GstInference/Benchmarks: Difference between revisions

From RidgeRun Developer Wiki
Line 194: Line 194:
       google.charts.load('current', {'packages':['corechart', 'bar']});
       google.charts.load('current', {'packages':['corechart', 'bar']});
       google.charts.setOnLoadCallback(drawStuff);
       google.charts.setOnLoadCallback(drawStuff);
 
     


       function drawStuff() {
       function drawStuff() {
Line 205: Line 205:


         var table_backend_platform_fps = google.visualization.arrayToDataTable([
         var table_backend_platform_fps = google.visualization.arrayToDataTable([
           ['Platform',
           ['Platform',                     //Colunm 0
           'InceptionV1 \n Tensorflow',
           'InceptionV1 \n Tensorflow',     //Colunm 1
           'InceptionV1 \n Tensorflow-Lite',
           'InceptionV1 \n Tensorflow-Lite',//Colunm 2
           'InceptionV2 \n Tensorflow',
           'InceptionV2 \n Tensorflow',     //Colunm 3
           'InceptionV2 \n Tensorflow-Lite',
           'InceptionV2 \n Tensorflow-Lite',//Colunm 4
           'InceptionV3 \n Tensorflow',
           'InceptionV3 \n Tensorflow',     //Colunm 5
           'InceptionV3 \n Tensorflow-Lite',
           'InceptionV3 \n Tensorflow-Lite',//Colunm 6
           'InceptionV4 \n Tensorflow',
           'InceptionV4 \n Tensorflow',     //Colunm 7
           'InceptionV4 \n Tensorflow-Lite',
           'InceptionV4 \n Tensorflow-Lite',//Colunm 8
           'TinyYoloV2 \n Tensorflow',
           'TinyYoloV2 \n Tensorflow',     //Colunm 9
           'TinyYoloV2 \n Tensorflow-Lite',
           'TinyYoloV2 \n Tensorflow-Lite', //Colunm 10
           'TinyYoloV3 \n Tensorflow',
           'TinyYoloV3 \n Tensorflow',     //Colunm 11
           'TinyYoloV3 \n Tensorflow-Lite'],
           'TinyYoloV3 \n Tensorflow-Lite'],//Colunm 12
           ['x86',      63.747, 22.781, 48.381, 14.164, 20.482, 12.164, 10.338, 10.164, 24.255, 12.164, 27.113, 10.164],
           ['x86',      63.747, 22.781, 48.381, 14.164, 20.482, 12.164, 10.338, 10.164, 24.255, 12.164, 27.113, 10.164],
           ['x86+GPU',  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
           ['x86+GPU',  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], //row 1
           ['TX2',      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
           ['TX2',      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], //row 2
           ['TX2-GPU',  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
           ['TX2-GPU',  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], //row 3
           ['Coral',    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
           ['Coral',    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], //row 4
           ['Coral+TPU', 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
           ['Coral+TPU', 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] //row 5
         ]);
         ]);
         var table_model_platform_fps = google.visualization.arrayToDataTable([
         var table_model_platform_fps = google.visualization.arrayToDataTable([
           ['Model',
           ['Model',                       //Colunm 0
          'Tensorflow \n x86',
            'Tensorflow \n x86',           //Colunm 1
          'Tensorflow \n x86+GPU',
            'Tensorflow \n x86+GPU',       //Colunm 2
          'Tensorflow \n TX2',
            'Tensorflow \n TX2',           //Colunm 3
          'Tensorflow \n TX2+GPU',
            'Tensorflow \n TX2+GPU',       //Colunm 4
          'Tensorflow \n Coral',
            'Tensorflow \n Coral',         //Colunm 5
          'Tensorflow \n Coral+TPU',
            'Tensorflow \n Coral+TPU',     //Colunm 6
          'Tensorflow-Lite \n x86',
