Rate this Page
★★★★★

torch.sym_max#

torch.sym_max(a,b)[source]#

SymInt-aware utility for max which avoids branching on a < b.Unlike builtins.max(), this only works for int/float, and it alwayspromotes to float if any argument is float (unlike builtins.max, whichwill faithfully preserve the type of the input argument).