- 07 11月, 2022 3 次提交
-
-
由 ykkk2333 提交于
add roll and roll_grad kernels and strided_slice and strided_slice_grad kernels, test=kunlun (#47368) * add stat tool * add roll and roll_grad kernels and strided_slice and strided_slice_grad kernels, test=kunlun
-
由 Sławomir Siwek 提交于
* init changes * bnorm * method signature * change order * bnorm * removed unused args
-
由 Sławomir Siwek 提交于
* remove fwd funcs * migrate conv grads
-
- 04 11月, 2022 10 次提交
-
-
由 Qi Li 提交于
* [NPU] add more attrs into npu storiages, test=develop * rename to storage_properties_initialized
-
由 zyfncg 提交于
* generate static graph code for some activation op * fix bug * fix infermeta of selected_rows
-
由 cyber-pioneer 提交于
* add sin double grad operator * add sin double grad test example * move sindoublegradopmaker to backward.yaml * fix sindoublegrad code * simplify sindoublegrad functor
-
由 houj04 提交于
* [XPU] add cumsum op. test=kunlun * try to fix linker. test=kunlun * try to fix linker. test=kunlun * try to fix linker. test=kunlun * debug. test=kunlun * update xpu.cmake. remove unnecessary codes. test=kunlun.
-
由 pangyoki 提交于
-
由 Sławomir Siwek 提交于
-
由 Piotr Paturej 提交于
* add extra attr property set * add type_info for all context * add onednn context to all context * fix context compile error * simplify conv kernel args * pass runtime attr into dev_ctx * fix marco error * clear conv_grad_kernel extra args * merge conv_grad_grad into conv_grad * clear conv2d_grad_grad extra attrs * clear yaml and eager extra attr * fix conv1d error * change to thread local * fix npu compile failed * try to fix windows compile failed * add conv2d onednn phi kernel * fix ci bugs (#36) * fix compile bugs (#38) * fix extra input transform bug (#39) * support dynamic created attr (#40) * reset extra info gen code * rm conv_grad_grad kernel * reimpl pass attr adapting * add int attr support * remove vector inputnames creating * fix map at error * Update paddle/phi/kernels/onednn/conv_grad_kernel.cc Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com> * remove useless extra attrs * replace mkldnn_engine by onednn_engine * Migrate pool+grad to PHI * Update paddle/fluid/operators/mkldnn/test_mkldnn_op_nhwc.cc Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com> * Update paddle/phi/kernels/onednn/pool_grad_kernel.cc Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com> * Update paddle/phi/kernels/onednn/pool_kernel.cc Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com> Co-authored-by: NChen Weihang <chenweihang@baidu.com> Co-authored-by: NYuanRisheng <yuanrisheng@baidu.com> Co-authored-by: NChen Weihang <chenwhpro@163.com> Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>
-
由 Sławomir Siwek 提交于
* add extra attr property set * add type_info for all context * add onednn context to all context * fix context compile error * simplify conv kernel args * pass runtime attr into dev_ctx * fix marco error * clear conv_grad_kernel extra args * merge conv_grad_grad into conv_grad * clear conv2d_grad_grad extra attrs * remove redundant imports * migrate softmax * clear yaml and eager extra attr * fix conv1d error * change to thread local * fix npu compile failed * try to fix windows compile failed * add conv2d onednn phi kernel * fix ci bugs (#36) * fix compile bugs (#38) * fix extra input transform bug (#39) * support dynamic created attr (#40) * reset extra info gen code * rm conv_grad_grad kernel * reimpl pass attr adapting * add int attr support * remove vector inputnames creating * merge dev * fix map at error * adjust attribute * adapt funcs to PHI * init * adjust imports * support postops * format codeblocks * revert changes to softmax Co-authored-by: NChen Weihang <chenweihang@baidu.com> Co-authored-by: NYuanRisheng <yuanrisheng@baidu.com>
-
由 ykkk2333 提交于
-
由 zhangyikun02 提交于
-
- 03 11月, 2022 8 次提交
-
-
由 wanghuancoder 提交于
-
由 zhouweiwei2014 提交于
-
由 Nyakku Shigure 提交于
* [CodeStyle][py2][U008] remove unnecessary args in `super()` * remove remained args * revert changes in test_pylayer_op * Revert "revert changes in test_pylayer_op" This reverts commit ff185a9ae738afac3b0264f61bde6c6b7f72e7c4. * revert some changes in example code
