Hi Guys, I Recently Started Working On Clearml. I Have A Pipeline Which Has Couple Of Tasks Added To It. The Pipeline Executes And Completes As Expected When I Use The Web App. I Am Trying To Start This Pipeline From The Sdk Api. I Have A Two Part Questio

Unanswered

Hi guys,
I recently started working on ClearML. I have a pipeline which has couple of tasks added to it. The pipeline executes and completes as expected when I use the Web App. I am trying to start this pipeline from the SDK API. I have a two part question

I use PipelineController.get to get the pipeline controller but when I try to enqueue the pipeline using PipelineController.enqueue using the controller from the get method and the mentioning the queue_name as well, I receive an error as below clearml.backend_interface.session.SendError: Action failed <400/110: tasks.enqueue/v1.0 (Invalid task status (Invalid status change): current_status=completed, new_status=queued)> . So, I thought that the pipeline which has been completed once previously would not change its status meaning, it wont re-run that would change the status of an already successfully completed pipeline. But I was able to re-run the pipeline on the WebApp. What am I doing wrong here?
The other question is how can add the Parameters programmatically to the new run of the pipeline, which is possible to do using the WebApp. I want to re-run this pipeline programmatically using different values.

  				
Posted 
	one year ago

					More
				  		
  Report
		
					DullCentipede49
				
					0
					 × 1

Write your answer

1K Views

0 Answers

one year ago

Hi Clearml Team, May I Ask Do I Have To Start An Agent On Services Mode Solely To Listen To The Queue Or Not If I Am Running On Clearml Hosted Server (Pro Tier)? My Use Case Is Full Automated Ml Pipeline Where I'D Like To Trigger Re-Train On Dataset Upda

Hi Guys, I’M Trying To Deploy Clearml To K8S. When I Try To Execute A Pipeline (From The Examples) On A Remote Worker, I Get An

Hello Clearml Community, I'M Using A Self-Hosted Clearml Setup, And While It Generally Performs Well, I Encounter Issues During Disruptions, Such As Temporary Network Problems. Here'S The Specific Scenario: I Have A Pipeline With Three Tasks. The First

Hi All, Juts Learning The Ropes Of Clearml Atm. And Am Doing A Really Simple Etl Pipeline: Raw Data -> Clean Data My Current Approach Is In One Script, I Add The Raw Data File To A Dataset In The Project: # Register_Raw.Ipynb

Hi, I Was Using Pipeline Controller To Run Pipeline Tasks With 2 Steps, The 1St Step Is Supposed To Create 51 Task And The 2Nd Task Will Compute Result Based On The 51St Tasks Output In The 1St Step. But I Only See 34 Tasks In Step 1, And Got Error Of The

Hello! Can You Help Me To Understand How Clearml Captures Provided Args? During The First Execution Of Pipeline As A Code Script I Provide Parameters Using `Argparse` With Default Values. And They Seem To Be Captured. But Later When I Try To Run Pipeline

Hey Team, I Had A Question About Executing The Pipelines Using Clearml Agent Setup In K8S. When We Define A Pipeline Script To Be Executing Remotely And Submit It To The Queue For Execution We See Following Things Happening

Hey, I Have Clearml Install On-Prem K8S. Anyonw Know How Can I Backup The Data , In Case Of Dr? I Do Not Care For Username \ Passwords \ Pipeline Settings, But The Data From All The Tasks Is Important For Us

Hello! I Need To Run Clearml Pipeline With Caching Of Steps. I Specify Cache_Executed_Step=True For Each Step, But My Steps Are Not Cached And Ended With The Status - Completed.