vilc
/
mindspore2022
forked from huawei/mindspore2022
0
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
42,205 Commits 18 Branches 22 Tags 3.6 GiB
Commit Graph

687 Commits

Author SHA1 Message Date
i-robot d12446c538
!32729 [MSLITE][CPU] AVX512/256/SSE/NENO Advanced packaging, and DivFusion Op and RealDiv Op (total 7) Refactoring and optimization 
Merge pull request !32729 from Greatpan/avx512_realdiv
 2022-04-09 12:57:13 +00:00
i-robot 57be55d99c
!32745 [MSLITE][CPU] cpu platform instruction support check, add ModelParallelRunner and log 
Merge pull request !32745 from Greatpan/x86cpu_info
 2022-04-09 12:35:05 +00:00
i-robot 4a646193d4
!32760 add selu kernerl cpu implement 
Merge pull request !32760 from zhuzhongrui/pub_master2
 2022-04-09 09:39:20 +00:00
greatpanc b7e1fd30d6 avx512 x86cpu info 2022-04-09 17:10:20 +08:00
i-robot 8f1bd1c1bb
!32299 [MS][LITE]update vs infer register and all_gather 
Merge pull request !32299 from luoyuan/update-vs-infer-register
 2022-04-09 08:46:05 +00:00
i-robot b3bf1a0fcb
!32757 [MS][LITE] fix windows 32bit build 
Merge pull request !32757 from sunsuodong/fix_win32_build
 2022-04-09 08:43:02 +00:00
i-robot 3f7879a2e3
!32763 add lerp kernerl base class cpu implement to DeprecatedNativeCpuKernelMod 
Merge pull request !32763 from zhuzhongrui/pub_master
 2022-04-09 08:13:54 +00:00
i-robot c6884369e0
!32754 fix cuda10 cusparse library compile issue 
Merge pull request !32754 from chenweifeng/fix-cuda10-cusparse-library-link-issue
 2022-04-09 07:44:35 +00:00
z00512249 3470565532 add selu kernerl cpu implement 2022-04-09 14:57:27 +08:00
z00512249 6b8d7c3157 add lerp kernerl base class cpu implement to DeprecatedNativeCpuKernelMod 2022-04-09 14:40:25 +08:00
i-robot 1a2235358a
!32626 Refactor KernelMod and its sub class. 
Merge pull request !32626 from TronZhang/make_kernel_mod_better
 2022-04-09 05:36:57 +00:00
i-robot fbf55f9cef
!32735 add lerp kernerl cpu implement 
Merge pull request !32735 from zhuzhongrui/pub_master
 2022-04-09 03:07:02 +00:00
sunsuodong 4f70145b43 fix win32 build 2022-04-08 19:09:26 -07:00
z00512249 e50a7bfe41 add lerp kernerl cpu implement 2022-04-09 09:57:10 +08:00
wilfChen 21ec487868 fix cuda10 cusparse library link issue 2022-04-09 09:42:41 +08:00
i-robot 9da4f06075
!32690 [MSLITE][CPU] cpu platform instruction support check 
Merge pull request !32690 from Greatpan/simd_instruct_check
 2022-04-08 09:16:59 +00:00
greatpanc d06eaa595d avx512 divfusion/realdiv op 2022-04-08 16:42:45 +08:00
i-robot 51e35f7faf
!32082 GraphKernel for Mindpsore-Lite 
Merge pull request !32082 from DeshiChen/0328_graph_kernel_lite
 2022-04-08 08:29:29 +00:00
tronzhang 27893cde7c change kernelmod interface and gpu/cpu kernel mod 2022-04-08 16:28:02 +08:00
i-robot 9589f88cd2
!32714 Add cpu implementation of IsClose operator. 
Merge pull request !32714 from hezhenhao1/add_isclose
 2022-04-08 07:27:07 +00:00
hezhenhao1 54d290a813 Add cpu implementation of IsClose operator. 2022-04-08 12:44:48 +08:00
i-robot 9e695d0d9c
!32587 [MS][LITE][ToD] VAE model bit exact 
Merge pull request !32587 from Nizzan/export_nizzan
 2022-04-08 03:30:23 +00:00
i-robot dc1304afee
!32508 [DynamicShape][GPU]fix a bug that the dropout op continued to create new Generators under dynamic shape 
Merge pull request !32508 from hanhuifeng/dyn_dropout
 2022-04-08 02:31:22 +00:00
i-robot 6b082e6ebe
!32416 RL add TensorsQueue 
Merge pull request !32416 from VectorSL/grads_queue_for_merge
 2022-04-08 01:32:28 +00:00
i-robot fa203ca52b
!32483 Refactor testcases for scatter_nd 
Merge pull request !32483 from zhujingxuan/scatter_test
 2022-04-08 01:26:50 +00:00
i-robot efd13aa953
!32551 Change the types of input and output of tcp message handler 
Merge pull request !32551 from chengang/add_retrieve_for_cgn
