floating-point half-precision fp16

hpc persistence mpi stl blitz dataset hdf5 armadillo cpp17 dlib blaze serialization-library half-precision itpp hdf5-library eigen3 hdf5-datasets ublas h5cpp

javascript half-precision fp16 binary16 float16array float16

pytorch half-precision re-id distributed-training model-parallel

gpu cublas precision gemm half-precision float16 p100 v100

android caffe opencl half-precision fp16

matlab matrix arithmetic ieee rounding half-precision fp16 binary16 bfloat16 arithmetic-formats precisions subnormal-numbers

cuda clang half-precision

macos swift ios tvos watchos cocoapods carthage swiftpm half-precision

go golang floating-point ieee754 half-precision binary16 float16

c z80 z80asm zx-spectrum half-precision float16

z80 zx-spectrum floating-point half-precision binary16 half bfloat16 bfloat

math graphics vector matrix game-development dimensions angle quaternion math-library half-precision templated gmtk math-toolkit

ieee754 half-precision binary16

half-precision mips-instructions

c assembly doge nasm half-precision

constexpr exponent floating-point 8bit float sfinae half-precision 16bit half mantissa 32bit 64bit double-precision significand 48bit