- 04 3月, 2021 1 次提交
-
-
由 Leo Chen 提交于
* add npu kernel for equal op * refine code * add more ut * update year
-
- 02 3月, 2021 3 次提交
-
-
由 Leo Chen 提交于
* init commit, add layer_norm npu kernel * fix typo * add unittest * add unittest * fix bug * fix bug * refine ut
-
由 Void Main 提交于
Refactor HCCLCommContext to be compatible with Paddle (#31359)
-
由 Leo Chen 提交于
* fix reading flags from env * fix problem caused by async run * support partial grad * support elementwise_add_grad npu kernel * add unittest * fix bug?
-
- 01 3月, 2021 3 次提交
- 26 2月, 2021 1 次提交
-
-
由 liym27 提交于
* [NPU] Support npu op: (1) pow (2) pow_grad * Support fp16
-
- 23 2月, 2021 1 次提交
-
-
由 Leo Chen 提交于
Fix compilation problem (#31100)
-
- 22 2月, 2021 1 次提交
-
-
由 Leo Chen 提交于
* add npu sub op * fix typo * rename test * fix bug * fix bug * add fp16 kernel * fix typo * support sub grad op * support elementwise_sub_grad op Co-authored-by: Nfrankwhzhang <frankwhzhang@126.com>
-
- 09 2月, 2021 3 次提交
-
-
由 Leo Chen 提交于
* support npu allocator * add npu device context * fix some compile problem * fix some compile problem * add npu info * compile ok * fix include dir * support naive_best_fit_allocator * run ut ok, bug failed to exit * call aclrtResetDevice before exit * fix aclFinilize * add system allocatot test * add selected_gpus in gtest * add tensor_test for npu * support npu op, initial commit * add npu stream * add elementwise_add_op * compile ok * fix typo * fix elementwise_add_op_npu_test * support op run * test can run but failed * change aclopExecuteV2 to aclopCompileAndExecute
-
由 Leo Chen 提交于
[feature] support npu operator
-
由 Leo Chen 提交于
[feature] support npu allocator
-
- 21 1月, 2021 2 次提交
-
-
由 gongweibao 提交于
Add Hccl program group
-
由 gongweibao 提交于
Add distribution supported
-
- 15 1月, 2021 4 次提交
-
-
由 hutuxian 提交于
-
由 石晓伟 提交于
-
由 Wojciech Uss 提交于
-
由 wawltor 提交于
* fix the rnn mask memory bug for out of read * update the code for the rnn
-
- 14 1月, 2021 2 次提交
-
-
由 taixiurong 提交于
-
由 ShenLiang 提交于
-
- 13 1月, 2021 1 次提交
-
-
由 GaoWei8 提交于
* softmax backward optimize
-
- 12 1月, 2021 8 次提交
-
-
由 JZ-LIANG 提交于
-
由 lidanqing 提交于
-
由 tangwei12 提交于
* rename sendrecv.proto to namespace paddle.distributed * split ps with distributed
-
由 Double_V 提交于
* fix elugradgrad test fail and error message opt * fix unitest,test=develop * Update prroi_pool_op.h fix error message * opt message,test=develop * fix ci fail,test=develop
-
由 Zhen Wang 提交于
* Fix the accuracy problem of allclose op when using float64 data type in static mode. * Format the code style.
-
由 yaoxuefeng 提交于
-
由 furnace 提交于
-
由 石晓伟 提交于
-
- 11 1月, 2021 9 次提交
-
-
由 Chengmo 提交于
* fix cmake Co-authored-by: NseiriosPlus <tangwei12@baidu.com>
-
由 石晓伟 提交于
-
由 liym27 提交于
Support vector<double> as type of op attribute and op set_value suppport vector<double> as value (#30126)
-
由 wangchaochaohu 提交于
-
由 AshburnLee 提交于
-
由 chentianyu03 提交于
* type promotion for grad * add type promotion for div op
-
由 liym27 提交于
-
由 WeiXin 提交于
* modify error message based on comments * edit code according to review. * Correct spelling according to review.
-
由 wawltor 提交于
add the op error message for the matmul xpu
-
- 10 1月, 2021 1 次提交
-
-
由 GaoWei8 提交于
* optimize softmax forward
-