Skip to content

[Issue #10584] fix: Use torch.round().long() for timestep comparison line 303 to handle floating-point precision#10587

Open
Liang-ZX wants to merge 1 commit intohuggingface:mainfrom Liang-ZX:fix-timestep-precision
+2-1

Commits