Polyaxon has an automatic tracking of run events to provide a uniform observability and monitoring for resources as well as statuses, events, and conditions of all runs.
Run status is at the heart of every run. It is used to represent the current state of an experiment or a notebook, and helps users to view passed stages of a run.
Polyaxon has now features for observability and monitoring for resources as well as statuses, events, and conditions.
The CLI has a command argument
--watch -w for watching statuses and events in real-time, same with the UI,
it streams the statuses and provides much deeper insight about the conditions of the underlying pods.
You can get instant information about any run’s lifecycle and stage:
You can enable resources tracking for your experiment to log MemCPU/GPU resource consumption