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.
1250 Commits
28 Branches
21 MB
0 Download
 
 
 
 
 
 
Tree: 2b183e25d5
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '2b183e25d5'
${ noResults }
graphengine/tests/ut/ge/graph/load
mindspore-ci-bot 09e5439215 !880 Eliminate variable_op_list_ 3 years ago
..
data_dumper_unittest.cc ge ut 4 years ago
davinci_model_unittest.cc !880 Eliminate variable_op_list_ 3 years ago
end_graph_task_unittest.cc Update GraphEngine to synchronize with latest Ascend driver software suite 4 May 2020 4 years ago
hccl_task_info_unittest.cc Modify UT 3 years ago
kernel_ex_task_info_unittest.cc Eliminate variable_op_list_ 3 years ago
kernel_task_info_unittest.cc Eliminate data_op_list_ 3 years ago
memcpy_addr_async_task_info_unittest.cc Modify UT 3 years ago
memcpy_async_task_info_unittest.cc Modify UT 3 years ago
model_helper_unittest.cc display model info 3 years ago
new_model_manager_data_inputer_unittest.cc initial version 4 years ago
new_model_manager_davinci_model_unittest.cc Memory optimization during model loading 3 years ago
new_model_manager_event_manager_unittest.cc initial version 4 years ago
new_model_manager_model_manager_aicpu_unittest.cc Free memory before return 3 years ago
new_model_manager_model_manager_unittest.cc ge ut 4 4 years ago
new_model_manager_task_build_unittest.cc initial version 4 years ago
new_op_test_utils.h Fix code check 3 years ago
output_net_output_unittest.cc ge ut 5 4 years ago
tbe_handle_store_unittest.cc initial version 4 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