[X] I had searched in the issues and found no similar feature requirement.
Description
does (kube)ray support scheduled jobs (e.g. through cronjobs)?
is it possible to chain different jobs/pipelines together, e.g. when a deployed modelA returns a specific value, then start a training pipeline for another model and so on. as in, is there the concept of workflows in ray?
is there any incremental (online) learning model implementation in ray library and if so, how to distribute it and checkpointing it ?
Search before asking
Description
does (kube)ray support scheduled jobs (e.g. through cronjobs)?
is it possible to chain different jobs/pipelines together, e.g. when a deployed modelA returns a specific value, then start a training pipeline for another model and so on. as in, is there the concept of workflows in ray?
is there any incremental (online) learning model implementation in ray library and if so, how to distribute it and checkpointing it ?
Use case
No response
Related issues
No response
Are you willing to submit a PR?