yujianfeng
|
85ff90c237
|
Add input shape condition for transpose_reshape fusion pass
|
5 years ago |
huanghui
|
12eaaf710b
|
mul_add_fusion pass supports when add's 2nd is mul
|
5 years ago |
YuJianfeng
|
aa6f808616
|
Add batch norm bert fission pass
|
5 years ago |
YuJianfeng
|
7185961e89
|
Enable BatchNorm fusion pass
|
5 years ago |
huanghui
|
0de0aee54e
|
Add BatchNormGrad split pass
|
5 years ago |
mindspore-ci-bot
|
3d9f1087c6
|
!695 Check topk supported before converting input to attr
Merge pull request !695 from YuJianfeng/master
|
5 years ago |
YuJianfeng
|
ce2a13fcda
|
Check topk supported before converting input to attr
|
5 years ago |
Wei Luning
|
f305d6f8a7
|
Revert "add pattern AdjustAllReduceMulAdd"
This reverts commit ea6958c50a.
|
5 years ago |
mindspore-ci-bot
|
928dcf4ec4
|
!554 Fix derelu fusion pass
Merge pull request !554 from huanghui/fix_derelu_fusion_pass
|
5 years ago |
huanghui
|
cd6e8d6542
|
fix ReluV2's mask shape in derelu fusion pass
|
5 years ago |
huanghui
|
14df771175
|
fix confusion_softmax_grad_rule pass
|
5 years ago |
liubuyu
|
672244e0ac
|
add keep_bn_fp32 parameter
|
5 years ago |
YuJianfeng
|
bc2df2c913
|
Fix inputs size and attr for AddN fission pass
|
5 years ago |
mindspore-ci-bot
|
71b63c3fcf
|
!246 [opt pass] AdjustAllReduceMulAdd
Merge pull request !246 from vlne-v1/I1E3PI-opt-pass-adjust-allreduce-apply-weight-decy-seq
|
5 years ago |
Wei Luning
|
ea6958c50a
|
add pattern AdjustAllReduceMulAdd
|
5 years ago |
YuJianfeng
|
5eb5379889
|
Add AdamApplyOne fusion pass
|
5 years ago |
mindspore-ci-bot
|
f69a668d98
|
!350 change tuple output to make tuple
Merge pull request !350 from lianliguang/chang-tuple-output-to-make-tuple
|
5 years ago |
lianliguang
|
00e4306518
|
convert all tuple output to maketuple
|
5 years ago |
mindspore-ci-bot
|
f98efafa16
|
!317 [IRFusion] add derelu_fusion pass
Merge pull request !317 from huanghui/derelu_fusion_pass
|
5 years ago |
wenchunjiang
|
ee5f3fa240
|
add pass to insert memcpy_async for get_next outputs
|
5 years ago |
laiyongqiang
|
3e05f50f5f
|
getnext_memcpy_elimination
|
5 years ago |
huanghui
|
b02e871c1a
|
[IRFusion] add derelu_fusion pass
|
5 years ago |
mindspore-ci-bot
|
94589ce611
|
!226 expend conv stride and dilation to 2d
Merge pull request !226 from wangnan39/expend_conv_stride_to_2d
|
5 years ago |
wangnan39@huawei.com
|
2604acedcb
|
extend conv stride and dilation to 2d
|
5 years ago |
mindspore-ci-bot
|
298a32f1e2
|
!132 Add confusion_mul_grad_fusion pass
Merge pull request !132 from huanghui/master
|
5 years ago |
huanghui
|
19ee376cd3
|
add confusion_mul_grad fusion pass
|
5 years ago |
YuJianfeng
|
7307c81f31
|
implement AddN fission pass
|
5 years ago |
biffex
|
62bbf560c6
|
constant duplicate mul for momentum
|
5 years ago |
buxue
|
7541d3b067
|
Develop op MaxPoolWithArgMax
|
5 years ago |
mindspore-ci-bot
|
a5d95e472e
|
!106 support comparison ops for python
Merge pull request !106 from amongo/SupportPythonOperators
|
5 years ago |
huangdongrun
|
461d8e3a08
|
add comparison ops
fix pylint
use scalar_lt primitive directly
fix review
|
5 years ago |
YuJianfeng
|
7f53bb062d
|
Optimize depend edge with make tuple input
|
5 years ago |
GinFung
|
468dbc3557
|
Add matmul biasadd fusion pass
|
5 years ago |
zhunaipan
|
930a1fb0a8
|
initial version
Signed-off-by: leonwanghui <leon.wanghui@huawei.com>
|
5 years ago |