i-robot
|
e14942017d
|
!31673 support tranpose with neg perm
Merge pull request !31673 from 范吉斌/transpose_neg
|
4 years ago |
i-robot
|
c1a9e313b0
|
!31281 gpu dynamic shape
Merge pull request !31281 from changzherui/gpu_dynamic_shape
|
4 years ago |
fan-jibin
|
37a658e4c2
|
support tranpose with neg perm
|
4 years ago |
changzherui
|
df3976d5a0
|
gpu dynamic shape
|
4 years ago |
ttudu
|
adb18b82ca
|
dynamic shape CPU & GPU
|
4 years ago |
i-robot
|
27e417de6a
|
!31271 Reconstruct the GPU invoking mode
Merge pull request !31271 from zong_shuai/reconstruct_gpu_kernel
|
4 years ago |
zong-shuai
|
d42daf0dcb
|
reconstruct_gpu_kernel
|
4 years ago |
hanhuifeng2020
|
089ce0386a
|
[DynamicShape][Gpu]support dynamic shape for the DCN network
|
4 years ago |
changzherui
|
47ef477fef
|
modify gpu opt adapter dynamic shape
|
4 years ago |
hanhuifeng2020
|
b571e67795
|
[DynamicShape][GPU]add dynamic shape support some ops for new network
|
4 years ago |
ttudu
|
dac6f5425d
|
delete GPU/CPUDynamicKernel
|
4 years ago |
ckey_Dou
|
e9679ca0bd
|
return default shape when max_shape is empty
|
4 years ago |
zhoufeng
|
f49b195c39
|
extract common as an independent shared library
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
|
4 years ago |
i-robot
|
b3c5943bf8
|
!30373 [DynamicShape][GPU]add dynamic shape support of Concat op and its backward for new Network
Merge pull request !30373 from hanhuifeng/dcn_dyn_ops_1
|
4 years ago |
hanhuifeng2020
|
662c51c019
|
[DynamicShape][GPU]add dynamic shape support of Concat and its backward for DCN
|
4 years ago |
jinjiali
|
b2d1bad62c
|
package the cuda operations as a dynamic link library
|
4 years ago |
i-robot
|
6f88bf7c75
|
!30324 rename scipy ops matrix_diag_part to matrix_diag_part_V3
Merge pull request !30324 from wuwenbing/master
|
4 years ago |
wenbean
|
638ba5b4d1
|
Fix ops matrix_diag_part different version interplay infershap between ascend and gpu/cpu
|
4 years ago |
lianliguang
|
5c8af66e53
|
support TensorShape ops to instead of DynamicShape ops
|
4 years ago |
yanghaoran
|
739c4b236a
|
回退 'Pull Request !30120 : Package the cuda operators as a dynamic link library'
|
4 years ago |
i-robot
|
8e21ffae65
|
!30120 Package the cuda operators as a dynamic link library
Merge pull request !30120 from jinjiali-kali/cuda_ops_3
|
4 years ago |
jinjiali
|
cd7db0f57f
|
package the cuda operations as a dynamic link library
|
4 years ago |
changzherui
|
83e6ea3d71
|
add gpu ResetResource
|
4 years ago |
wenbean
|
fddedb03c8
|
Matrix ops move to arrays dir, use scure func
|
4 years ago |
marui
|
d47e79b04c
|
Refactor ccsrc directories and CMakeLists files
|
4 years ago |