commit | 3bd198b649a0cb30bd01f2daf5da7a4d00cfffb2 | [log] [tgz] |
---|---|---|
author | qiaoli <[email protected]> | Thu Jul 27 23:20:58 2023 +0000 |
committer | qiaoli <[email protected]> | Thu Jul 27 23:21:03 2023 +0000 |
tree | 5861de0ba4651a75bafe7097b5665fed86eb6903 | |
parent | 7cbdbedccabdce9d78dd1e97a6481d15afc450ef [diff] |
Allow ondevicepersonalization module to use fp16_headers Bug: 241799297 Test: mma Change-Id: Ie2f1b6ab5f20312f84266d98e9cc7f6245469efd
Header-only library for conversion to/from half-precision floating point formats
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.