Paddle

Author	SHA1	Message	Date
Zhaolong Xing	8acd745c25	[Ernie GPU Optim]: Fuse three fc to multihtead matmul (#22486 ) * 1. optim multihead matmul: fuse three fc to multihtead matmul test=develop * fix conflict test=develop * fix comments test=develop	6 years ago
zhaoyuchen2018	8859ddd6cf	Refine multihead kernel, align block to 32 (#21961 ) * Refine multihead kernel, align block to 32 test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com> * Refine log comments test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com>	6 years ago
zhaoyuchen2018	a5a8d14414	Fix softmax cuda bug (#21720 ) * Fix softmax cuda bug * Refine multihead log and softmax logic	6 years ago
Yiqun Liu	03ba0fdae6	Move the codes of fused operators to operators/fused directory. (#20881 ) * Move the codes of fused operators to operators/fused directory. test=develop * Correct the op name in cmake. * Change the use of PADDLE_ENFORCE. test=develop	6 years ago