He Wei
|
43e0967024
|
Decouple ir::Tensor class from python
|
5 years ago |
yujianfeng
|
7ad877a948
|
Add Split fission pass
|
5 years ago |
WilliamLian
|
e3a26c2229
|
fix transdata insert cast
|
5 years ago |
mindspore-ci-bot
|
294449697a
|
!2179 Add sort by index for each group of AllReduce
Merge pull request !2179 from YuJianfeng/all_reduce_index
|
5 years ago |
yujianfeng
|
f15cb6b7c9
|
Add sort by index for each group of AllReduce
|
5 years ago |
huanghui
|
1243022f01
|
fix the back outputs of BNTrainingUpdateV3 may cover the front whitch is empty
|
5 years ago |
huanghui
|
ffe074b661
|
fix exception when FusedBatchNorm's first input 's shape dims not equal 4
|
5 years ago |
huanghui
|
88eec2b894
|
fix single-batchnorm-fission && softmax-grad-ext-fusion pass
|
5 years ago |
huanghui
|
118496b3ec
|
enhance insert memcpy
|
6 years ago |
mindspore-ci-bot
|
0624263496
|
!1940 Fix when batchnorm has just 4 outputs, it will be not fission
Merge pull request !1940 from huanghui/single-batchnorm-fission-pass
|
5 years ago |
huanghui
|
36d1aadf1c
|
fix when Batchnorm's output is 0,1,2,4, fission doesn't work
|
5 years ago |
WilliamLian
|
9808e47663
|
change checkAicpu to CheckAICPU & add charge Scalar function to charge the input or output is scalar
|
5 years ago |
huanghui
|
b4c0ed4b36
|
add signle batchnorm fission pass
|
6 years ago |
huanghui
|
cf87218fb7
|
place layernormgrad split pass before kernel select
|
6 years ago |
mindspore-ci-bot
|
59683a1d90
|
!1692 Fix topk bug for fasterrcnn
Merge pull request !1692 from meixiaowei/master
|
6 years ago |
yujianfeng
|
e6f1cfa581
|
Check the input size of BatchNorm before fission in bert
|
6 years ago |
meixiaowei
|
1778ec0135
|
fix topk bug
|
6 years ago |
yujianfeng
|
ee087bdf60
|
Check the size of topk input names before converting input to attr
|
6 years ago |
WilliamLian
|
9d364f509f
|
fix bug of transdata transpose split
|
6 years ago |
mindspore-ci-bot
|
7e2e7ad52b
|
!1367 modify topk split pass
Merge pull request !1367 from jjfeing/master
|
6 years ago |
WilliamLian
|
338d7c1ac1
|
decoupled of insert transdata and deal ref and split transdata
|
6 years ago |
jjfeing
|
654f09037e
|
fix topk split
|
6 years ago |
yujianfeng
|
e702e0bce8
|
Add tuple_getitem check for outputs of bn
|
6 years ago |
yujianfeng
|
d6f043b020
|
Fix batch norm bert fission for control depend case
|
6 years ago |
huanghui
|
c4af71e236
|
add LarsV2 fission pass
|
6 years ago |
yujianfeng
|
2b061c84c7
|
Add batch_norm_grad infer fisson
|
6 years ago |
WilliamLian
|
691b0648e3
|
convert unsupported kernel in aicore to aicpu
|
6 years ago |
yujianfeng
|
4a0ddaef59
|
Support specifying reshape type for batchnorm fused op
|
6 years ago |
YuJianfeng
|
aa6f808616
|
Add batch norm bert fission pass
|
6 years ago |
huanghui
|
0de0aee54e
|
Add BatchNormGrad split pass
|
6 years ago |
mindspore-ci-bot
|
3d9f1087c6
|
!695 Check topk supported before converting input to attr
Merge pull request !695 from YuJianfeng/master
|
6 years ago |
YuJianfeng
|
ce2a13fcda
|
Check topk supported before converting input to attr
|
6 years ago |
huanghui
|
9e29224af9
|
move add_memcpy_async pass and transdata_split pass
|
6 years ago |
YuJianfeng
|
bc2df2c913
|
Fix inputs size and attr for AddN fission pass
|
6 years ago |
wenchunjiang
|
ee5f3fa240
|
add pass to insert memcpy_async for get_next outputs
|
6 years ago |
YuJianfeng
|
7307c81f31
|
implement AddN fission pass
|
6 years ago |
zhunaipan
|
930a1fb0a8
|
initial version
Signed-off-by: leonwanghui <leon.wanghui@huawei.com>
|
6 years ago |