zengzitao
62458b5636
adapt graph kernel for cpu
4 years ago
zengzitao
7d6c6b17bb
fix master warnings
4 years ago
tronzhang
134ffa9efe
support tot ops
4 years ago
ms_yan
36a8886ca2
Revert "[feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset"
This reverts commit b077aa1cab .
Revert "[feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset"
This reverts commit 4e6f7dc97d .
delete pass_registry_test.cc
comment hiai_nlu_model_multi.pb related line
4 years ago
djc
4e6f7dc97d
[feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset
4 years ago
i-robot
ff07de80b4
!21603 Fix static-check warnings
Merge pull request !21603 from DeshiChen/0809_clean
4 years ago
dayschan
9add26ad99
Add expanders in c++ code
transplant the op expander code from python to c++, base on LiteGraph.
the c++ expander will be called in priority if it was registered in OpExpanderFactory.
add two examples, BiasAdd and ExpandDims.
remove BiasAdd from python expanders.
since the ExpandDims is also imported by other ops (e.g. BatchNorm), we don't remove it now.
4 years ago
dayschan
fff39e7451
fix static check problems
4 years ago
zengzitao
b8095efcab
gk fix warnings in master
4 years ago
chenlei_autodiff
0271535429
[GraphKernel] fix bert and add graph kernel ops.
4 years ago
i-robot
ab78bcf16f
!19081 add ops IsNan
Merge pull request !19081 from lingyunli63/add_ops_IsNan
4 years ago
Yang Jiao
582669e41c
add expander checkings
4 years ago
lingyunli63
50a66ae476
isnan isfinite isinf squaresumall identity oneslik
4 years ago
yanglf1121
c30b1e6d06
add graph kernel div, floordiv, mod, floormod, floor
4 years ago
wangrao124
7cddde47b0
!215 add ops: LogicalNot, LogicalAnd, LogicalOr, NotEqual, EqualCount, Asinh, Acosh
* add ops: LogicalNot, LogicalAnd, LogicalOr, NotEqual, EqualCount, Asinh, Acosh
4 years ago
chenlei_autodiff
7d55cef106
[GraphKernel] add sponge ops.
4 years ago
zengzitao
aa019a639f
support complex in gpu about graph_kernel
4 years ago
i-robot
92675f7572
!18747 Add expander for op AddN
Merge pull request !18747 from DeshiChen/0618_expand_addn
4 years ago
dayschan
149dab39c5
Add expander for AddN; update akg submodule
4 years ago
lishanni513
4a64bc6bc5
Enable TensorCore for Bert-Base on GPU, enable akg-conv2d for ssd on GPU
4 years ago
zhengzuohe
448d1cdbb6
Update submodule mindspore/akg
Enable Graph Kernel for W&D Host-Device Mode and ResNext50 on GPU
Enable akg-conv2d in yolov3_darknet on GPU
4 years ago
i-robot
5d5f38c501
!18428 [GraphKernel] Enable Auto-Mixed-Precision & TensorCore for AlexNet on GPU
Merge pull request !18428 from lishanni/master_alexnet
4 years ago
lishanni513
bddfd9ac65
Enable Auto-Mixed-Precision & TensorCore for AlexNet on GPU
4 years ago
i-robot
04f42b9901
!18629 stop mutmul to mul on GPU
Merge pull request !18629 from lingyunli63/stop_matmul_to_mul_GPU
4 years ago
lingyunli63
e810ebb3d0
matmul to mul only work on aicore
4 years ago
zengzitao
8064de7931
fix maximum_grad and minimum_grad input_shape not equal to output_shape bug
4 years ago
i-robot
07f58b0b46
!17626 matmul to mul
Merge pull request !17626 from lingyunli63/matmul_to_mul
4 years ago
lingyunli63
4f34e537a0
replace matmul/batchmatmul by mul when k is 1
4 years ago
Yang Jiao
4d1f8d1d00
fix expander bugs
4 years ago
mindspore-ci-bot
7e5d68991a
!17680 Avoid overflow of in realdiv
From: @wenfangpei
Reviewed-by: @gaoxiong1,@ckey_dou
Signed-off-by: @ckey_dou
4 years ago
wenfangpei
fa531a473d
avoid overflow in realdiv
4 years ago
Yang Jiao
d9a60e0d28
fix static warnings
4 years ago
mindspore-ci-bot
84859aba23
!16695 bug fix in lamb
From: @wenfangpei
Reviewed-by: @gaoxiong1,@ckey_dou
Signed-off-by: @ckey_dou
4 years ago
wenfangpei
fd84c20a6a
bug fix in lamb
4 years ago
looop5
68f55e1e93
expand conv2d when input format is DefaultFormat but attr format is NHWC
4 years ago
looop5
dd81f47271
add conv2d expander
use data_format instead of format in op_infer
add pad and unpad inside conv2d expander
fix pad value
add limit in conv2d expander
4 years ago
zengzitao
516487c9cd
fix log(inputs), inputs less than 0.0, cause loss inf bug
4 years ago
mindspore-ci-bot
08959ddb74
!16224 [GraphKernel]Fix wide & deep ps mode precision error
From: @zengzitao
Reviewed-by: @gaoxiong1,@ckey_dou
Signed-off-by: @ckey_dou
4 years ago
zengzitao
15cf743ede
rewrite sigmoid_cross_entropy_with_logits expander to avoid overflow
4 years ago
wenfangpei
2e2e850b4a
adapt batchmatmul for GraphKernel in ascend
4 years ago
mindspore-ci-bot
ed539597c2
!15415 [GraphKernel]adapt for logsoftmax in ascend
From: @wenfangpei
Reviewed-by: @gaoxiong1,@ckey_dou,@gaoxiong1,@ckey_dou
Signed-off-by: @ckey_dou,@ckey_dou
4 years ago
wenfangpei
4174a7b38f
expanders of some fusion ops
4 years ago
wenfangpei
db8256e61f
adapt for logsoftmax in ascend
4 years ago
wenfangpei
c41875b318
adapt expanders of some ops from gpu to ascend
4 years ago
mindspore-ci-bot
52e7f51970
!15741 [GraphKernel] batchnorm expander supports when first input is float16
From: @looop5
Reviewed-by: @gaoxiong1,@dylangeng
Signed-off-by: @dylangeng
4 years ago
looop5
24f441ba33
batchnorm expander supports when first input is float16
4 years ago
chenlei_autodiff
fd227bb448
[graph kernel] clean code for expanders.
4 years ago
mindspore-ci-bot
168c64b60d
!15648 [GraphKernel] negative axis in Squeeze expander.
From: @chenlei_autodiff
Reviewed-by: @gaoxiong1,@anyrenwei
Signed-off-by: @anyrenwei
4 years ago
zengzitao
8dcff8d83c
refactor tile op and in expander open on gpu
4 years ago
chenlei_autodiff
b419f60b0d
[GraphKernel] negative axis in Squeeze expander.
4 years ago