search

dusty-nv / ros_deep_learning

Deep learning inference nodes for ROS / ROS2 with support for NVIDIA Jetson and TensorRT
866 stars 258 forks source link

How to use command to do inference #85

Open YaoJiawei329 opened 2 years ago

YaoJiawei329 commented 2 years ago

Hi, all. I am a newbie on ROS and deep learning, wonder how to use these params in /ros_deep_leraning/launch/imagenet.ros1.launch. I just want to run a classification demo on a video(mp4) or a picture(jpg), resnet18. I tried this command

$ roslaunch ros_deep_learning imagenet.ros1.launch input:=file:///home/nvidia/test_videos/test.mp4 output:=file:///home/nvidia/test_result.mp4 model_name:=resnet18

but it turns out this error every try: [ERROR] [1629873729.496301287]: [setParam] Failed to contact master at [localhost:11311]. Retrying... (I had run $ roscore before it.)

I can correctly do inference in jetson-inference. On Xavier, Jetpack 4.5, ROS melodic.

Can someone give me an example or tell me how to fix the ERROR. Thanks very much!