server
server copied to clipboard
Add testing for decoupled model use case
Testing added for the specific use case of launching a separate thread for decoupled models. Fixed a small issue for the CI as well.
PYBE: https://github.com/triton-inference-server/python_backend/pull/358