Commit Graph

224 Commits

Author SHA1 Message Date
i-robot 20f3a911f1
!34094 Remove the context option: 'enable_sparse'.
Merge pull request !34094 from 张清华/opt
2022-05-20 00:52:05 +00:00
Zhang Qinghua d7762f1c8d Add bprop_return_sparse flag for sparse bprop primitive and remove context option: 'enable_sparse'. 2022-05-19 17:31:57 +08:00
fujianzhao 7fcb75beaa [feat] [assistant] [ops] [I51VS2] Add new operator nn.AdaptiveMaxPool1d 2022-05-18 15:50:44 +08:00
fujianzhao cb56dd895d [feat] [assistant] [ops] [I51VS1] Add new operator nn.AdaptiveAvgPool1d 2022-05-16 16:19:24 +08:00
i-robot c07819796c
!34221 [feat] [assistant] [ops] [I51VSC] Add new operator nn.HuberLoss
Merge pull request !34221 from fujianzhao/huberloss
2022-05-13 02:57:19 +00:00
i-robot bfd377ca3c
!34125 add nn.ConstantPad1d nn.ConstantPad2d nn.ConstantPad3d and nn.ZeroPad2d
Merge pull request !34125 from ZhidanLiu/master
2022-05-13 01:53:23 +00:00
ZhidanLiu fdd6f7c759 add nn.ConstantPad1d nn.ConstantPad2d nn.ConstantPad3d and nn.ZeroPad2d 2022-05-12 10:34:52 +08:00
fujianzhao 4fd3433d23 [feat] [assistant] [ops] [I51VSC] Add new operator nn.HuberLoss 2022-05-11 15:39:02 +08:00
fujianzhao 5bb57674da [feat] [assistant] [ops] [I51VSB] Add new operator nn.Hardtanh 2022-05-11 14:23:11 +08:00
liqiliang 0fabbc4ccb Add a nn cell for softsign. 2022-05-10 17:09:33 +08:00
He Wei 8b0d4e7751 Refactor optimizers to support flatten parameters 2022-04-20 20:23:31 +08:00
He Wei 0d12199928 Enable flatten_gradients for Adam optimizer 2022-04-15 19:49:57 +08:00
He Wei 723543eb5d Modify Optimizer to support flatten parameters
1. Provide method to create Parameter from Tensor without copy;
2. Adapt optimizer.py to support flattened parameters;
3. Add unit test cases for parameter and optimizer.
2022-04-07 14:11:32 +08:00
xiao_yao1994 94df0869ad uni_backend: add ut ge test 2022-03-10 16:08:36 +08:00
Margaret_wangrui 462c38813d [ME] Add parameter name check. 2022-02-14 10:07:43 +08:00
liuyang_655 15a153a40d Modify error info 2022-01-06 01:28:53 -05:00
wanyiming ab65f4ac98 addut 2021-11-26 09:48:34 +08:00
chenzhuo d07f24462c add examples of grad, jvp and vjp 2021-11-16 14:15:26 +08:00
chenzhuo f36edec3ff add function jvp 2021-11-12 08:44:41 +08:00
Zhang Qinghua b43693be23 Fix fallback and cell kwargs issues. 2021-10-27 14:06:02 +08:00
Bingya Weng be20f00299 modify ut of probability 2021-10-14 19:33:34 +08:00
huanghui ba66c0d491 add security isolate for save_graphs 2021-09-14 10:24:12 +08:00
i-robot 76fac95591 !23069 Modify error content
Merge pull request !23069 from huchunmei/master
2021-09-11 03:50:52 +00:00
huchunmei c75773cf37 Modify error content 2021-09-10 15:21:33 +08:00
yanghaoran 0ced8775c4 upgrade Ascend software package 09 Sep 21 2021-09-09 20:30:22 +08:00
huangxinjing 6cea07f749 Add args check 2021-09-08 09:12:35 +08:00
yao_yf 82889ec56b fixed sparse attention 2021-08-31 16:11:56 +08:00
i-robot ff2e34b9f9 !22509 Add ut and st for Jvp, Vjp and Grad
Merge pull request !22509 from LiangZhibo/fix
2021-08-30 08:21:54 +00:00
Zhang Qinghua a137fa1d0b Optimize the Executors routines.
- Fix the key generating.
- Distinguish the executors.
2021-08-28 17:36:12 +08:00
l00591931 98fad9f08e Add st and ut for jvp vjp and grad 2021-08-28 17:34:23 +08:00
zhihenghu ce12c02343 Add Sparse Attention
adjut the file structure and name

Deleted extra information

Do some formatting work

Add test case and fix some document

fix imports
2021-08-28 11:50:58 +08:00
huangxinjing 62496d75f3 less the interface exposed 2021-08-25 21:22:40 +08:00
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
2021-08-23 01:46:38 +08:00
djc 4e6f7dc97d [feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset 2021-08-22 13:39:37 +08:00
huangxinjing d777742904 1. Move the class to mindspore.parallel, support activation sharding 2021-08-20 14:34:10 +08:00
huangxinjing 18044aff0f 1. Add docstring, elimitate attention mask, tuple append the deocoder return layer past
split attention inputs from two into three

Add input check and modify test case

Fix check error and add check to transformer inputs

add input check to the transformers
2021-08-16 21:08:48 +08:00
i-robot 2650aae9ba !21351 complex support
Merge pull request !21351 from zhouyaqiang0/complex_support
2021-08-12 09:39:15 +00:00
zhouyaqiang b81ba100b7 add complex number support 2021-08-12 10:36:43 +08:00
jiangzhenguang a3bb780ce4 optimize run memory 2021-08-10 20:45:25 +08:00
huangxinjing 615d1a179d Add transformer layer 2021-07-29 11:10:43 +08:00
huangbingjian ad8f73f32b set default context mode to GRAPH_MODE 2021-06-09 14:40:49 +08:00
huanghui 29f632613f optimize log 2021-06-04 12:32:44 +08:00
huanghui 5acf3c8ade [dfx]optimize exception info print 2021-05-31 14:57:29 +08:00
wangnan39@huawei.com 2761776699 add check for optimizer parameters 2021-05-17 22:23:03 +08:00
jiaqi cdb7330818 diceloss bug 2021-03-09 16:23:45 +08:00
Zhang Qinghua df866f7248 Add TopoSort Rhs First attribute for special CNode, such as Depend CNode with isolated nodes. 2021-03-06 12:39:36 +08:00
Zhang Qinghua 8b8c59f01e Optimize the compile performance in Parser, FG, Manager and Renormalize:
---
Remove the routine of handling isolated nodes in Renormalize.
Add isolated nodes from Parser&Resolver.
Modify isolated nodes handling in FG&Manager.
Optimize the renormalize routines.
Other optimizations.
2021-03-04 16:01:54 +08:00
Jiaqi 112b8488c5 add rmse loss 2021-02-25 17:03:07 +08:00
Jiaqi fd1a4726b7 add focal loss 2021-02-22 19:33:26 +08:00
Jiaqi 952e245000 add multclass diceloss 2021-02-19 15:43:30 +08:00