#include "tensorstore/util/bfloat16.h"
BFloat16 tensorstore::expm1(BFloat16 x);

Computes \(e\) raised to the given power, minus 1 (\(e^x-1\)).