sunsuodong
20311c30b3
fix vs build
2022-02-21 00:39:19 -08:00
i-robot
e4a86db2e8
!30213 [MSLITE][DEVELOP] fix VS bug
...
Merge pull request !30213 from yangruoqi713/master
2022-02-21 08:19:25 +00:00
zengxianglong
5925a53600
split quant task
2022-02-21 16:02:08 +08:00
i-robot
a5242d33e1
!30181 support TensorShape ops
...
Merge pull request !30181 from lianliguang/change-dynshape-to-tensorshape
2022-02-21 07:36:40 +00:00
lz
1056328d94
micro refactor
2022-02-21 15:36:18 +08:00
i-robot
52574812b3
!30279 Rename matrix_triangular_solve
...
Merge pull request !30279 from zhujingxuan/master
2022-02-21 07:28:23 +00:00
i-robot
072fc770d0
!29963 dataset: add vision api document
...
Merge pull request !29963 from ms_yan/code_docs_cn_api
2022-02-21 07:23:06 +00:00
wenbean
638ba5b4d1
Fix ops matrix_diag_part different version interplay infershap between ascend and gpu/cpu
2022-02-21 15:05:09 +08:00
ms_yan
90f665726a
add part of vision api doc
2022-02-21 14:48:53 +08:00
i-robot
56a90003c4
!30296 fix DynamicMatmul4x4x16AIWI c++ && copy bias bug and float_mode bug
...
Merge pull request !30296 from yeyunpeng2020/dynamic_quant
2022-02-21 06:35:22 +00:00
i-robot
6c301b6e1f
!30194 modify initializer for master
...
Merge pull request !30194 from lilei/modify_initializer_for_master
2022-02-21 06:23:50 +00:00
i-robot
d4298e9136
!29820 moe_topk routing
...
Merge pull request !29820 from wangshengnan123/moe_topk_routing
2022-02-21 06:21:55 +00:00
i-robot
7386612515
!29820 moe_topk routing
...
Merge pull request !29820 from wangshengnan123/moe_topk_routing
2022-02-21 06:21:55 +00:00
i-robot
916328a221
!30206 optimize syncbatchnorm format problem
...
Merge pull request !30206 from huoxinyou/syncbatchnorm
2022-02-21 06:17:43 +00:00
i-robot
fda6fce473
!30242 [ME] Remove warning log when Parameter is not set name.
...
Merge pull request !30242 from Margaret_wangrui/remove_warning
2022-02-21 06:14:20 +00:00
yujianfeng
7e941cfef5
Fix the bprop mindirs generated with sequence ref_attr_name
2022-02-21 11:36:23 +08:00
Zhang Qinghua
1357f1fe7a
Transform if tail call to parallel call.
2022-02-21 11:28:06 +08:00
ZPaC
247ca73d1a
Optimize code in graph splitter for rpc nodes.
2022-02-21 11:27:02 +08:00
i-robot
97bc0ca366
!30085 code_docs_add_transformer_api_CN
...
Merge pull request !30085 from lichen/code_docs_add_transformer_api
2022-02-21 03:11:35 +00:00
i-robot
8ad4b6c3ec
!30047 fix fuzz and clean codex
...
Merge pull request !30047 from qinzheng/fix_fuzz
2022-02-21 03:10:26 +00:00
wangshengnan12@huawei.com
3a27ee3533
code_docs_add_transformer_api
2022-02-21 10:41:16 +08:00
yao_yf
4b79d4c425
auto parallel adasum uts and checks
2022-02-21 10:41:09 +08:00
i-robot
6ebee5ff2c
!30195 Update boost cn api.
...
Merge pull request !30195 from linqingke/code_docs_api
2022-02-21 02:39:32 +00:00
lilei
8e2b5515a4
modify initializer for master
2022-02-21 10:36:15 +08:00
i-robot
6a22e23855
!30210 add cho_factor && cho_solve grad st
...
