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.
2222 Commits
28 Branches
21 MB
0 Download
 
 
 
 
 
 
Tree: ff2f8e720e
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'ff2f8e720e'
${ noResults }
graphengine/ge
zhengyuanhua 2328aa0ef4 aic error 3 years ago
..
analyzer analyzer log 3 years ago
client format log optimize 3 years ago
common aic error 3 years ago
engine_manager Change check_supported interface. 3 years ago
executor aic error 3 years ago
ge_local_engine fixed compiled issue for proto files 3 years ago
ge_runtime Replace rtLabelGotoEx by rtLabelSwitchByIndex 3 years ago
generator fix optional input bug 3 years ago
graph aic error 3 years ago
host_cpu_engine add netoupt alwways 3 years ago
host_kernels modified: ge/host_kernels/gather_v2_kernel.cc 3 years ago
hybrid aic error 3 years ago
inc update code from yellow zone 20201013 3 years ago
init update geloge and report errormessage 3 years ago
ir_build report errormsg 3 years ago
model multi-thread online infer 3 years ago
offline ge code of fuzz build 3 years ago
omm report errormsg 3 years ago
opskernel_manager report errormsg 3 years ago
plugin/engine !1261 aicpu算子优先级提升 3 years ago
proto modified insert_op.proto 3 years ago
session report errormsg 3 years ago
single_op !1468 fix repeat rtmalloc device mem 3 years ago
stub python3,7 modify python3 3 years ago
CMakeLists.txt aic error 3 years ago
README.md update cmake 4 years ago
ge_inference.mk ge code of fuzz build 3 years ago
ge_runner.mk ge code of fuzz build 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