AlertBlackbird30 :

--remote-gateway [REMOTE_GATEWAY] Advanced: Specify gateway ip/address to be passed to interactive session (for use with k8s ingestion / ELB)I see this in clearml-session - what’s the intent here?

AgitatedDove14 - i am disabling pytorch like above but still see auto models . I even see model import when running evaluation from a loaded model

AgitatedDove14 - looks like the serving is doing the savemodel stuff?

https://github.com/allegroai/clearml-serving/blob/main/clearml_serving/serving_service.py#L554

sure, will do AlertBlackbird30

Thanks AgitatedDove14 - is the datasets feature pretty new?

AgitatedDove14 - just saw about start_remotely - https://clear.ml/docs/latest/docs/references/sdk/automation_controller_pipelinecontroller#start_remotely

This means the services agent will take care of taking it to completion right?

Or WARNING should be param not found, but using General/param etc

I have convention of root task for experiments and a sub-project Pipelines for running “sandboxed” pipeline task. So I do some magic like this

It does have click==8.0.1

Good question 🙂

this is what I am seeing in the logs:

` No tasks in queue 9154efd8a1314550b1c7882981720861
No tasks in Queues, sleeping for 5.0 seconds
No tasks in queue 9154efd8a1314550b1c7882981720861
No tasks in Queues, sleeping for 5.0 seconds
No tasks in queue 9154efd8a1314550b1c7882981720861
No tasks in Queues, sleeping for 5.0 seconds
No tasks in queue 9154efd8a1314550b1c7882981720861
No tasks in Queues, sleeping for 5.0 seconds
K8S Glue pods monitor: Failed parsing kubectl output...

But if haven’t done a task yet

The agent ip? Generally what’s the expected pattern to deploy and scale this for multiple models?

AgitatedDove14 - these instructions are out of date? https://allegro.ai/clearml/docs/docs/deploying_clearml/clearml_server_kubernetes_helm.html

Yes. It's a pickle file that I have added via OutputModel

OK i found what’s happening:

I had an additional Task.init() - just the blank one, to get the project name. Adding the disable to that as well fixed the issue

I am on 1.0.3

I am doing something like this with a yaml based pipelines DSL

and then do Task.current_task() it is None as expected

Any updates on trigger and schedule docs 🙂

I always have my notebooks in git repo but suddenly it's not running them correctly.

And other question is clearml-serving ready for serious use?

AlertBlackbird30 - i don’t understand why it can’t be a focus though. Probably missing some context.

Yeah will use that

Also the pipeline ran as per this example - https://github.com/allegroai/clearml/blob/master/examples/pipeline/pipeline_controller.py

“notebook as a service” right

Task.add_requirements would fit the bill yeah, thanks

Initially thought use_current_task in Dataset.create does it, but that seems to set the DataprocessingTask itself to the current task

Only one. Will replicate it in detail and see what’s actually up

Something like the Gitlab Kubernetes Executor could work well.

This Docker Mode - https://allegro.ai/clearml/docs/docs/use_cases/clearml_agent_use_case_examples.html#docker-mode