Good Morning! I Have A Quick Question Regarding Clearml Dataset. When Creating A New Dataset, You Have The Option To Create A Description With

Answered

good morning!
I have a quick question regarding ClearML Dataset. When creating a new dataset, you have the option to create a description with .set_description but how do i get that description back in a script without using the UI? is there a something equivalent to .get_description ?

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					MassiveGoldfish6
				
					0
					 × 1

Votes Newest

Answers

Hi @<1547028031053238272:profile|MassiveGoldfish6> , I think you can access the description by fetching the dataset and searching inside the different attributes. You can use dir() in python to see what's attached. I think you can look for Dataset.data after you fetch it

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					CostlyOstrich36
				
					0

Write your answer

2K Views

1 Answer

2 years ago

Hey Guys, If I Set File_Upload_Retries Arbitrarily High Like 1000, Should I Be Good To Upload A Large Dataset >1Gb On Spotty Connections?

Is There A Way To Tag A Task/Dataset "Folder" With The Sdk? In The Ui There Is An Option To Add A Tag To A Certain Project Folder, But In Python It Seems That It Can Be Done Only On Single Task Runs.

Heya, Good Day To Everyone, I'M Finding Myself Facing This Random Error With A Very Opaque Backtrace When Attempting To Squash Two Distinct Versions Of The Same Dataset, Does It Ring Any Bells?

Good Day To All! Any Clue How To Make Upload Dataset To Minio S3? I Have Setup A Custom S3 In Clearml.Conf As Per Docs:

Hello Community! Is There An Option To Only Download A Part Of A Dataset With .Get_Local_Copy()? I Imagine Something Like This, But I Can'T Find The Right Way To Do It.

Is Setting Up Clearml Server On Same Machine That Is Supposed To Be Our Gpu Worker A Good Idea? I Wanted To Have Every Dataset Locally Using Clearml-Data Cli, How Would Gpu Worker Handle Data Fetching, Would It Go Straight From Disc?