commit | 7cea9a0f2055f812f590716813104a16dd56a367 | [log] [tgz] |
---|---|---|
author | Haibo Huang <hhb@google.com> | Thu May 14 04:07:46 2020 -0700 |
committer | Haibo Huang <hhb@google.com> | Fri May 15 00:20:18 2020 +0000 |
tree | 30b48e320586c9d4e5beec06e7d903dc3fe55d94 | |
parent | 2ea5ac4e63f6f5c84d75b8da4ec3d8ad332bc26b [diff] | |
parent | 4dfe081cf6bcd15db339cf2680b9281b8451eeb3 [diff] |
Upgrade FP16 to 4dfe081cf6bcd15db339cf2680b9281b8451eeb3 Exempt-From-Owner-Approval: upgrade Change-Id: I2f0b6248a85ce7ac92982ad91f619e84c941f36c
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.