GstInference/Benchmarks: Difference between revisions

Revision as of 22:34, 6 July 2020

GstInference Benchmarks

The following benchmarks were run with a source video (1920x1080@60). With the following base GStreamer pipeline, and environment variables:

$ VIDEO_FILE='video.mp4'
$ MODEL_LOCATION='graph_inceptionv1_tensorflow.pb'
$ INPUT_LAYER='input'
$ OUTPUT_LAYER='InceptionV1/Logits/Predictions/Reshape_1'

The environment variables were changed accordingly with the used model (Inception V1,V2,V3 or V4)

GST_DEBUG=inception1:1 gst-launch-1.0 filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model inceptionv1 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER net.src_model ! perf ! fakesink -v

The Desktop PC had the following specifications:

Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz
8 GB RAM
Cedar [Radeon HD 5000/6000/7350/8350 Series]
Linux 4.15.0-54-generic x86_64 (Ubuntu 16.04)

The Jetson Xavier power modes used were 2 and 6 (more information: Supported Modes and Power Efficiency)

View current power mode:

$ sudo /usr/sbin/nvpmodel -q

Change current power mode:

sudo /usr/sbin/nvpmodel -m x

Where x is the power mode ID (e.g. 0, 1, 2, 3, 4, 5, 6).

Summary

Desktop PC	CPU Library
Model	Framerate	CPU Usage
Inception V1	11.89	48
Inception V2	10.33	65
Inception V3	5.41	90
Inception V4	3.81	94

Jetson Xavier (15W)	CPU Library		GPU Library
Model	Framerate	CPU Usage	Framerate	CPU Usage
Inception V1	8.24	86	52.3	43
Inception V2	6.58	88	39.6	42
Inception V3	2.54	92	17.8	25
Inception V4	1.22	94	9.4	20

Jetson Xavier (30W)	CPU Library		GPU Library
Model	Framerate	CPU Usage	Framerate	CPU Usage
Inception V1	6.41	93	66.27	72
Inception V2	5.11	95	50.59	62
Inception V3	1.96	98	22.95	44
Inception V4	0.98	99	12.14	32

Framerate

CPU Usage

TensorFlow Lite Benchmarks

FPS measurement

CPU usage measurement

Test benchmark video

The following video was used to perform the benchmark tests.
To download the video press right click on the video and select 'Save video as' and save this in your computer.

Test benchmark video

ONNXRT Benchmarks

FPS Measurements

CPU Load Measurements

Test benchmark video

The following video was used to perform the benchmark tests.
To download the video press right click on the video and select 'Save video as' and save this in your computer.

Test benchmark video

Previous: Example Applications/DispTec

Index

Next: Model Zoo

@@ Line 178: / Line 178: @@
    }
    </style>
-</style>
+<div id="Buttons_Model" style="margin: auto; width: 1300px; height: auto;">
-  <div id="Buttons_Model" style="margin: auto; width: 1300px; height: auto;">
    <button class="button" id="show_inceptionv1">Show InceptionV1 </button>
    <button class="button" id="show_inceptionv2">Show InceptionV2 </button>
@@ Line 189: / Line 188: @@
 <br><br>
 <div id="chart_div" style="margin: auto; width: 800px; height: 500px;"></div>
+<br><br>
+<div id="Buttons_Backend" style="margin: auto; width: 600px; height: auto;">
+<button class="button" id="show_onnxrt">Show ONNXRT </button>
+</div>
+<div id="chart_div1" style="margin: auto; width: 800px; height: 500px;"></div>
 <br><br>
@@ Line 195: / Line 200: @@
        google.charts.setOnLoadCallback(drawStuff);
        function drawStuff() {
          var chartDiv = document.getElementById('chart_div');
+        var chartDiv1 = document.getElementById('chart_div1');
          var table_backend_platform_fps = google.visualization.arrayToDataTable([
@@ Line 209: / Line 216: @@
            ['x86', 47.8702, 32.7236, 12.092, 5.2632, 16.03, 18.3592]
          ]);
+        var table_model_platform_fps = google.visualization.arrayToDataTable([
+          ['Model',                        //Colunm 0
+            'ONNXRT \n x86',               //Colunm 1
+          ['InceptionV1', 47.8702], //row 1
+          ['InceptionV2', 32.7236], //row 2
+          ['InceptionV3', 12.092], //row 3
+          ['InceptionV4', 5.2632], //row 4
+          ['TinyYoloV2',  16.03], //row 5
+          ['TinyYoloV3',  18.3592]  //row 6
+        ]);
          var materialOptions = {
-           width: 320,
+           width: 350,
+          chart: {
+            title: 'Model vs Platform per backend',
+          },
+          series: {
+          },
+          axes: {
+            y: {
+              distance: {side: 'left',label: 'FPS'}, // Left y-axis.
+            }
+          }
+        };
+       var materialOptions1 = {
+          width: 400,
            chart: {
-             title: 'Model Vs Platform per backend',
+             title: 'Model vs backend per platform',
            },
            series: {
@@ Line 225: / Line 254: @@
          var materialChart_fps = new google.charts.Bar(chartDiv);
+        var materialChart1_fps = new google.charts.Bar(chartDiv1);
          view_fps = new google.visualization.DataView(table_backend_platform_fps);
+        view1_fps = new google.visualization.DataView(table_model_platform_fps);
          function drawMaterialChart() {
            var materialChart_fps = new google.charts.Bar(chartDiv);
+          var materialChart1_fps = new google.charts.Bar(chartDiv1);
            materialChart_fps.draw(table_backend_platform_fps, google.charts.Bar.convertOptions(materialOptions));
+          materialChart1_fps.draw(table_model_platform_fps, google.charts.Bar.convertOptions(materialOptions1));
            init_charts();
          }
@@ Line 236: / Line 269: @@
            view_fps.hideColumns([2,3,4,5,6]);
            materialChart_fps.draw(view_fps, materialOptions);
+          view_fps.setColumns([0,1]);
+          view_fps.hideColumns([2,3,4,5,6]);
+          materialChart1_fps.draw(view1_fps, materialOptions1);
          }