-
由 sneaxiy 提交于
-
由 Sławomir Siwek 提交于
* add extra attr property set * add type_info for all context * add onednn context to all context * fix context compile error * simplify conv kernel args * pass runtime attr into dev_ctx * fix marco error * clear conv_grad_kernel extra args * merge conv_grad_grad into conv_grad * clear conv2d_grad_grad extra attrs * remove redundant imports * migrate softmax * clear yaml and eager extra attr * fix conv1d error * change to thread local * fix npu compile failed * try to fix windows compile failed * add conv2d onednn phi kernel * fix ci bugs (#36) * fix compile bugs (#38) * fix extra input transform bug (#39) * support dynamic created attr (#40) * reset extra info gen code * rm conv_grad_grad kernel * reimpl pass attr adapting * add int attr support * remove vector inputnames creating * merge dev * fix map at error * adjust attribute * adapt funcs to PHI Co-authored-by: NChen Weihang <chenweihang@baidu.com> Co-authored-by: NYuanRisheng <yuanrisheng@baidu.com>
-
由 zhangkaihuo 提交于
-
由 zhouweiwei2014 提交于
-
由 YuanRisheng 提交于
-
- 02 11月, 2022 8 次提交
-
-
由 zhangbo9674 提交于
* fix ci bug * test
-
由 Tao Luo 提交于
-
由 Chen Weihang 提交于
* add storage properties for npu * fix compile failed * fix api name mismatch * polish design
-
由 YuanRisheng 提交于
* Standardise batch norm * standardize conv3d and depwise_conv2d * fix ci bugs
-
由 zhouweiwei2014 提交于
-
由 zyfncg 提交于
* support generating static code of high order grad op by yaml * polish code
-
由 houj04 提交于
* [XPU] add int64 support for slice and subtract. test=kunlun * try to fix xpu compile. test=kunlun * try to fix xpu compile. test=kunlun * try to fix xpu compile. test=kunlun * remove unnecessary modification. test=kunlun
-
由 Tian Zheng 提交于
* Add build option for CUDNN Frontend API * Fix review comments * Change namespace for cudnn_frontend.h
-
- 01 11月, 2022 8 次提交
-
-
由 Siming Dai 提交于
-
由 limingshu 提交于
* first commit * transpose_kernel_optimization * first complishment of transpose op * second commit * refine code logics of tranpose_kernel * refine transpose kernel * first commit * fix DtoD copy bugs for hip * refine code according to the PR advice * change dim to int64_t type. * fix some type error
-
由 YuanRisheng 提交于
* standard_api * add hardtanh
-
由 xiongkun 提交于
* Einsum Support Complex * code fix * add unittest for complex grad with einsum * set rtol=1e-4 * fix
-
由 Wang Xin 提交于
-
由 zyfncg 提交于
* gene static graph code for ceil, expm1 op * gene static graph code for some activation op * fix bug * revert doc of silu and logsigmoid
-
由 Chen Weihang 提交于
* add extra attr property set * add type_info for all context * add onednn context to all context * fix context compile error * simplify conv kernel args * pass runtime attr into dev_ctx * fix marco error * clear conv_grad_kernel extra args * merge conv_grad_grad into conv_grad * clear conv2d_grad_grad extra attrs * clear yaml and eager extra attr * fix conv1d error * change to thread local * fix npu compile failed * try to fix windows compile failed * add conv2d onednn phi kernel * fix ci bugs (#36) * fix compile bugs (#38) * fix extra input transform bug (#39) * support dynamic created attr (#40) * reset extra info gen code * rm conv_grad_grad kernel * reimpl pass attr adapting * add int attr support * remove vector inputnames creating * fix map at error * Update paddle/phi/kernels/onednn/conv_grad_kernel.cc Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com> * remove useless extra attrs * replace mkldnn_engine by onednn_engine Co-authored-by: NYuanRisheng <yuanrisheng@baidu.com> Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>
-
由 umiswing 提交于
-
- 31 10月, 2022 3 次提交
-
-
由 YuanRisheng 提交于
* standard api * fix ci bugs * fix ci bugs * fix ce bugs
-
由 xiongkun 提交于
* add unittest for einsum-v2-trace and diagonal * repeat labels. * einsum support repeated labels. * forward is ok for diagonal and undiagonalized. TODO: check backward is ok by our theorem. * backward is ok! * fix by PR suggestions. * fix ci error * fix ci error * fix ci warning
-
由 ronnywang 提交于
* [CustomDevice] GetCCLComm add custom device support * update * update * update
-