This following describes how to deploy an RTiC server docker Image on any Linux machine.
Before you start, verify the Pre-requisites for running RTiC.
(1) Login into Docker Hub – In order to pull the RTiC Inference Engine Docker image, run the docker login command in a CLI terminal on the machine on which to deploy RTiC Inference Engine. This can be any machine, such as a local machine. When the prompt is displayed, use the password (access token) issued by Deci.
$ docker login --username deciusers $ Password: <access_token> Login Succeeded
Note – If you do not already have the password (access token) issued by Deci, display the Deci Lab by clicking the Lab tab at the top of the page, select the Optimized Version model to be deployed and click the Deploy button. The access token appears in the displayed window.
(2) Pull RTiC Docker – Use the following command to pull either the RTiC-GPU or the RTiC-CPU image according to the target hardware environment on which the model will be run (CPU or GPU).
docker pull deciai/rtic-gpu:latest
docker pull deciai/rtic-cpu:latest
Note – Pulling the docker may take several minutes, during which you will see the pulling process running on your screen. It ends when all lines are show PULL COMPLETE mode.
Updated 5 months ago