How Can I Stop Clearml From Uploading Temporary Models? I Am Running Cross_Validation, Training A Bunch Of Models In A Loop Like This:

Answered

How can I stop clearml from uploading temporary models?

I am running cross_validation, training a bunch of models in a loop like this:

models = []
for X_train, Y_train, X_test, Y_test in cv:
  model = TempModel()
  model.fit(X_train, Y_train)
  model.eval(X_test, Y_test)
  models.append(model)

final_model = get_final_model(models)
joblib.dump(final_model, 'model.joblib')

What happens: ClearML uploads each model from inside the loop
What I want: only upload the model I explicitly dump, the final model

How can I control this behaviour?

  				
Posted 
	one year ago

					More
				  		
  Report
		
					AdventurousButterfly15
				
					0
					 × 1

Votes Newest

Answers 4

Hi @<1523702786867335168:profile|AdventurousButterfly15>

I am running cross_validation, training a bunch of models in a loop like this:

Use the wildcard or disable all together:

task = Task.init(..., auto_connect_frameworks={"joblib": False})

You can also do

task = Task.init(..., auto_connect_frameworks={"joblib": ["realmodelonly.pkl", ]})

  				
Posted 
	one year ago

					More
				  		
  Report
		
					AgitatedDove14
				
					0
					 × 1

@<1523701205467926528:profile|AgitatedDove14> thanks!

  				
Posted 
	one year ago

					More
				  		
  Report
		
					AdventurousButterfly15
				
					0
					 × 1

"realmodelonly.pkl"

should be the full path, or just the file name?

  				
Posted 
	one year ago

					More
				  		
  Report
		
					AdventurousButterfly15
				
					0
					 × 1

should be the full path, or just the file name?

just file name, this is basically fname matching

  				
Posted 
	one year ago

					More
				  		
  Report
		
					AgitatedDove14
				
					0
					 × 1

Write your answer

1K Views

4 Answers

one year ago