            'Tensorflow-Lite \n x86',       //Colunm 7
          'Tensorflow-Lite \n x86+GPU',
            'Tensorflow-Lite \n x86+GPU',   //Colunm 8
          'Tensorflow-Lite \n TX2',
            'Tensorflow-Lite \n TX2',       //Colunm 9
          'Tensorflow-Lite \n TX2+GPU',
            'Tensorflow-Lite \n TX2+GPU',   //Colunm 10
          'Tensorflow-Lite \n Coral',
            'Tensorflow-Lite \n Coral',     //Colunm 11
          'Tensorflow-Lite \n Coral+TPU'],
            'Tensorflow-Lite \n Coral+TPU'],//Colunm 12
           ['InceptionV1', 63.747, 0, 0, 0, 0, 0, 22.781, 0, 0, 0, 0, 0],
           ['InceptionV1', 63.747, 0, 0, 0, 0, 0, 22.781, 0, 0, 0, 0, 0], //row 1
           ['InceptionV2', 48.381, 0, 0, 0, 0, 0, 14.164, 0, 0, 0, 0, 0],
           ['InceptionV2', 48.381, 0, 0, 0, 0, 0, 14.164, 0, 0, 0, 0, 0], //row 2
           ['InceptionV3', 20.482, 0, 0, 0, 0, 0, 12.164, 0, 0, 0, 0, 0],
           ['InceptionV3', 20.482, 0, 0, 0, 0, 0, 12.164, 0, 0, 0, 0, 0], //row 3
           ['InceptionV4', 10.338, 0, 0, 0, 0, 0, 10.164, 0, 0, 0, 0, 0],
           ['InceptionV4', 10.338, 0, 0, 0, 0, 0, 10.164, 0, 0, 0, 0, 0], //row 4
           ['TinyYoloV2',  24.255, 0, 0, 0, 0, 0, 12.164, 0, 0, 0, 0, 0],
           ['TinyYoloV2',  24.255, 0, 0, 0, 0, 0, 12.164, 0, 0, 0, 0, 0], //row 5
           ['TinyYoloV3',  27.113, 0, 0, 0, 0, 0, 10.164, 0, 0, 0, 0, 0]
           ['TinyYoloV3',  27.113, 0, 0, 0, 0, 0, 10.164, 0, 0, 0, 0, 0] //row 6
         ]);
         ]);
         var table_model_backend_fps = google.visualization.arrayToDataTable([
         var table_model_backend_fps = google.visualization.arrayToDataTable([
         ['Model',
         ['Model',                         //Colunm 0
           'Tensorflow \n x86',
           'Tensorflow \n x86',           //Colunm 1
           'Tensorflow \n x86+GPU',
           'Tensorflow \n x86+GPU',       //Colunm 2
           'Tensorflow \n TX2',
           'Tensorflow \n TX2',           //Colunm 3
           'Tensorflow \n TX2+GPU',
           'Tensorflow \n TX2+GPU',       //Colunm 4
           'Tensorflow \n Coral',
           'Tensorflow \n Coral',         //Colunm 5
           'Tensorflow \n Coral+TPU',
           'Tensorflow \n Coral+TPU',     //Colunm 6
           'Tensorflow-Lite \n x86',
           'Tensorflow-Lite \n x86',       //Colunm 7
           'Tensorflow-Lite \n x86+GPU',
           'Tensorflow-Lite \n x86+GPU',   //Colunm 8
           'Tensorflow-Lite \n TX2',
           'Tensorflow-Lite \n TX2',       //Colunm 9
           'Tensorflow-Lite \n TX2+GPU',
           'Tensorflow-Lite \n TX2+GPU',   //Colunm 10
           'Tensorflow-Lite \n Coral',
           'Tensorflow-Lite \n Coral',     //Colunm 11
           'Tensorflow-Lite \n Coral+TPU'],
           'Tensorflow-Lite \n Coral+TPU'],//Colunm 12
           ['InceptionV1', 63.747, 0, 0, 0, 0, 0, 22.781, 0, 0, 0, 0, 0],
           ['InceptionV1', 63.747, 0, 0, 0, 0, 0, 22.781, 0, 0, 0, 0, 0], //row 1
           ['InceptionV2', 48.381, 0, 0, 0, 0, 0, 14.164, 0, 0, 0, 0, 0],
           ['InceptionV2', 48.381, 0, 0, 0, 0, 0, 14.164, 0, 0, 0, 0, 0], //row 2
           ['InceptionV3', 20.482, 0, 0, 0, 0, 0, 12.164, 0, 0, 0, 0, 0],
           ['InceptionV3', 20.482, 0, 0, 0, 0, 0, 12.164, 0, 0, 0, 0, 0], //row 3
           ['InceptionV4', 10.338, 0, 0, 0, 0, 0, 10.164, 0, 0, 0, 0, 0],
           ['InceptionV4', 10.338, 0, 0, 0, 0, 0, 10.164, 0, 0, 0, 0, 0], //row 4
           ['TinyYoloV2',  24.255, 0, 0, 0, 0, 0, 12.164, 0, 0, 0, 0, 0],
           ['TinyYoloV2',  24.255, 0, 0, 0, 0, 0, 12.164, 0, 0, 0, 0, 0], //row 5
           ['TinyYoloV3',  27.113, 0, 0, 0, 0, 0, 10.164, 0, 0, 0, 0, 0]
           ['TinyYoloV3',  27.113, 0, 0, 0, 0, 0, 10.164, 0, 0, 0, 0, 0] //row 6
         ]);
         ]);


