i-robot
|
fd8eda24c4
|
!19904 fix fusion type
Merge pull request !19904 from laiyongqiang/fusionType
|
4 years ago |
i-robot
|
49c8d6a27e
|
!19789 Add Diag and DiagPart op for Ascend.
Merge pull request !19789 from liuxiao93/add-Diag-and-DiagPart
|
4 years ago |
LaiYongqiang
|
898a48d6ea
|
fix fusion type
|
4 years ago |
i-robot
|
63672de8e6
|
!20038 fix ReduceEltwiseFusionPass
Merge pull request !20038 from yuchaojie/ub_fusion
|
4 years ago |
i-robot
|
5cde059ef3
|
!19854 clearing the untrusted code of quantization
Merge pull request !19854 from Erpim/master
|
4 years ago |
yuchaojie
|
7f75f8a21e
|
fix ReduceEltwiseFusionPass
|
4 years ago |
liuxiao93
|
064ead40e1
|
Add Diag and DiagPart op for Ascend.
|
4 years ago |
i-robot
|
305d5236a7
|
!19588 fix some ub fusion patterns
Merge pull request !19588 from yuchaojie/ub_fusion
|
4 years ago |
Erpim
|
300456a2d7
|
clearing the untrusted code of quantization
|
4 years ago |
i-robot
|
5d6cac71d2
|
!19382 Support Complex in GPU about graph kernel
Merge pull request !19382 from ZengZitao/complex_gk
|
4 years ago |
zengzitao
|
aa019a639f
|
support complex in gpu about graph_kernel
|
4 years ago |
yuchaojie
|
b3420b2cff
|
fix some ub fusion patterns
|
4 years ago |
liuyihong
|
0f10ad9677
|
Add transform_op_optimizer
This submission implements the MinCut algorithm.
|
4 years ago |
yuchaojie
|
fdbfdc56a5
|
fix bnupdate_eltwise_eltwise_fusion_pass
|
4 years ago |
caifubi
|
c25de59f77
|
fix allreduce fusion index bug
|
4 years ago |
i-robot
|
57464dd61b
|
!19249 fix group FracZ transdata in multi-graph scene
Merge pull request !19249 from yuchaojie/op_select
|
4 years ago |
i-robot
|
d0bab79fde
|
!19252 add dump pb and nccl protected
Merge pull request !19252 from limingqi107/bug_fix3
|
4 years ago |
i-robot
|
d20b3d8faf
|
!19235 SliceGrad unify mindir adaption for CJ frontend
Merge pull request !19235 from yuchaojie/unify_ir
|
4 years ago |
limingqi107
|
70d156562a
|
add dump pb and nccl protected
|
4 years ago |
yuchaojie
|
45810c2adc
|
fix group FracZ transdata in multi-graph scene
|
4 years ago |
yuchaojie
|
76075c0b36
|
SliceGrad unify mindir adpation for Cangjie frontend
|
4 years ago |
chenlei_autodiff
|
b22c84a77d
|
[GraphKernel] enable graph kernel for bert optimizers and shufflenetv2 on GPU.
|
4 years ago |
i-robot
|
f8560b6109
|
!18919 Add TensorCopySlices op
Merge pull request !18919 from caifubi/master-tensor-copy-slices
|
4 years ago |
i-robot
|
92675f7572
|
!18747 Add expander for op AddN
Merge pull request !18747 from DeshiChen/0618_expand_addn
|
4 years ago |
caifubi
|
629111c6d3
|
Add TensorCopySlice Op
|
4 years ago |
i-robot
|
cc8d760c8b
|
!18980 Add NewCNodeWithInfos for kernel graph
Merge pull request !18980 from LiangZhibo/master
|
4 years ago |
dayschan
|
149dab39c5
|
Add expander for AddN; update akg submodule
|
4 years ago |
i-robot
|
2ffb014dc9
|
!18950 adapt for multi-frontend cj
Merge pull request !18950 from qingshanxiaozi/adapt_for_cj
|
4 years ago |
qingshanxiaozi
|
4aaac53e21
|
adapt for multi-frontend cj
Signed-off-by: qingshanxiaozi <qingshanxiaozi@163.com>
|
4 years ago |
dingpeifei
|
63784e49f5
|
upgrade_ascend_0626_mindspore
|
4 years ago |
l00591931
|
c7bc867ae4
|
Add CloneCNodeWithInfos for kernel graph
|
4 years ago |
i-robot
|
80003511a7
|
!18906 support group conv2d in pynative
Merge pull request !18906 from yuchaojie/op_select2
|
4 years ago |
yuchaojie
|
bf13a8a01e
|
support group conv2d in pynative
|
4 years ago |
i-robot
|
311e59c190
|
!17297 dynamic_shape_pipeline
Merge pull request !17297 from wanyiming/dynamic_shape_pipeline
|
4 years ago |
i-robot
|
865846a2a2
|
!18829 fix dynamic parameter change
Merge pull request !18829 from jjfeing/master
|
4 years ago |
i-robot
|
a87fd539d4
|
!18726 Fix cudnn malloc memory failed.
Merge pull request !18726 from linqingke/gpu_memory
|
4 years ago |
jjfeing
|
851f4b46dd
|
fix dynamic parameters
|
4 years ago |
linqingke
|
d0e2fabf1f
|
fix cudnn malloc memory failed.
|
4 years ago |
i-robot
|
76ef295b5c
|
!18764 move OutputUsedNum attr setting from fusion passes to ub_pattern_fusion
Merge pull request !18764 from yuchaojie/ub_fusion2
|
4 years ago |
i-robot
|
0c360ea2d6
|
!16766 Add recompute fuse
Merge pull request !16766 from lingyunli63/recompute_fuse
|
4 years ago |
yuchaojie
|
fc80ba655f
|
move OutputUsedNum attr setting from fusion passes to ub_pattern_fusion
|
4 years ago |
i-robot
|
0966595cfd
|
!18538 support FracZ format conversion when conv2d group > 1
Merge pull request !18538 from yuchaojie/op_select
|
4 years ago |
yuchaojie
|
689158f79b
|
FracZ format conversion when conv2d group > 1
|
4 years ago |
lingyunli63
|
a995bea507
|
recompute_fuse
|
4 years ago |
wanyiming
|
b37c99b112
|
dynamic
|
4 years ago |
kswang
|
3247c00555
|
optimize heter memcpy
|
5 years ago |
i-robot
|
35b15bda0b
|
!18465 clean code for memory reuse
Merge pull request !18465 from laiyongqiang/mem_clean
|
4 years ago |
i-robot
|
19fcb3070c
|
!18457 code clean enterprise
Merge pull request !18457 from zhaosida/code_clean_enterprise
|
4 years ago |
LaiYongqiang
|
d4d6fb940d
|
memory reuse code clean
|
4 years ago |
i-robot
|
f3d799f1b1
|
!18402 [GraphKernel] Enable TensorCore for training wide&deep and deepfm on GPU
Merge pull request !18402 from lishanni/master
|
4 years ago |