Commit Graph

271 Commits

Author SHA1 Message Date
mindspore-ci-bot f3aaf9b20f !16103 [MS][LITE] move mindir proto to core
From: @jianghui58
Reviewed-by: @zhoufeng54,@jpc_chenjianping
Signed-off-by: @jpc_chenjianping
2021-05-10 11:08:53 +08:00
jianghui58 d5ecee383a move mindir proto to core 2021-05-08 16:40:42 +08:00
wangshuide2020 9f058af6d8 add adapter of Erf, Expm1, Inv, etc. for graphengine. 2021-05-08 16:00:55 +08:00
mindspore-ci-bot eeb32f410e !15365 All the descriptions of batch normal under the mindspore folder have been uniformly updated to "Batch Normalization"
From: @dinglinhe123
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
2021-04-20 09:20:41 +08:00
dinglinhe 865cf68243 All the descriptions of batch normal under the mindspore folder have been uniformly updated to 'Batch Normalization' 2021-04-19 16:37:51 +08:00
changzherui 69b58e1ef5 modify export onnx input 2021-04-19 15:46:16 +08:00
changzherui 7781f135cc modify topk adapter 2021-04-15 18:53:10 +08:00
wangshuide2020 a0a260dca1 add adapter of Conv3D and Conv3DTranspose operators for graphengine. 2021-04-14 21:56:54 +08:00
changzherui f1543b69f2 modify adapter tensor size limit for ma 2021-04-12 23:22:42 +08:00
wangshuide2020 4e8bfc2862 1.add adapter of Mod, MaxPool3D and BCEWithLogitsLoss operators for graphengine.
2.fix the backword of matrixinverse.
2021-04-09 16:52:17 +08:00
mindspore-ci-bot e2260a2f09 !14612 remove ControlDepend
From: @huangbingjian
Reviewed-by: 
Signed-off-by:
2021-04-07 18:36:29 +08:00
huangbingjian 63a89925ff remove ControlDepend and its use 2021-04-07 11:30:40 +08:00
changzherui e9ab8016eb add read air proto func 2021-04-07 00:46:28 +08:00
mindspore-ci-bot d00ad5f11b !14504 Add onnx exporter error info
From: @liuyang_655
Reviewed-by: @zh_qh,@zhunaipan
Signed-off-by: @zh_qh
2021-04-02 08:54:41 +08:00
liuyang_655 8a3c74a722 add error info 2021-04-01 16:30:00 +08:00
changzherui c7323b39b3 modify export mindir tuple[] 2021-03-31 19:56:42 +08:00
mindspore-ci-bot dfd368a574 !12757 Change all io_format in master
From: @liangzhibo
Reviewed-by: @kingxian
Signed-off-by: @kingxian
2021-03-25 10:20:44 +08:00
zhoufeng ec7f9f395a fix cmake lint
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
2021-03-21 15:43:33 +08:00
l00591931 7a683ed293 Change all io_format in master 2021-03-20 08:41:08 +08:00
mindspore-ci-bot 17654135aa !13233 modify attr_target when load and export
From: @wangnan39
Reviewed-by: 
Signed-off-by:
2021-03-17 14:45:44 +08:00
wangnan39@huawei.com 93b1243806 modify attr_target when load and export 2021-03-16 19:38:56 +08:00
changzherui 44f8fbc219 mod Conv2DBackpropInput adapter 2021-03-14 00:47:46 +08:00
Zhang Qinghua f73f2d73f3 Set abstract value for reserved Depend node when merge Depend nodes. 2021-03-11 09:17:59 +08:00
mindspore-ci-bot 2b2964e6dd !12925 [MS][Maskrcnn][310infer]Maskrcnn 310 infer failed in halfway
From: @lanzhineng
Reviewed-by: 
Signed-off-by:
2021-03-09 16:43:41 +08:00
lanzhineng c8899e1f59 fix 310 maskrcnn infer failed in halway 2021-03-08 14:28:48 +08:00
l00591931 680324f225 Change make tuple in core.ops 2021-03-05 10:54:33 +08:00
mindspore-ci-bot e9a5d0248d !12573 export large mindir model
From: @changzherui
Reviewed-by: 
Signed-off-by:
2021-03-02 10:20:49 +08:00
changzherui b0070fe089 export large mindir model 2021-03-01 12:22:04 +08:00
l00591931 65cbb7b08f Change io_format in adapter 2021-02-25 10:14:24 +08:00
He Wei 7d9a783993 [auto-monad] Support side-effects by auto-monad
The basic idea is: exploits data dependency to control the execution order
of side-effect operations, and keep the semantics of ANF unchanged.

