Commit Graph

87 Commits

Author SHA1 Message Date
xulei 7a7927b537 update mindspore install scripts 2022-04-08 15:48:05 +08:00
xulei 0703b9d134 use conda-forge channel in conda create 2022-04-07 09:47:09 +08:00
i-robot 2cf884634f
!32385 update owners
Merge pull request !32385 from kisnwang/modify-owners
2022-04-02 02:11:53 +00:00
kswang 296fb69b54 update owners 2022-04-01 15:25:54 +08:00
xulei b80e55e22e install te topi packages in ascend install scripts 2022-04-01 11:14:19 +08:00
xulei cefb1b9ac7 [scripts] check mindspore version 2022-03-31 11:10:02 +08:00
i-robot 5532fab42c
!32153 realize scale out rollback
Merge pull request !32153 from tan-wei-cheng-3260/develop-twc-master
2022-03-30 19:17:36 +00:00
xulei 5cd0495eb0 fix pip auto install scripts 2022-03-30 09:27:27 +08:00
i-robot 893a34a3b3
!32149 fix mindspore install version in conda and pip scripts
Merge pull request !32149 from xulei/scripts
2022-03-29 03:34:32 +00:00
twc 2f068a4218 realize scale out rollback 2022-03-29 10:53:52 +08:00
xulei 83ccc1026d fix mindspore install version in conda and pip scripts 2022-03-29 10:16:56 +08:00
zhengyuanhua d89d8fd3f7 add 310 arm package 2022-03-28 19:38:28 +08:00
yanghaoran 8295364de0 modify dockerfile 2022-03-25 14:40:37 +08:00
xulei 1a7370dfb1 update auto install scripts 2022-03-24 10:45:34 +08:00
xulei 38a1ed4f73 add auto install scripts 2022-02-24 11:23:31 +08:00
Liu_Xuu f3bdedb432 [MSLITE] add release package for tensorrt 0209_01 2022-02-09 17:05:20 +08:00
yanghaoran 7e24a2bf51 remove deprecated script 2022-01-25 10:12:44 +08:00
zhengjun10 c15307dab2 support rm minddata 2022-01-22 14:45:19 +08:00
i-robot 6eae53eb34
!29057 fix ISSUE I4QCJM
Merge pull request !29057 from tan-wei-cheng-3260/develop-fix
2022-01-18 01:12:26 +00:00
twc 5d40613d52 fix ISSUE I4QCJM 2022-01-17 14:21:03 +08:00
i-robot 3a4744ff27
!28995 unify infer and train compile binary
Merge pull request !28995 from xiaoyao/master
2022-01-17 04:43:56 +00:00
i-robot 4c81d815e7
!29135 add ascend release pkg
Merge pull request !29135 from zhengyuanhua/br1
2022-01-15 09:17:23 +00:00
yanghaoran a5e8fa59a8 update dockerfile for 1.6.0 2022-01-15 15:48:30 +08:00
zhengyuanhua a2ae2a4598 add ascend release pkg 2022-01-15 15:38:20 +08:00
xiao_yao1994 54fc038f0a unify infer and train compile binary 2022-01-13 17:04:34 +08:00
i-robot 85a0216154
!28787 fix fl pclint
Merge pull request !28787 from tan-wei-cheng-3260/develop-fix
2022-01-12 02:01:21 +00:00
twc a48722c078 fix pclint 2022-01-11 14:08:22 +08:00
xiao_yao1994 0843910260 unify ge and vm backend 2022-01-11 11:53:58 +08:00
yanghaoran 762dc0b08a a conda script for euleros environments to prepare MindSpore runtime environment 2022-01-07 15:56:32 +08:00
i-robot f097e72be5
!28539 [MS][LITE]fix release pack
Merge pull request !28539 from gongdaguo/fix_release_pack
2022-01-06 01:22:28 +00:00
Bowen Ma 9b9d688c4c update auto install scripts 2022-01-05 15:19:28 +08:00
yanghaoran 0161b10f64 a script for euleros environments to prepare MindSpore runtime environment 2022-01-05 10:17:16 +08:00
gongdaguo 294718042f fix release package 2022-01-04 17:11:11 +08:00
Bowen Ma 7b1b133d1f update ubuntu + gpu with pip/conda install script. 2021-12-29 18:22:14 +08:00
Bowen Ma b94bdc8fdd update cuda with pip install script on ubuntu 18.04 2021-12-28 15:37:13 +08:00
Bowen Ma 3d1f4e6c70 update auto install scripts for pip/conda under ubuntu 2021-12-27 10:09:49 +08:00
gongdaguo 0e2d1d9749 fix package 2021-12-21 09:26:17 +08:00
zengzitao 6f4aebbe11 adapt ms_cpu ci for graphkernel 2021-12-09 19:44:09 +08:00
twc 649376a243 add fl restful tool to script 2021-12-06 17:38:38 +08:00
He Wei 41dcac9c49 Replace std::unordered_map/set with robin-hood-hashing
Robin-hood-hashing (https://github.com/martinus/robin-hood-hashing)
is considered faster then std::unordered_map/set,
so we use it to improve mindspore performance.