 2022-04-08 01:25:14 +00:00
i-robot f2f93f9f08
!32599 Add cusparse library 
Merge pull request !32599 from chenweifeng/sparse-matrix-add-gpu
 2022-04-08 01:23:23 +00:00
greatpanc 81a2c79c25 x86 simd instruction checkout 2022-04-08 09:09:39 +08:00
nizzan 0d0fd7e4c2 Change file for bit exact in VAE model 2022-04-07 18:47:50 +03:00
i-robot 9c0e3ca887
!32627 [MSLITE][CPU] AVX512/256/SSE/NENO Advanced packaging, and BiasAdd Op Refactoring and optimization 
Merge pull request !32627 from Greatpan/avx512_bias_add
 2022-04-07 11:30:09 +00:00
i-robot 8f2b833081
!32424 Add CPU implementation of ScatterNdMul operator. 
Merge pull request !32424 from hezhenhao1/add_scatter_mul
 2022-04-07 09:16:59 +00:00
i-robot 4adf5064fc
!32414 RL add gpu kernel: batch read write 
Merge pull request !32414 from VectorSL/read_write_op
 2022-04-07 02:16:34 +00:00
hanhuifeng2020 bff52aac5a [DynamicShape][GPU]fix a bug that the dropout op continued to create new Generators under dynamic shape 2022-04-07 10:09:33 +08:00
i-robot 76f0e16327
!32605 Update output size of reshape op when dynamic shape 
Merge pull request !32605 from zhengzuohe/reshape_output_update
 2022-04-07 02:05:44 +00:00
VectorSL 0f9148ea5b add tensorsqueue gpu kernel 2022-04-07 10:02:54 +08:00
i-robot 6f4976d619
!32498 enable multi thread when run maxpool forward to obtain index 
Merge pull request !32498 from zhangbuxue/use_multi_thread_when_run_maxpool_forward_to_obtain_index
 2022-04-07 01:42:21 +00:00
zhujingxuan 5303f20e89 Refactor testcases for scatter_nd. 2022-04-07 09:16:58 +08:00
hezhenhao1 33beadec59 Add CPU implementation of ScatterNdMul operator. 2022-04-07 09:13:00 +08:00
wilfChen 825ee3d57f add cusparse library 2022-04-07 09:09:34 +08:00
i-robot a6d8b185e7
!32616 set value node fracz group 
Merge pull request !32616 from jjfeing/master
 2022-04-07 01:07:59 +00:00
i-robot 7cd2dcbdfd
!32591 optimize the cpu kernel select error info 
Merge pull request !32591 from limingqi107/bug_fix3
 2022-04-06 18:35:17 +00:00
jjfeing f3a03b5775 only set value node group with mul 2022-04-06 20:16:40 +08:00
zhengzuohe 738f019940 update output size of reshape op when dynamic shape 2022-04-06 18:36:59 +08:00
limingqi107 056afa975c optimize the cpu kernel select error info 2022-04-06 17:31:55 +08:00
z00512249 821f514e10 add TensorScatterMul and opt TensorScatterBianryOp's old codes 2022-04-06 15:56:03 +08:00
i-robot d3e5b08302
!32550 add precision flag for op which raise or reduce precision 
Merge pull request !32550 from liubuyu/code_dex
 2022-04-06 07:05:05 +00:00
i-robot 081957a829
!32545 [lite]support prefetch to optimize matmul-arm 
Merge pull request !32545 from 徐安越/primitive
 2022-04-06 06:46:35 +00:00
buxue deabea9211 enable multi thread when run maxpool forward to obtain index 2022-04-06 13:06:07 +08:00
i-robot 9e0d533ce4
!32495 add conv2dbackprobfilter + mul pass 
Merge pull request !32495 from jjfeing/master
 2022-04-06 03:10:57 +00:00
dayschan 2026faa695 GraphKernel for MS-lite 
1. move the lite_adapter files to lite directory.
created a directory "graph_kernel" in "mindspore/lite/tools".
we use 3 directory "converter", "runtime" and "common" to hold the different files.

2. modify lite makefiles to support GraphKernel.
a) define a micro "MSLITE_ENABLE_GRAPH_KERNEL" when graphkernel is enabled by building option "-K on".
b) converter's makefile: build the GraphKernel's core code and graph_kernel adapter for converter,
includes some backend's code of mindspore.
c) lite/src/CMakeLists: build graph_kernel adapter for benchmark

3. move the graph_kernel_pass_manager.cc from "graph_kernel/adapter" to "graph_kernel/core".
the original two graph_kernel_pass_manager files are similar, so we unify them.
 2022-04-06 11:02:12 +08:00