History

hcl 7be3fd486c init from android-12.1.0_r8		2023-01-09 17:11:35 +08:00
..
bench	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
cmake	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
include	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
test	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
.gitignore	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
Android.bp	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
CMakeLists.txt	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
LICENSE	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
METADATA	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
MODULE_LICENSE_MIT	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
OWNERS	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
README.md	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
TEST_MAPPING	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
configure.py	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00
confu.yaml	init from android-12.1.0_r8	2023-01-09 17:11:35 +08:00

README.md

FP16

Header-only library for conversion to/from half-precision floating point formats

Features

Supports IEEE and ARM alternative half-precision floating-point format
- Property converts infinities and NaNs
- Properly converts denormal numbers, even on systems without denormal support
Header-only library, no installation or build required
Compatible with C99 and C++11
Fully covered with unit tests and microbenchmarks

Acknowledgements

The library is developed by Marat Dukhan of Georgia Tech. FP16 is a research project at Richard Vuduc's HPC Garage lab in the Georgia Institute of Technology, College of Computing, School of Computational Science and Engineering.

This material is based upon work supported by the U.S. National Science Foundation (NSF) Award Number 1339745. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect those of NSF.