Hi There, I Have A Batch Prediction Task That Load A Model Published On Clearml.

Answered

Hi there,
I have a batch prediction Task that load a model published on ClearML.
input_model = InputModel(model_id=model_id) model_path = input_model.get_local_copy()The problem is that every time it finish the batch prediction, ClearML upload the model again (see the image)
This creates a lot of entries in the models page.
How can I avoid this?

  				
Posted 
	one year ago

					More
				  		
  Report
		
					IrritableGiraffe81
				
					0
					 × 1

Votes Newest

Answers 2

Hi IrritableGiraffe81
Can you share a code snippet ?
Generally I would try
task = Task.init(..., auto_connect_frameworks={"pytorch': False, 'tensorflow': False)

  				
Posted 
	one year ago

					More
				  		
  Report
		
					AgitatedDove14
				
					0
					 × 1

Thanks Martin, your suggestion solves the problem.
👍

  				
Posted 
	one year ago

					More
				  		
  Report
		
					IrritableGiraffe81
				
					0
					 × 1

Write your answer

956 Views

2 Answers

one year ago