hummingbird
/
graphengine
Not watched
2
0
Fork 0
Code
Releases 13 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
2984 Commits
28 Branches
21 MB
0 Download
 
 
 
 
 
 
Tree: 4be8820566
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '4be8820566'
${ noResults }
graphengine/ge
wqtshg 4be8820566 delete compiling macros 3 years ago
..
analyzer Normalize include files... 3 years ago
client modify ge common so 3 years ago
common fix 3 years ago
engine_manager Add aicore_task_compiler.cc to executor 3 years ago
executor modify ge common so 3 years ago
ge_local_engine Fix multi control from one node 3 years ago
ge_opt_info fix 3 years ago
ge_runtime !1892 fix sc 3 years ago
generator modify ge common so 3 years ago
graph !1997 assign graph memory max size and variable memory max size adaptively 3 years ago
host_kernels modify ge common so 3 years ago
hybrid !1999 fix coverity 3 years ago
inc Normalize include files... 3 years ago
init modify ge common so 3 years ago
ir_build modify ge common so 3 years ago
offline delete compiling macros 3 years ago
opskernel_manager Add aicore_task_compiler.cc to executor 3 years ago
plugin/engine Normalize include files... 3 years ago
session modify ge common so 3 years ago
single_op !1975 single op of AICPU support null tensor 3 years ago
stub python3,7 modify python3 3 years ago
CMakeLists.txt delete compiling macros 3 years ago
README.md update cmake 4 years ago
ge_inference.mk modify ge common so 3 years ago
ge_runner.mk modify ge common so 3 years ago
module.mk update ge folder 4 years ago

README.md

Delete task

Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.

cancel
confirm

图引擎模块(GE)是MindSpore的一个子模块，其代码由C++实现，位于前端模块ME和底层硬件之间，起到承接作用。图引擎模块以ME下发的图作为输入，然后进行一系列的深度图优化操作，最后输出一张可以在底层硬件上高效运行的图。GE针对昇腾AI处理器的硬件结构特点，做了特定的优化工作，以此来充分发挥出昇腾AI处理器的强大算力。在进行模型训练/推理时，GE会被自动调用而用户并不感知。GE主要由GE API和GE Core两部分组成，详细的架构图如下所示

C++ Text Makefile Shell CMake other