Skip to content

Commit a7e0b22

Browse files
committed
[UPD]update doc
1 parent fbd2851 commit a7e0b22

File tree

3 files changed

+21
-19
lines changed

3 files changed

+21
-19
lines changed

doc/cn/development/profiling.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -42,7 +42,7 @@
4242

4343
首次运行先利用快捷键`Command + Shift + K`对工程进行清理,再执行快捷键`Command + R`运行。在界面上点击Run按钮,界面会显示model目录下所有模型的CPU和GPU耗时情况。iPhone7真机运行结果如下图。
4444

45-
<div align=left ><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/development/resource/ios_benchmark_result.jpg" width = "75%" height = "75%"/>
45+
<div align=left ><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/development/resource/ios_benchmark_result.jpg" width = "50%" height = "50%"/>
4646

4747
PS:
4848

doc/cn/user/demo.md

Lines changed: 15 additions & 15 deletions
Original file line numberDiff line numberDiff line change
@@ -21,40 +21,40 @@
2121

2222
如下图点击TNNExamples工程,找到工程设置`Signing & Capabilities`,点击Team选项卡选择`Add an Account...`
2323

24-
<div align=left><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/user/resource/ios_add_account_demo.jpg">
24+
<div align=left ><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/user/resource/ios_add_account_demo.jpg" width = "75%" height = "75%"/>
2525

2626
在如下界面输入Apple ID账号和密码,添加完成后回到`Signing & Capabilities`界面,并在Team选项卡中选中添加的账号。如果没有Apple ID也可以通过`Create Apple ID`选项根据相关提示进行申请。
2727

2828
`PS:申请Apple ID无需付费,可以即时通过,通过后才可在真机上运行APP调试`
2929

30-
<div align=left><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/user/resource/ios_set_account.jpg">
30+
<div align=left ><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/user/resource/ios_set_account.jpg" width = "75%" height = "75%"/>
3131

3232
4. 真机运行
3333

3434
4.1 修改`Bundle Identitifier`
3535

3636
如图在现有`Bundle Identifier`后随机添加后缀(限数字和字母),避免个人账户遇到签名冲突。
3737

38-
<div align=left><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/user/resource/ios_set_bundleid_demo.jpg">
39-
38+
<div align=left ><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/user/resource/ios_set_bundleid_demo.jpg" width = "75%" height = "75%"/>
39+
4040
4.2 验证授权
41-
41+
4242
首次运行先利用快捷键`Command + Shift + K`对工程进行清理,再执行快捷键`Command + R`运行。如果是首次登陆Apple ID,Xcode会弹框报如下错误,需要在iOS设备上根据提示进行授权验证。一般来说手机上的授权路径为:设置 -> 通用 -> 描述文件与设备管理 -> Apple Development选项 -> 点击信任
43-
44-
<div align=left><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/user/resource/ios_verify_certificate_demo.jpg">
45-
43+
44+
<div align=left ><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/user/resource/ios_verify_certificate_demo.jpg" width = "75%" height = "75%"/>
45+
4646
4.3 运行结果
47-
47+
4848
首次运行先利用快捷键`Command + Shift + K`对工程进行清理,再执行快捷键`Command + R`运行。默认界面为人脸检测,可以点击右上角编辑按钮切换图像分类等不同功能。
49-
49+
5050
PS:
51-
51+
5252
a) 由于GPU和CPU加速原理不同,具体模型的GPU性能不一定比CPU高,与具体机型、模型结构以及工程实现有关。欢迎大家参与到TNN开发中,共同进步。
53-
53+
5454
b) TNNSDKSample.h中的宏TNN_SDK_USE_NCNN_MODEL默认为0,运行TNN模型,可以设置为1来运行ncnn模型。
55-
55+
5656
c) 如遇到`Unable to install...`错误提示,请在真机设备上删除已有的TNNExamples,重新运行安装。
57-
57+
5858
d) 真机运行时,如果遇到CodeSign错误`Command CodeSign failed with a nonzero exit code`,可参看issue20 `iOS Demo运行步骤说明`
5959

6060
### Demo运行效果
@@ -99,7 +99,7 @@ b) TNNSDKSample.h中的宏TNN_SDK_USE_NCNN_MODEL默认为0,运行TNN模型,
9999

100100
### 运行效果
101101
1. 人脸检测-图片
102-
102+
103103
模型来源:https://github.com/Linzaer/Ultra-Light-Fast-Generic-Face-Detector-1MB
104104

105105
效果示例:华为P30, ARM 单线程 32.2359ms

doc/cn/user/visual.md

Lines changed: 5 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,10 @@
1-
<div align=center ><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/user/resource/tnn-netron.jpg" width = "75%" height = "75%"/>
21

32

3+
<div align=left ><img src="https://raw.githubusercontent.com/darrenyao87/tnn-models/master/doc/cn/user/resource/tnn-netron.jpg" width = "75%" height = "75%"/>
4+
45
[Netron](https://github.com/lutzroeder/netron)是一款轻量好用、兼容Tensorflow、Python、Caffe等多种模型的可视化模型,同时它还具有跨平台支持、Python模块支持的能力。
56

67
TNN提供了基于Netron的可视化工具,当前正计划提交Pull Request合并,提交通过前可通过如下链接进行下载:
7-
[Linux Netron for TNN]()
8-
[MacOS Netron for TNN]()
8+
[Windows Netron for TNN](https://raw.githubusercontent.com/darrenyao87/tnn-models/master/netron/Netron%20Setup%204.1.1.exe.zip)
9+
[Linux Netron for TNN]() To Add
10+
[MacOS Netron for TNN]() To Add

0 commit comments

Comments
 (0)