Merge pull request !30210 from zhuzhongrui/pub_master3
2022-02-21 01:41:55 +00:00
chenping
531dd01b3e
310 codex clean
2022-02-21 09:38:44 +08:00
hezhenhao1
73ad595fa3
Refactor value and type checking for mindspore.scipy module.
2022-02-21 09:35:59 +08:00
He Wei
946e83dc0e
Use steady_clock() instead of gettimeofday() for profiling
...
1. std::chrono::steady_clock() is monotonic clock which is better for profiling.
2. Using c++ standard library is more portable than using gettimeofday().
2022-02-21 09:34:18 +08:00
lianliguang
5c8af66e53
support TensorShape ops to instead of DynamicShape ops
2022-02-21 09:33:32 +08:00
i-robot
33800c7f31
!29471 [MS][LITE][develop] code review
...
Merge pull request !29471 from sunsuodong/code_review_master
2022-02-21 01:20:22 +00:00
yeyunpeng2020
3e8ac68bc6
fix DynamicMatmul4x4x16AIWI c++ && copy bias bug and float_mode bug
2022-02-21 09:19:59 +08:00
xuanyue
0f72ace8cf
optimize gather op
2022-02-21 09:19:45 +08:00
xuanyue
eb3ad986d1
optimize matmul's logic
2022-02-21 09:15:53 +08:00
greatpanc
1ff5cf0e99
remove some big gpu model
2022-02-20 20:15:43 +08:00
zhujingxuan
911e4660ce
Rename matrix_triangular_solve to be compatible with tensorflow's MatrixTriangularSolve OP
2022-02-20 17:12:10 +08:00
i-robot
d0d932c94f
!30255 [master]Add testcases about side effect & Execute in vm of pipeline + control + mutil-target
...
Merge pull request !30255 from chenfei_mindspore/add-side-effect-case
2022-02-19 10:42:13 +00:00
i-robot
c2b7e15b5b
!30264 fix activation grad infer
...
Merge pull request !30264 from zhaodezan/master
2022-02-19 10:31:47 +00:00
i-robot
996e7bffaa
!30155 AllReduce input and output size aligned by 512 in ascend device context
...
Merge pull request !30155 from laiyongqiang/master_allreduce
2022-02-19 09:56:54 +00:00
i-robot
d4e765fb69
!30254 modify landscape Chinese api
...
Merge pull request !30254 from Songyuanwei/api
2022-02-19 08:59:48 +00:00
i-robot
24609fd15f
!30153 bugfix; fix 310 compile error when multi thread
...
Merge pull request !30153 from zhengyuanhua/br1
2022-02-19 08:18:58 +00:00
i-robot
2be5bf62dd
!30230 710 models: crnn faster_rcnn bugfix
...
Merge pull request !30230 from chenping/runtime
2022-02-19 08:18:57 +00:00
i-robot
3215493944
!30229 Commit code in 1.6.
...
Merge pull request !30229 from gaoyong10/runtime_second12
2022-02-19 08:14:41 +00:00
i-robot
fe07b896ed
!30068 add vision chinese api
...
Merge pull request !30068 from 刘勇琪/master
2022-02-19 07:55:17 +00:00
chenfei
1652592529
add side effect cases
2022-02-19 15:38:48 +08:00
zhaodezan
e746d250c6
fix activation grad infer
2022-02-19 15:35:20 +08:00
i-robot
0acbf5d67f
!30029 remove mutil_stream for kernel by kernel
...
Merge pull request !30029 from baihuawei/seq_kernel_by_kernel
2022-02-19 06:59:10 +00:00
i-robot
e4d107be80
!29896 Construct cache dir with rank_id directly
...
Merge pull request !29896 from laiyongqiang/cache_dir
2022-02-19 06:41:14 +00:00
yangruoqi713
54e6a75875
[MSLITE][DEVELOP] fix VS bug
2022-02-19 10:59:47 +08:00
xuanyue
106a41eee1
optimize reduce op
2022-02-19 10:36:28 +08:00
i-robot
288bbd832a
!30207 [MS][LITE]sync fuzz
...
Merge pull request !30207 from mengyuanli/bugfix
2022-02-19 02:09:27 +00:00