Metrics provided by Seldon core for monitoring ML models' performance in production

[that-rahul-guy]

Hello, For monitoring the ml models in production (deployed via seldon core) I came across the following link which talks about integration with Prometheus: https://docs.seldon.io/projects/seldon-core/en/latest/analytics/analytics.html As far as I understood seldon will expose the metrics on an endpoint /metrics and Prometheus will collect them and then we can use Grafana for further visualizations..

I want to understand what all metrics will the be readily available in /metrics for observing the model performance in production. Is there a list which mentions all the metrics being calculated and exposed by seldon at /metrics endpoint? Also is there any way I can add custom metrics as per requirements?

Please guide me to some resources (if any) or comment below.

Thanks.

[ukclivecox]

You can create custom metrics as outlined here. Does this fulfil your requirements?

[github-actions[bot]]

This issue is stale because it has been open 10 days with no activity. Remove stale label or comment or this will be closed in 5 days.

[ukclivecox]

closing for now - please also look at v2