@Lovedeep Singh The first two conditions that should be satisfied are:
- the docker image must be created based on prompt flow base image,
mcr.microsoft.com/azureml/promptflow/promptflow-runtime-stable:<newest_version>
. You can find the newest version here. - the environment definition must include the
inference_config
.
If your image is already created on top of base image, then the current flow.dag.yaml
file should point to the correct location of this image.
The second requirement of inference_config
this should be added after the image field in your current yaml.
inference_config:
liveness_route:
port: 8080
path: /health
readiness_route:
port: 8080
path: /health
scoring_route:
port: 8080
path: /score
If this answers your query, do click Accept Answer
and Yes
for was this answer helpful. And, if you have any further query do let us know.