Fix a typo on simplePosetEstimation.m in GPU deployment.

Regenerate all HTML files to ship as a toolbox.

Author: tohruk

doc/SimplePoseNetEstimationOnJetson.md

@@ -1,4 +1,4 @@
-# Deploy Simple Pose Estimation on NVIDIA® Jetson™ Using GPU Coder™
+# Deploy Simple Pose Estimation on NVIDIA(R) Jetson(TM) Using GPU Coder(TM)
 
 
 ![image_0_png.jpg](SimplePoseNetEstimationOnJetson_images/image_0_png.jpg)
@@ -29,7 +29,28 @@ Running on a Jetson AGX Xavier.
 
 ```matlab:Code
 clear;
-hwobj = jetson('192.168.55.1', 'nvidia', 'nvidia');
+hwobj = jetson('tohru-jetson-nano', 'nvidia', 'nvidia');
+```
+
+
+```text:Output
+Checking for CUDA availability on the Target...
+Checking for 'nvcc' in the target system path...
+Checking for cuDNN library availability on the Target...
+Checking for TensorRT library availability on the Target...
+Checking for prerequisite libraries is complete.
+Gathering hardware details...
+Checking for third-party library availability on the Target...
+Gathering hardware details is complete.
+ Board name         : NVIDIA Jetson TX1
+ CUDA Version       : 10.0
+ cuDNN Version      : 7.3
+ TensorRT Version   : 5.0
+ GStreamer Version  : 1.14.5
+ V4L2 Version       : 1.14.2-1
+ SDL Version        : 1.2
+ Available Webcams  : MicrosoftÂ(R) LifeCam Cinema(TM)
+ Available GPUs     : NVIDIA Tegra X1
 ```
 
 # **Verify the GPU Environment**
@@ -41,12 +62,24 @@ Use the [coder.checkGpuInstall](matlab:doc('coder.checkGpuInstall')) function an
 
 ```matlab:Code
 envCfg = coder.gpuEnvConfig('jetson');
-envCfg.DeepLibTarget = 'tensorrt';
+envCfg.DeepLibTarget = 'cudnn';
 envCfg.DeepCodegen = 1;
 envCfg.HardwareObject = hwobj;
 coder.checkGpuInstall(envCfg);
 ```
 
+
+```text:Output
+互換性のある GPU               : パス 
+CUDA 環境                  : パス 
+	ランタイム     : パス 
+	cuFFT     : パス 
+	cuSOLVER  : パス 
+	cuBLAS    : パス 
+cuDNN 環境                 : パス (警告: 深層学習コード生成は、cuDNN v7.5 を使用してテストされました。指定された cuDNN ライブラリの v7.3 に完全な互換性がない可能性があります。)
+深層学習 (cuDNN) コードの生成      : パス 
+```
+
 # **Prepare Pose Estimation for Stand-Alone Deployment**
 
 
@@ -61,13 +94,41 @@ d = imageDisplay(hwobj);
 image(d,img);
 ```
 
+
+![figure_0_png.jpg](SimplePoseNetEstimationOnJetson_images/figure_0_png.jpg)
+
 # Test Pose Estimation Function on MATLAB
 
 ```matlab:Code
 clear hwobj d w envCfg;
 figure;
 set(gcf,'Visible','on');
 simplePoseEstimation
+```
+
+
+```text:Output
+Checking for CUDA availability on the Target...
+Checking for 'nvcc' in the target system path...
+Checking for cuDNN library availability on the Target...
+Checking for TensorRT library availability on the Target...
+Checking for prerequisite libraries is complete.
+Gathering hardware details...
+Checking for third-party library availability on the Target...
+Gathering hardware details is complete.
+ Board name         : NVIDIA Jetson TX1
+ CUDA Version       : 10.0
+ cuDNN Version      : 7.3
+ TensorRT Version   : 5.0
+ GStreamer Version  : 1.14.5
+ V4L2 Version       : 1.14.2-1
+ SDL Version        : 1.2
+ Available Webcams  : MicrosoftÂ(R) LifeCam Cinema(TM)
+ Available GPUs     : NVIDIA Tegra X1
+```
+
+
+```matlab:Code
 % Need to Ctrl+C to stop the running program.
 ```
 
@@ -77,25 +138,83 @@ simplePoseEstimation
 cfg = coder.gpuConfig('exe');
 cfg.Hardware = coder.hardware('NVIDIA Jetson');
 cfg.Hardware.BuildDir = '~/remoteBuildDir';
-cfg.DeepLearningConfig = coder.DeepLearningConfig('tensorrt');
+cfg.DeepLearningConfig = coder.DeepLearningConfig('cudnn');
 cfg.GenerateExampleMain = 'GenerateCodeAndCompile';
 codegen('-config ',cfg,'simplePoseEstimation.m','-report');
 ```
 
+
+```text:Output
+Warning: 無限ループのため、関数 'simplePoseEstimation' が終了しません。
+
+警告 ==> simplePoseEstimation 行: 65 列: 1
+コードの生成が成功しました (警告あり):レポートの表示
+```
+
 # Run the Pose Estimation on the Target
 
 
-To run the generated executable on the target, use the MATLAB® runApplication function.
+To run the generated executable on the target, use the MATLAB(R) runApplication function.
 
