into the FX graph
when they occur. Instead, we accumulate them in the ShapeEnv, and in this
pass insert them into the graph as proper tests.