opt_einsum_fx icon indicating copy to clipboard operation
opt_einsum_fx copied to clipboard
verified publisher icon Linux-cpp-lisp

[Feature] Accumulate scalar constants in better-than-`double` precision

Open Linux-cpp-lisp opened this issue 4 years ago • 0 comments

When accumulating scalar constants at graph optimization time, arbitrary precision arithmetic should be used to ensure the best result.

Linux-cpp-lisp avatar Mar 15 '21 02:03 Linux-cpp-lisp