5 Commits

Author	SHA1	Message	Date
Orion Reblitz-Richardson	9ec0a2aef4	fbshipit-source-id: ba600fcd2b5cefc7621357bdeb05e24cea02e5af	2018-06-27 04:50:56 -07:00
Orion Reblitz-Richardson	1d5780d42c	Remove Apache headers from source. ... * LICENSE file contains details, so removing from individual source files.	2018-03-27 13:10:18 -07:00
Yangqing Jia	8286ce1e3a	Re-license to Apache ... Summary: Closes https://github.com/caffe2/caffe2/pull/1260 Differential Revision: D5906739 Pulled By: Yangqing fbshipit-source-id: e482ba9ba60b5337d9165f28f7ec68d4518a0902	2017-09-28 16:22:00 -07:00
Junjie Bai	4fddc04054	Use the same schema of switching to device reduce sum for SumSqrElements ... Summary: Based on benchmark script located at `caffe2/experiments/python/device_reduce_sum_bench.py`, device reduce sum is slower for N <= 10000, so we only switch to use device reduce for large N in SumElements. This diff applies the same schema for SumSqrElements. Reviewed By: jamesr66a Differential Revision: D5369868 fbshipit-source-id: ae13a611aff9d3464d1c4950ee155c740a2da339	2017-07-05 10:52:17 -07:00
Junjie Bai	f3a59aedff	Use cub::DeviceReduce for faster math::Sum CUDA version ... Summary: Port SumElements and softmax_ops.cu to use device reduce sum Reviewed By: akyrola Differential Revision: D5351881 fbshipit-source-id: ca9604186c261ffcb1480da2a17baab8a4809372	2017-06-30 15:04:06 -07:00