I have followed the following blog from NVIDIA to setup Triton Inference Server
I was able to create Kubernetes Deployment and Kubernetes Services.
But when I was trying to access the clusterIP of the kubernetes Service I am getting ConnectionRefusedError
The script used for this is taken from GitHub - triton-inference-server/client: Triton Python, C++ and Java client libraries, and GRPC-generated client examples for go, java and scala.
And the command used to run the script is following
python3 client/src/python/examples/image_client.py
-u TRITON_HTTP_ENDPOINT
-m inception_graphdef
-s INCEPTION
-x 1
-c 1
car.jpg
TRITON_HTTP_ENDPOINT is replaced with the Cluster-IP of kubernetes service and port 8003
The screenshot of error message is attached .