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.
3167 Commits
28 Branches
21 MB
0 Download
 
 
 
 
 
 
Tree: b56450bde6
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'b56450bde6'
${ noResults }
graphengine/inc/framework/common
yanghaoran 19b430b275 upgrade Ascend software package 8 Sep 22 2 years ago
..
debug upgrade Ascend software package 8 Sep 22 2 years ago
helper upgrade Ascend software package 8 Sep 22 2 years ago
op upgarde Ascend software package 1 Sep 22 2 years ago
profiling upgrade Ascend software package 8 Sep 22 2 years ago
fmk_error_codes.h upgarde Ascend software package 1 Sep 22 2 years ago
fmk_types.h fix missing hearders 2 years ago
ge_compiler_options.h fix missing hearders 2 years ago
ge_format_util.h upgrade Ascend package 2 Dec 21 2 years ago
ge_inner_error_codes.h upgrade Ascend package 11 Feb 22 2 years ago
ge_types.h upgrade Ascend software package 8 Sep 22 2 years ago
ge_visibility.h upgrade Ascend software package 10 Aug 22 2 years ago
l2_cache_optimize.h upgrade Ascend package 11 Feb 22 2 years ago
op_types.h update Ascend package 09 Jun 22 2 years ago
profiling_definitions.h upgrade Ascend software package 8 Sep 22 2 years ago
runtime_tensor_desc.h upgrade Ascend package 18 May 22 2 years ago
scope_guard.h upgrade Ascend package 24 Feb 22 2 years ago
string_util.h upgarde Ascend software package 1 Sep 22 2 years ago
taskdown_common.h upgrade Ascend software package 8 Sep 22 2 years ago
types.h upgarde Ascend software package 1 Sep 22 2 years ago
util.h upgrade Ascend software package 8 Sep 22 2 years ago
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