Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
as350144
Mace
提交
aba6b576
Mace
项目概览
as350144
/
Mace
与 Fork 源项目一致
Fork自
Xiaomi / Mace
通知
2
Star
1
Fork
1
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
0
列表
看板
标记
里程碑
合并请求
0
DevOps
流水线
流水线任务
计划
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
Mace
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
0
Issue
0
列表
看板
标记
里程碑
合并请求
0
合并请求
0
Pages
DevOps
DevOps
流水线
流水线任务
计划
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
流水线任务
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
提交
aba6b576
编写于
6月 26, 2018
作者:
L
liutuo
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
check docs
上级
47894517
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
13 addition
and
13 deletion
+13
-13
README.md
README.md
+9
-9
docs/development/contributing.md
docs/development/contributing.md
+1
-1
docs/getting_started/create_a_model_deployment.rst
docs/getting_started/create_a_model_deployment.rst
+3
-3
未找到文件。
README.md
浏览文件 @
aba6b576
...
@@ -10,31 +10,31 @@
...
@@ -10,31 +10,31 @@
[
中文
](
README_zh.md
)
[
中文
](
README_zh.md
)
**Mobile AI Compute Engine**
(or
**MACE**
for short) is a deep learning inference framework optimized for
**Mobile AI Compute Engine**
(or
**MACE**
for short) is a deep learning inference framework optimized for
mobile heterogeneous computing platforms. The design
is focused
on the following
mobile heterogeneous computing platforms. The design
focuses
on the following
targets:
targets:
*
Performance
*
Performance
*
The runtime is highly optimized with NEON, OpenCL and Hexagon, and
*
The runtime is highly optimized with NEON, OpenCL and Hexagon, and
[
Winograd algorithm
](
https://arxiv.org/abs/1509.09308
)
is introduced to
[
Winograd algorithm
](
https://arxiv.org/abs/1509.09308
)
is introduced to
speed up the convolution operations.
Except for the
inference speed, the
speed up the convolution operations.
Besides the fast
inference speed, the
initialization
speed is also intensively optimized
.
initialization
part is also intensively optimized to be faster
.
*
Power consumption
*
Power consumption
*
Chip dependent power options like big.LITTLE scheduling, Adreno GPU hints are
*
Chip dependent power options like big.LITTLE scheduling, Adreno GPU hints are
included as advanced API.
included as advanced API
s
.
*
Responsiveness
*
Responsiveness
*
UI responsiveness gu
rantee is sometimes obligatory when ru
ning a model.
*
UI responsiveness gu
arantee is sometimes obligatory when run
ning a model.
Mechanism like automatically breaking OpenCL kernel into small units is
Mechanism like automatically breaking OpenCL kernel into small units is
introduced to allow better preemption for the UI rendering task.
introduced to allow better preemption for the UI rendering task.
*
Memory usage and library footprint
*
Memory usage and library footprint
*
Graph level memory allocation optimization and buffer reuse
is
supported.
*
Graph level memory allocation optimization and buffer reuse
are
supported.
The core library tries to keep minium external dependencies to keep the
The core library tries to keep minium external dependencies to keep the
library footprint small.
library footprint small.
*
Model protection
*
Model protection
*
Model protection is
one
the highest priority feature from the beginning of
*
Model protection is the highest priority feature from the beginning of
the design. Various techniques are introduced like coverting models to C++
the design. Various techniques are introduced like coverting models to C++
code and literal obfuscations.
code and literal obfuscations.
*
Platform coverage
*
Platform coverage
*
A good coverage of recent Qualcomm, MediaTek, Pinecone and other ARM based
*
A good coverage of recent Qualcomm, MediaTek, Pinecone and other ARM based
chips. CPU runtime is also comp
ita
ble with most POSIX systems and
chips. CPU runtime is also comp
ati
ble with most POSIX systems and
archetectures with limited performance.
archetectures with limited performance.
## Getting Started
## Getting Started
...
@@ -44,7 +44,7 @@ targets:
...
@@ -44,7 +44,7 @@ targets:
## Performance
## Performance
[
MACE Model Zoo
](
https://github.com/XiaoMi/mace-models
)
contains
[
MACE Model Zoo
](
https://github.com/XiaoMi/mace-models
)
contains
several common neural networks
models and
built daily against a list of mobile
several common neural networks
and models which will be
built daily against a list of mobile
phones. The benchmark result can be found in the CI result page.
phones. The benchmark result can be found in the CI result page.
## Communication
## Communication
...
...
docs/development/contributing.md
浏览文件 @
aba6b576
...
@@ -4,7 +4,7 @@ Contributing guide
...
@@ -4,7 +4,7 @@ Contributing guide
License
License
-------
-------
The source file should contain
s
a license header. See the existing files
The source file should contain a license header. See the existing files
as the example.
as the example.
Python coding style
Python coding style
...
...
docs/getting_started/create_a_model_deployment.rst
浏览文件 @
aba6b576
...
@@ -6,15 +6,15 @@ file.
...
@@ -6,15 +6,15 @@ file.
One deployment file describes a case of model deployment,
One deployment file describes a case of model deployment,
each file will generate one static library (if more than one ABIs specified,
each file will generate one static library (if more than one ABIs specified,
there will be one static library for each). The deployment file can contain
s
there will be one static library for each). The deployment file can contain
one or more models, for example, a smart camera application may contain
s
face
one or more models, for example, a smart camera application may contain face
recognition, object recognition, and voice recognition models, which can be
recognition, object recognition, and voice recognition models, which can be
defined in one deployment file),
defined in one deployment file),
Example
Example
----------
----------
Here is an
deployment file example used by
Android demo application.
Here is an
example deployment file used by an
Android demo application.
TODO: change this example file to the demo deployment file
TODO: change this example file to the demo deployment file
(reuse the same file) and rename to a reasonable name.
(reuse the same file) and rename to a reasonable name.
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录