Getting started with TI Jacinto 7 Edge AI/Demos/Classification

From RidgeRun Developer Connection
Jump to: navigation, search

Run the Classification app

/opt/edge_ai_apps/apps_cpp/bin/Release/app_image_classification \
-m /opt/edge_ai_apps/models/classification/TFL-CL-000-mobileNetV1-mlperf \
-i /opt/edge_ai_apps/data/images/%04d.jpg \
-o output/classification_%d.jpg


  • The help will show you the flags options
[docker] root@j7-evm:/opt/edge_ai_apps/apps_cpp/build# /opt/edge_ai_apps/apps_cpp/bin/Release/app_image_classification --help
 
# 
# /opt/edge_ai_apps/apps_cpp/bin/Release/app_image_classification PARAMETERS [OPTIONAL PARAMETERS]
# OPTIONS:
#  --model       |-m Path to the model directory.
#  [--input      |-i Source to gst pipeline camera or file.]
#  [--output     |-o Set gst pipeline output display or file.]
#  [--device     |-d Device name for camera input.]
#  [--index      |-u Start index for multiple file input output.]
#  [--frame      |-f Framerate of gstreamer pipeline for image input.]
#  [--no-curses  |-n Disable curses report.]
#  [--connector  |-c Connector id to select output display.]
#  [--log-level  |-l Logging level to enable. [0: DEBUG 1:INFO 2:WARN 3:ERROR]. Default is 2.
#  [--help       |-h]
# 
# 
# (c) Texas Instruments 2021
#
Retrieved from "http://developer.ridgerun.com/wiki/index.php?title=Getting_started_with_TI_Jacinto_7_Edge_AI/Demos/Classification&oldid=35719"