1. robin_hood head file in `third_party/robin_hood/include`;
2. In `utils/hash_map.h` and `utils/hash_set.h`, we define:
 - mindspore::HashMap as an alias of robin_hood::unordered_map;
 - mindspore::HashSet as an alias of robin_hood::unordered_set;
3. Replace:
 - `#include <unordered_map>` --> `#include "utils/hash_map.h"`;
 - `#include <unordered_set>` --> `#include "utils/hash_set.h"`;
 - `std::unordered_map` --> `mindspore::HashMap`;
 - `std::unordered_set` --> `mindspore::HashSet`;
 - `map.insert(std::pair(key, value))` --> `map.emplace(key, value)`;
 - `[] (const std::pair<K, V> &p) {..} ` --> `[] (const auto &p) {..} `;
4. Fix issues found by switch to robin_hood:
 - AnfNodeConfig hash and equal;
 - Fix a bug in `Slice::operator==()`;
 - Fix a bug in `CNode::HasPrimalAttr()`;
 - Fix map.erase() usage bugs: `map.erase(iter++)` --> `iter = map.erase(iter)`;
 - Fix some iterator invalidated problem;
5. Some std::unordered_map/set can not replace by robin_hood:
 - As parameter of functions that exposed to python by pybind11;
 - Use bad hash that cause robin_hood::map over_flow, such as AbstractBasePtrListHasher;
6. Update cpp unit tests;
7. Add build option '-F' to enable robin_hood, default on.
2021-11-24 10:47:40 +08:00
sunsuodong 1615edc359 add c api 2021-11-23 05:11:45 -08:00
tanghuikang f0995e7899 Add Exception log when AscendMemoryPool::AllocTensorMem fail 2021-11-16 09:28:28 +08:00
hwjiaorui d6b2a34a69 MindRT Ascend Devcie Context 2021-11-09 17:23:57 +08:00
He Wei fd8bbb4d96 Optimize isa() and type id related code.
1. Let 'isa<T>()' calls 'IsSameTypeId()' when T is final,
   and we add 'final' mark to some final classes derived from Base;

2. Generate type id at compile time using constexpr hash function,
   remove TypeIdManager and related code;

3. Add a tool script 'scripts/check_tid.sh' to check tid uniqueness;

4. Improve type search in 'proto_exporter.cc' and 'convert_utils_py.cc';

5. Add some missed MS_DECLARE_PARENT;

6. Ensure calls of 'parent_t::IsFromTypeId()' are all inlined.
2021-11-03 14:14:44 +08:00
yanghaoran d00fc2971b fix docker script due to pillow update 2021-10-20 11:22:35 +08:00
yanghaoran e9605ffefb add docker & conda scripts for 1.5.0 2021-10-13 10:43:43 +08:00
yanghaoran 6b07b95e81 fix ascend conda errors 2021-10-12 12:04:45 +08:00
i-robot 06b0beced7 !22731 GraphKernel supports CPU
Merge pull request !22731 from DeshiChen/0901_graphkernel_cpu
2021-09-30 09:36:35 +00:00
yanghaoran dde579e3eb fix conda problems 2021-09-30 14:26:00 +08:00
dayschan 32ecd8ee79 GraphKernel supports CPU
only Linux system is supported now.

change the default value of `ENABLE_AKG` to off, and controlled by option `-K`.
the `ENABLE_AKG` is auto enabled when `ENABLE_GPU` or `ENABLE_D` is on.
since now, we can use `ENABLE_AKG` to control the compilation of graphkernel
and akg codes.

fix usage description for option "-K", it should be "[-K on|off]".

LLVM is required by akg for cpu kernels, so AKG for cpu is default disabled now.
2021-09-29 23:34:16 +08:00