Hello All, I Am Trying To Report A Confusion Matrix For My Output Model As Follows:

Answered

Hello all,

I am trying to report a confusion matrix for my Output Model as follows:

output_model.report_confusion_matrix(title = "Validation Confusion Matrix", 
                                        series = "Validation Data", 
                                        matrix = best_total_cm, 
                                        xaxis = "Predicted Labels", 
                                        yaxis = "True Labels", 
                                        xlabels = list(classes.keys()), 
                                        ylabels = list(classes.keys()),)

However, when I update the weights later in the script I get the following warning:

2023-07-18 14:15:34,407 - clearml.Metrics - ERROR - Action failed <400/131: events.add_batch/v1.0 (Events not added: Event must have a 'task' field=1)>

The weights successfully upload but the confusion matrix cannot be seen on the clearml web app. Does anyone know what could be causing this?

  				
Posted 
	one year ago

					More
				  		
  Report
		
					ThoughtfulSeahorse27
				
					0
					 × 1

Votes Newest

Answers 6

What does it look like when you instantiate the output_model object?

  				
Posted 
	one year ago

					More
				  		
  Report
		
					HarebrainedToad56
				
					0
					 × 1

That looks good to me, not sure

  				
Posted 
	one year ago

					More
				  		
  Report
		
					HarebrainedToad56
				
					0
					 × 1

Ok, thanks for trying

  				
Posted 
	one year ago

					More
				  		
  Report
		
					ThoughtfulSeahorse27
				
					0
					 × 1

Hi @<1592326527742119936:profile|ThoughtfulSeahorse27> , this indeed seems like a bug - can you please open a GitHub issue? 🙂

  				
Posted 
	one year ago

					More
				  		
  Report
		
					SuccessfulKoala55
				
					0
					 × 1

Reporting the confusion matrix AFTER updating the weights fixed the issue.

  				
Posted 
	one year ago

					More
				  		
  Report
		
					ThoughtfulSeahorse27
				
					0
					 × 1

task = Task.init(project_name=params['project name'], 
                    task_name= "-".join([params['task name'], params['net']['model'], 'lr', str(params['opt']['lr'])]),
                    auto_connect_frameworks={"tensorflow":False,"tensorboard": False,'pytorch': False,})
    

output_model = OutputModel(task=task, 
                        framework="PyTorch", 
                        config_dict = params,
                        tags = [params['project name']] + params['tags'],)

  				
Posted 
	one year ago

					More
				  		
  Report
		
					ThoughtfulSeahorse27
				
					0
					 × 1

Write your answer

1K Views

6 Answers

one year ago