ClearML FAQ | Hi, I Am Trying To Run Experiment From Clearml Web Ui. I Did Experiment Copy, Enqueue, But In The Execution Log I See That It Runs Command

Answered

Hi, I Am Trying To Run Experiment From Clearml Web Ui. I Did Experiment Copy, Enqueue, But In The Execution Log I See That It Runs Command

Hi, I am trying to run experiment from ClearML web ui. I did experiment copy, enqueue, but in the execution log I see that it runs command
[.]$ /home/exx/.clearml/venvs-builds/3.8/bin/python -u train.py but I need to add experiment=my_config after train.py. Is there any way to do it from UI?
Thank you

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					MortifiedDove27
				
					0
					 × 1

Votes Newest

Answers 31

, I need to understand it what happens when I press "Enqueue" In web UI and set it to default queue

The Task ID is pushed into the execution queue (from the UI / backend that is it), Then you have clearml-agent running on Your machine, the agent listens on queue/s and pulls jobs from queue.
It will pull the Task ID from the queue, setup the environment according to the Task (i.e. either inside a docker container or in a new virtual-env), clone the code/apply uncommitted changes install the python packages etc. then it will spin the code which will use the configuration in the UI (instead of logging into the UI, when executed manually)
Make sense ?

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					AgitatedDove14
				
					0
					 × 1

Show more results

Write your answer

3K Views

31 Answers

2 years ago

one month ago

Tags

Similar posts

Hi All, I Have A Question Regarding Clearml Task Initialization With Multithreading. I'M Using Python'S Joblib Library And The Parallel Class To Run An Experiment In Multiple Parallel Threads. The Experiment Runs To Completion But I Get Incomplete Mllogge

Hello Everyone, Thanks For The Fruitful Conversations! I Am Trying To Run A Worker And A Queue On A Locally Hosted Clearml Server. But We I Enqueue An Experiment, It Always Collapses With This, Any Ideas What To Do? I Set He Worker Up With Clearml-Agent

Hi, I Noticed That My Experiments Fail If I Try To Clone Them And Enqueue Them After I Already Pushed Another Commits (Following The Commit Of The Cloned Experiment). I Am Getting The Error

Good Day All. I'M Working With The Pipelinecontroller For Creating Pipelines From Tasks. One Thing I'Ve Noticed Is That If I Do Pipe.Start_Locally() It Does Run The Pipeline But It Still Seems To Enqueue The Tasks And Runs Them On The Remote Agents. Am I

Hi, Is There A Solution For User-Roles? I Want To Give Access To Clearml To A User Outside The Ds Team, But Only To View Experiment Results. I Don'T Want Them To Accidently Clone/Archive/Enqueue Anything. Is This An Option?

After Recent Clearml Server Update, Whenever I Clone An Experiment, The Default Project For The Draft Copy Is The First Project In The List. Previously, It Would Be The Project Which I Am Cloning This Experiment From. This Was Much More Convenient. Is Thi

Hey Folks, When I Am Running The Experiments, I Currently Do Not See The Console Logs Coming Up For The Experiment Runs. I Am Using

Hi, I Am Trying To Clone An Experiment. Using The Server Gui, I Select 'Clone' And Then 'Enqueue'. In The Console Window, I See That Clearml Makes Sure The Environment Is Installed, And Then It Goes Into A 'Completed' Status Although The Experiment Did N

Hello, Can I Somehow Make An Exact Copy Of Completed Task Including Scalars, Plots, Artifacts, Etc.? I Do Not Plan Run The Experiment Again, I Just Want To Add Some Metrics Evaluated On A New Dataset To The Copy Of The Task. Thank You In Advance.

Hi Folks, Occasionally When I Clone A Job And Enqueue It, Instead Of Being Processed By The Expected Queue, A New Queue (With Some Id That Looks Like An Hash) Is Created Instead, And The Experiment Hangs In A "Pending" State. When This Happens, If I Abor