Now I am using a docker (that has pytorch and cuda inside : pytorch/pytorch:1.8.1-cuda10.2-cudnn7-devel) because we need to use GPU , then i install tiago package inside that docker. In general docker build and rocker run docker container complete.
Inside docker container
When I run roslaunch a simulation roslaunch tiago_gazebo tiago_gazebo.launch public_sim:=true robot:=steel
Gazebo ran but without tiago model inside, on the terminal there is a error like fail when load URDF model. More detail about error here:
Can you have any suggestion for me? to fix the issue.
Or suggest me other way to make dockerfile that can use GPU ( cuda, Pytorch ). I tried to use tiago docker (docker base on osrf/ros:melodic-desktop-full-bionic ) it work good with tiago package but it doesn't have cuda and pytorch inside. Can anyone suggest me install cuda inside tiago docker?
Many thanks
Dat
Now I am using a docker (that has pytorch and cuda inside :
pytorch/pytorch:1.8.1-cuda10.2-cudnn7-devel
) because we need to use GPU , then i install tiago package inside that docker. In general docker build and rocker run docker container complete.Inside docker container When I run roslaunch a simulation roslaunch tiago_gazebo tiago_gazebo.launch public_sim:=true robot:=steel Gazebo ran but without tiago model inside, on the terminal there is a error like fail when load URDF model. More detail about error here:
Can you have any suggestion for me? to fix the issue.
Or suggest me other way to make dockerfile that can use GPU ( cuda, Pytorch ). I tried to use tiago docker (docker base on osrf/ros:melodic-desktop-full-bionic ) it work good with tiago package but it doesn't have cuda and pytorch inside. Can anyone suggest me install cuda inside tiago docker? Many thanks Dat