You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
We are trying to use Python Backend of Triton server. We are using Model.py and writing our logic and model usage inside it. I want to know how can we use metrics in Python backend so that it can be retrieved from deployed Triton Server.
Describe the solution you'd like
Model.py file could accompdate something like
Import triton_python_backend_utils as pb_utils
.....
execute()
internal_process_key = pb_utils.Metrics("key_for_prometheus")
Someprocess....
internal_process_key.inc()
Describe alternatives you've considered
Not yet as, could not find how this could be used
The text was updated successfully, but these errors were encountered:
Hi @AniForU, the custom metrics support in Python backend utilizes the in-process Triton C API under the hood, so you should be able to retrieve the custom metrics by any of the metrics endpoint.
Is your feature request related to a problem? Please describe.
We are trying to use Python Backend of Triton server. We are using Model.py and writing our logic and model usage inside it. I want to know how can we use metrics in Python backend so that it can be retrieved from deployed Triton Server.
Describe the solution you'd like
Model.py file could accompdate something like
Import triton_python_backend_utils as pb_utils
.....
execute()
internal_process_key = pb_utils.Metrics("key_for_prometheus")
Someprocess....
internal_process_key.inc()
Describe alternatives you've considered
Not yet as, could not find how this could be used
The text was updated successfully, but these errors were encountered: