lizhenyu
4269dcece5
ps cache support save checkpoint
5 years ago
mindspore-ci-bot
ffe61081d3
!10189 fix shape type error when dynamic_kernel shape type is compute_depend
From: @liubuyu
Reviewed-by: @zhoufeng54,@kisnwang
Signed-off-by: @kisnwang
5 years ago
wilfChen
09e10e18bb
momentum weightdecay fusion
5 years ago
liubuyu
4d75d7b992
fix shape type error
5 years ago
mindspore-ci-bot
d8a64b4ac4
!9796 Add SpaceToDepth fission pass to fix bug when data type is float16.
From: @liu_xiao_93
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
5 years ago
liuxiao93
2bbd97d334
Add SpaceToDepth fission pass.
5 years ago
jjfeing
1984cf8e20
unify mindir
5 years ago
mindspore-ci-bot
be4e91339f
!9661 gpu relu optimize
From: @wilfchen
Reviewed-by: @cristoval,@limingqi107
Signed-off-by: @limingqi107
5 years ago
wilfChen
c1d3bd2160
relu optimize
5 years ago
zhouyuanshen
e9aca01620
add support to reduceAny and reduceAll on gpu
5 years ago
mindspore-ci-bot
32444fbbd5
!8870 hccl send receive op
From: @huaweib
Reviewed-by: @jjfeing,@kisnwang
Signed-off-by: @kisnwang
5 years ago
liubuyu
e3fa342d72
support 3d format
5 years ago
baihuawei
7d09dff880
add hccl send recv
5 years ago
TFbunny
5e19a642f9
fix and add testcase for dynamic shape scatteradd/update transpose
5 years ago
mindspore-ci-bot
c78683a411
!8981 gatherv2 pad optimizer in dynamic shape scene
From: @yao_yf
Reviewed-by: @stsuteng,@kisnwang
Signed-off-by: @stsuteng
5 years ago
yao_yf
444cb99b40
gather_v2 pad optimizer pass
5 years ago
liuxiao93
584e241e29
Adapt ops LinSpace for Ascend.
5 years ago
lizhenyu
094f0b2a07
bugfix:fused batch norm op's input channel nums should be a multiple of 4
5 years ago
fangzehua
69ce58425d
fix reshape dynamic and emb
5 years ago
LianLiguang
bb6148661f
change mixedprecision of pynative
5 years ago
liuxiao93
d471ac491e
Adapt DynamicGRUV2 forward for Ascend new backend.
5 years ago
jjfeing
3feffc7d62
fix ubfusion bug
5 years ago
mindspore-ci-bot
a5b0d13141
!8079 support GNMT net fix dynamic rnn grad fission pass
Merge pull request !8079 from liubuyu/op_support
5 years ago
liubuyu
662976a75d
dynamic rnn fission pass v2
5 years ago
liuxiao93
45d343257b
Add DynamicGRU.
5 years ago
hwjiaorui
3698b9fd54
register proximal adagrad ds
import proximal adaagrad ds
tiling map
style check bug
add op set
5 years ago
VectorSL
509b25ef1e
gpu nhwc
5 years ago
kswang
74c7bdd471
fix segmentfault with fused sparse ftrl
5 years ago
kswang
ece27f313e
enable async run
5 years ago
caifubi
d3b978147f
Ascend Dynamic Shape
5 years ago
mindspore-ci-bot
21c5607fca
!6971 cudnn inplace optimizer
Merge pull request !6971 from chenweifeng/tensoradd_inplace
5 years ago
wilfChen
b420b6cda7
cudnn inplace optimizer
5 years ago
liubuyu
8af3250477
support dynamic_rnn and dynamic_rnn_grad op
5 years ago
mindspore-ci-bot
7b3873559f
!5883 support for frac_zn_lstm
Merge pull request !5883 from liubuyu/master
5 years ago
dayschan
37a48f6aac
GraphKernel supports GPU
1. Update akg submodule
2. Refactor akg_kernel_build, akg_ascend_kernel_build, akg_gpu_kernel_build
3. Add akg_kernel_json_decoder to support converting kernel_json to AnfNode.
4. Add GraphKernel Cost Model. (mindspore/_extends/graph_kernel)
5. Add some GraphKernel passes to GpuSession, move these passes to backend/optimizer/graph_kernel.
6. Add global id for ir files.
7. Fix bug in ConstInputToAttr.
5 years ago
liubuyu
23a298ca81
support new format frac_zn_lstm
5 years ago
wuxuejian
bd527a331d
update aicpu proto and update module: graphengine
Support Dynamic Shape Aicpu Run Package
adapt tensorengin modify, fix ub fusion
5 years ago
wilfChen
13dd31f56c
reorder fused optimizer
5 years ago
mindspore-ci-bot
1944b8e53b
!5612 Resnet50 pattern Fusion
Merge pull request !5612 from chenweifeng/BatchNormAddReluGrad
5 years ago
wilfChen
5316061fa3
gpu resnet50 fusion
5 years ago
yujianfeng
4b77f6b53c
Add AdamApplyOneWithDecayAssign fusion pass
5 years ago
WilliamLian
097f53bed9
add attr for transdata node
5 years ago
limingqi107
5b76e8f3d7
gpu add format transform pass
5 years ago
lizhenyu
7ddddc41a9
add FusedBatchNoramEx gpu kernel
5 years ago
gukecai
66e7b02b4b
independent stream parallel
5 years ago
mindspore-ci-bot
3fb58fcbe4
!4585 add gpu nccl broadcast
Merge pull request !4585 from baihuawei/broadcast
5 years ago
gukecai
6362e954df
Revert "independent stream parallel"
This reverts commit adb6ff6c78 .
5 years ago
baihuawei
b9ebd9c280
add gpu nccl broadcast
5 years ago
wuyongkang
78611b5d5b
fix static check problems
5 years ago
mindspore-ci-bot
162a356fc8
!4345 new ctrl for parallel and iteration num
Merge pull request !4345 from gukecai/parallel-genmask
5 years ago