The ControlDepend primitive is removed and there are two primitives added:

1. UpdateState:
```
  a = Assign(para, value)
```
became:
```
  a = Assign(para, value, u)
  u = UpdateState(u, a)
```

2. Load:
```
  x = Add(para, value)
```
became:
```
  p = Load(para, u)
  x = Add(p, value)
  u = UpdateState(u, p)
```
2021-02-08 09:01:15 +08:00
mindspore-ci-bot 228a64de0f !11390 convert part of prim attr from str to int
From: @wangnan39
Reviewed-by: @kingxian
Signed-off-by:
2021-02-07 19:15:11 +08:00
wangnan39@huawei.com 2f79905f60 convert attr from str to enum 2021-02-07 17:03:17 +08:00
jinyaohui 30a27b2adb modify Gelu、FastGelu to GeLU and FastGeLU 2021-02-05 17:19:52 +08:00
mindspore-ci-bot aebe263dce !11895 unify mindir for different backend: the output num of optimizer ops, the backward of concat
From: @wangnan39
Reviewed-by: 
Signed-off-by:
2021-02-05 16:27:19 +08:00
mindspore-ci-bot c9aaa70b39 !12092 Change L2Norm, merge from r1.1 to master
From: @liangzhibo
Reviewed-by: @ginfung
Signed-off-by:
2021-02-05 14:13:12 +08:00
mindspore-ci-bot ad5b033cc5 Change L2Norm, r1.1 to master 2021-02-05 09:42:43 +08:00
jinyaohui 8022f9a6ed modify pack to stack 2021-02-04 18:54:31 +08:00
wangnan39@huawei.com cd9173fdfd unify the output num of optimizer ops 2021-02-04 17:46:00 +08:00
mindspore-ci-bot b189f177bb Change tuple_getitem to TupleGetItem and some other ops, merge from r1.1 to master 2021-02-02 15:54:15 +08:00
l00591931 9ec100d069 Change TensorAdd to Add, from r1.1 to master 2021-02-01 17:53:52 +08:00
mindspore-ci-bot 9fa0499fa0 Change GatherV2 to Gather r1.1 to master 2021-01-30 14:13:47 +08:00
mindspore-ci-bot 9d00e30ed0 !11544 modify export mindir for type tupletensor
From: @changzherui
Reviewed-by: @zh_qh,@zhoufeng54
Signed-off-by: @zh_qh
2021-01-26 15:51:20 +08:00
xsmq 73b7154e55 fix cmakelint error 2021-01-23 21:04:29 +08:00
changzherui 8b119950b3 modify export mindir type tupletensor 2021-01-22 16:54:52 +08:00
yuchaojie 611a9a3654 unify Conv's attr pad_list 2021-01-21 15:31:28 +08:00
yuchaojie 1932d87a26 update some op's attr name 2021-01-20 09:16:43 +08:00
yuchaojie b51b3a6764 update Pool's attr kernel_size, pad_mode 2021-01-18 12:10:20 +08:00
wangshuide2020 d574de4c42 add adapter of Asin, Asinh, Atan etc. operators for graphengine. 2021-01-09 16:34:40 +08:00
jiangzhenguang d3f42e7d6b add nll_loss operation. 2021-01-06 09:11:35 +08:00
mindspore-ci-bot 142f9c2d3e !9069 register fusion operator for lamb optimizer
From: @shibeiji
Reviewed-by: @c_34,@liangchenghui
Signed-off-by: @liangchenghui
2020-12-31 09:46:25 +08:00