Hi Community, How Can I Prevent Clearml Creating A New Experiment, Each Time I Interrupt And Restart Training On The Same Task? I'M Training Yolov8 And Clearml Docker Usage Is Up To 30Gb. I Can'T See A Yaml Config Parameter For This.

Answered

Hi community, how can I prevent ClearML creating a new experiment, each time I interrupt and restart training on the same task? I'm training YOLOv8 and clearML docker usage is up to 30gb. I can't see a YAML config parameter for this.

  				
Posted 
	one year ago

					More
				  		
  Report
		
					ConvolutedLeopard95
				
					0

Votes Newest

Answers 2

Hi @<1539780305588588544:profile|ConvolutedLeopard95> , you mean you want to re-use the same experiment so it won't create another experiment object in the system? Not sure what you mean by docker usage

  				
Posted 
	one year ago

					More
				  		
  Report
		
					CostlyOstrich36
				
					0

Hey @<1539780305588588544:profile|ConvolutedLeopard95> , unfortunately this is not built-in into the YOLOv8 tracker. Would you mind opening an issue on the YOLOv8 github page and atting me? (I'm thepycoder on github)

I can then follow up the progress on it, because it makes sense to expose this parameter through the yaml.

That said, to help you right now, please change this line in the YOLOv8 codebase to enable overriding of tasks. This will happen ONLY if the previous task has no artifacts.

  				
Posted 
	one year ago

					More
				  		
  Report
		
					ExasperatedCrab78
				
					0
					 × 1

Write your answer

587 Views

2 Answers

one year ago