11 Commits (621a4085b9aff772ffc9c5018bcd8b55db609006)

Author	SHA1	Message	Date
Zhou Wei	7817003795	Optimize the error messages of paddle CUDA API (#23816) ... * Optimize the error messages of paddle CUDA API, test=develop * fix the error messages of paddle CUDA API, test=develop * Refactoring PADDLE_ENFORCE_CUDA_SUCCESS, and apply to curand/cudnn/cublas/NCCL,test=develop * remove build_ex_string,test=develop * merge conflict,test=develop	5 years ago
FlyingQianMM	443a713c9e	add backward gradient computation for op argsort (#22203) ... * add backward gradient computation for op argsort test=developo * use pre-commit test=develop	6 years ago
Aurelius84	5cb2c74127	add register op_data_type of pad/expand_as et.al (#21718) ... * add register op_data_type test=develop * fix register bug in isfinite op test=develop * rm int int64_t in pad2d gradKernel test=develop	6 years ago
zhaoyuchen2018	b16274556a	Add dscending for argsort (#21400) ... * Add ascending for argsort * Refine api doc description. * Refine descending description * Add int32 logic to speedup when data is small size. * Remove int32 opt as not support in python	6 years ago
zhaoyuchen2018	08c19c585d	Improve argsort performance. (#21267) ... * Improve argsort performance. - Give 200000 data to compute argsort on v100, can speed up ~190x before opt cost: 0.53s after opt cost:0.0027s - Add fp16 support * Refine error message * Refine code test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com>	6 years ago
Tao Luo	02270b3eb1	remove unused assert.h (#19529) ... test=develop	6 years ago
Yibing Liu	9386ac0a40	Enhance cuda code & unittest for argsort_op	7 years ago
Yibing Liu	92cfa2be3a	Avoid using dynamic array in cuda kernel	7 years ago
Yibing Liu	94e72ea6e7	Support more negative axes in argsort_op	7 years ago
Yibing Liu	42645ff779	Compute target index on gpu	7 years ago
Yibing Liu	6ee22c4f71	Add gpu kernel for argsort op	7 years ago