Notes about floating point arithmetic.
Currently interested in applications of low-precision arthimetic to scientific computing. Looking for a floating point implementation to simulate low precision plus a math library with elementary functions.
- GMP (GNU Multiple Precision Arithmetic Library) https://gmplib.org/
- MPFR https://www.mpfr.org/
- QPyTorch https://github.com/Tiiiger/QPyTorch
- FlexFloat (C library) https://github.com/oprecomp/flexfloat or FloatX (C++ headers) https://github.com/oprecomp/FloatX