CVS-162115 - Deprecate OVMS

geti_sdk/deployment/deployed_model.py

@@ -266,6 +266,10 @@ def load_inference_model(
                     f"requests: {optimal_requests}"
                 )
         else:
+            logging.warning(
+                "Model inference through OpenVINO model server is DEPRECATED and will be removed in the future. "
+                "Please use local inference instead for more stable results."
+            )
             # Connect to an OpenVINO model server instance
             model_name = generate_ovms_model_name(project=project, model=self)
             model_address = generate_ovms_model_address(

geti_sdk/deployment/resources/OVMS_README.md

@@ -1,6 +1,5 @@
 # Deploying Geti models with OpenVINO Model Server (OVMS)
-> NOTE: Be advised that is an experimental feature which may be subject to change in
-> the future
+> Note: This feature is deprecated and is not recommended for use.
 
 This README describes how to set up an OpenVINO Model Server for any Intel® Geti™
 project. Please note that it is meant as an example only, and the configuration used