Line 271: Line 271:
           width: 900,
           width: 900,
           chart: {
           chart: {
             title: 'Backend Vs Platform per model',
             title: 'Model Vs Platform per backend',
           },
           },
           series: {
           series: {
           },
           },
           axes: {
           axes: {
Line 288: Line 287:
           },
           },
           series: {
           series: {
            0: { axis: 'distance' }, // Bind series 0 to an axis named 'distance'.
           },
           },
           axes: {
           axes: {
Line 302: Line 300:
           },
           },
           series: {
           series: {
            0: { axis: 'distance' }, // Bind series 0 to an axis named 'distance'.
           },
           },
           axes: {
           axes: {
Line 340: Line 337:
         }
         }


 
        // REF_MODEL
         /*Select the backend to filter the data to show in the chart*/
         /*Select the Model that you want to show in the chart*/
         var show_inceptionv1 = document.getElementById('show_inceptionv1');
         var show_inceptionv1 = document.getElementById('show_inceptionv1');
         show_inceptionv1.onclick = function () {
         show_inceptionv1.onclick = function () {
Line 378: Line 375:
           materialChart_fps.draw(view_fps, materialOptions);
           materialChart_fps.draw(view_fps, materialOptions);
         }
         }
        /*  REF_BACKEND  */
         /*Select the backend to filter the data to show in the chart*/
         /*Select the backend to filter the data to show in the chart*/
         var button = document.getElementById('show_tensorflow');
         var button = document.getElementById('show_tensorflow');
Line 391: Line 389:
           materialChart1_fps.draw(view1_fps, materialOptions1);
           materialChart1_fps.draw(view1_fps, materialOptions1);
         }
         }
        /*  REF_PLATFORM  */
         /*Select the Platform to filter the data to show in the chart*/
         /*Select the Platform to filter the data to show in the chart*/
         var show_x86 = document.getElementById('show_x86');
         var show_x86 = document.getElementById('show_x86');
Line 428: Line 427:
           materialChart2_fps.draw(view2_fps, materialOptions2);
           materialChart2_fps.draw(view2_fps, materialOptions2);
         }
         }
         drawMaterialChart();
         drawMaterialChart();
    // listen for error
    google.visualization.events.addListener(chart, 'error', function (err) {
      // check error
      console.log(err.id, err.message);
      // remove error
      google.visualization.errors.removeError(err.id);
      // remove all errors
      google.visualization.errors.removeAll(container);
    });
     };
     };
</script>
</script>
</html>
</html>
Line 477: Line 464:


<script>
<script>
 
            google.charts.load('current', {'packages':['corechart', 'bar']});
       google.charts.setOnLoadCallback(drawStuff1);
       google.charts.setOnLoadCallback(drawStuff1);
        
        
Line 559: Line 546:
           },
           },
           series: {
           series: {
            0: { axis: 'distance' }, // Bind series 0 to an axis named 'distance'.
           },
           },
           axes: {
           axes: {
Line 573: Line 559:
           },
           },
           series: {
           series: {
            0: { axis: 'distance' }, // Bind series 0 to an axis named 'distance'.
           },
           },
           axes: {
           axes: {
Line 587: Line 572:
           },
           },
           series: {
           series: {
            0: { axis: 'distance' }, // Bind series 0 to an axis named 'distance'.
           },
           },
           axes: {
           axes: {
Line 626: Line 610:




         /*Select the backend to filter the data to show in the chart*/
        // REF_MODEL CPU
         /*Select the Model that you want to show in the chart*/
         var show_inceptionv1_cpu = document.getElementById('show_inceptionv1_cpu');
         var show_inceptionv1_cpu = document.getElementById('show_inceptionv1_cpu');
         show_inceptionv1_cpu.onclick = function () {
         show_inceptionv1_cpu.onclick = function () {
Line 663: Line 648:
           materialChart.draw(view, materialOptions);
           materialChart.draw(view, materialOptions);
         }
         }
        /*  REF_BACKEND  CPU*/
         /*Select the backend to filter the data to show in the chart*/
         /*Select the backend to filter the data to show in the chart*/
         var button = document.getElementById('show_tensorflow_cpu');
         var button = document.getElementById('show_tensorflow_cpu');
Line 676: Line 662:
           materialChart1.draw(view1, materialOptions1);
           materialChart1.draw(view1, materialOptions1);
         }
         }
        /*  REF_PLATFORM  CPU*/
         /*Select the Platform to filter the data to show in the chart*/
         /*Select the Platform to filter the data to show in the chart*/
         var show_x86_cpu = document.getElementById('show_x86_cpu');
         var show_x86_cpu = document.getElementById('show_x86_cpu');

Revision as of 14:19, 13 April 2020




Previous: Example Applications/DispTec Index Next: Model Zoo


GstInference
GstInference
Introduction
Getting started
Supported architectures

InceptionV1

InceptionV3

YoloV2

AlexNet
Supported backends

Caffe

Metadata and Signals
Overlay Elements
Utils Elements
Legacy pipelines
Example pipelines
Example applications
Benchmarks
Model Zoo
Project Status
Contact Us

Coral from Google



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

thumb
thumb

CPU Usage

thumb
thumb

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


Previous: Example Applications/DispTec Index Next: Model Zoo



Retrieved from "https://developer.ridgerun.com/wiki/index.php?title=GstInference/Benchmarks&oldid=29220"