 
 
 ```matlab:Code
 clear hwobj;
 hwobj = jetson
+```
+
+
+```text:Output
+Checking for CUDA availability on the Target...
+Checking for 'nvcc' in the target system path...
+Checking for cuDNN library availability on the Target...
+Checking for TensorRT library availability on the Target...
+Checking for prerequisite libraries is complete.
+Gathering hardware details...
+Checking for third-party library availability on the Target...
+Gathering hardware details is complete.
+ Board name         : NVIDIA Jetson TX1
+ CUDA Version       : 10.0
+ cuDNN Version      : 7.3
+ TensorRT Version   : 5.0
+ GStreamer Version  : 1.14.5
+ V4L2 Version       : 1.14.2-1
+ SDL Version        : 1.2
+ Available Webcams  : MicrosoftÂ(R) LifeCam Cinema(TM)
+ Available GPUs     : NVIDIA Tegra X1
+hwobj = 
+  jetson のプロパティ:
+
+       DeviceAddress: 'tohru-jetson-nano'
+                Port: 22
+           BoardName: 'NVIDIA Jetson TX1'
+         CUDAVersion: '10.0'
+        cuDNNVersion: '7.3'
+     TensorRTVersion: '5.0'
+          SDLVersion: '1.2'
+         V4L2Version: '1.14.2-1'
+    GStreamerVersion: '1.14.5'
+             GPUInfo: [1x1 struct]
+          WebcamList: {'MicrosoftÂ(R) LifeCam Cinema(TM)'}
+
+```
+
+
+```matlab:Code
 pid = runApplication(hwobj,'simplePoseEstimation');
 ```
 
 
+```text:Output
+### Launching the executable on the target...
+Executable launched successfully with process ID 6938.
+Displaying the simple runtime log for the executable...
+
+Note: For the complete log, run the following command in the MATLAB command window:
+system(hwobj,'cat /home/nvidia/remoteBuildDir/MATLAB_ws/R2020a/C/matlab/pose-estimation-github/work/simplePoseEstimation.log')
+```
+
+
 ```matlab:Code
 %killApplication(hwobj,'simplePoseEstimation');
 ```

doc/SimplePoseNetEstimationOnJetson_images/figure_0_png.jpg

@@ -12,7 +12,7 @@ This example shows how to predict human poses using a pretrained mode with a web
 # Load Camera and Pretrained Network
 
 
-Connect to the camera and load a pretrained GoogLeNet network. You can use any pretrained network at this step. The example requires MATLAB Support Package for USB Webcams, and Deep Learning Toolbox™ Model *for GoogLeNet Network*. If you do not have the required support packages installed, then the software provides a download link.
+Connect to the camera and load a pretrained GoogLeNet network. You can use any pretrained network at this step. The example requires MATLAB Support Package for USB Webcams, and Deep Learning Toolbox(TM) Model *for GoogLeNet Network*. If you do not have the required support packages installed, then the software provides a download link.
 
 
 

doc/SimplePoseNetEstimationWithWebcamExample.md

@@ -31,7 +31,7 @@ As a training dataset, we use COCO\hyperref{72EF1FDB}{[2]}. COCO's annotation da
 
 ```matlab:Code
 % COCO datset root directory.
-cocoDir = 'D:\coco';
+cocoDir = 'D:\Dataset\coco';
 if ~exist(cocoDir,"dir")
     error("Dataset cannot be found.")
 end
@@ -334,7 +334,7 @@ initialLearnRate = 0.001;
 
 
 
-Train on a GPU if one is available. Using a GPU requires Parallel Computing Toolbox™ and a CUDA® enabled NVIDIA® GPU with compute capability 3.0 or higher.
+Train on a GPU if one is available. Using a GPU requires Parallel Computing Toolbox(TM) and a CUDA(R) enabled NVIDIA(R) GPU with compute capability 3.0 or higher.
 
 
 
@@ -824,7 +824,11 @@ end
 end
 ```
 
-  
+
+
+**Heatmaps visualization:**
+
+
 
 ```matlab:Code
 function Iout = visualizeHeatmaps(heatmaps, I)
@@ -843,7 +847,11 @@ Iout = im2uint8(Iout);
 end
 ```
 
-  
+
+
+**Keypoints visualization:**
+
+
 
 ```matlab:Code
 function Iout = visualizeKeyPoints(I,joints,skeleton,bboxes)

doc/SimplePoseNetTrainingExample.md

@@ -3,14 +3,9 @@
 %
 % Copyright 2019 The MathWorks, Inc.
 
-function simplePoseEstimation
-%#codegen
-%
-% Copyright 2019 The MathWorks, Inc.
-
 persistent simplePoseNet;
 if isempty(simplePoseNet)
-    simplePoseNet = coder.loadDeepLearningNetwork('..\train\simplePoseNet.mat','simplePoseNet');
+    simplePoseNet = coder.loadDeepLearningNetwork('simplePoseNet.mat','simplePoseNet');
 end
 
 hwobj = jetson; % To redirect to the code generatable functions.
@@ -31,8 +26,8 @@
 
     % Detect keypoints
     output = simplePoseNet.predict(Icrop);
-    [scores,idx] = max(output,[],[1,2],'linear');
-    [y,x,~,~] = ind2sub(size(output),idx);
+    [scores,idx_out] = max(output,[],[1,2],'linear');
+    [y,x,~,~] = ind2sub(size(output),idx_out);
     keypoints = permute(cat(1,x,y),[3,1,2]);
 
     % Display image.