Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Answered
Hello Everyone, Is It Possible To Perform A Optimization Task Based On An Executed Task? In The Pipeline I Handover The Task Id Of An Executed Training Task To The Optimization Task (E.G. Parameter_Override='General/Taskidtrainyolov5': '${Trainyolov5.Id}

Hello everyone,

is it possible to perform a optimization task based on an executed task?
In the pipeline I handover the task id of an executed training task to the optimization task
(e.g. parameter_override='General/taskIdTrainYolov5': '${trainYolov5.id}'}
During the execution of the optimization task I get the following error:
clearml.backend_interface.session.SendError: Action failed <400/101: tasks.clone/v1.0 (Invalid task id: id=)> (task=, new_task_name=...)

  
  
Posted one year ago
Votes Newest

Answers 3


Hi @<1555000563244994560:profile|OutrageousSealion55> ! How do you pass base_task_id in the HyperParamterOptimizer ?

  
  
Posted one year ago

Hi @<1555000563244994560:profile|OutrageousSealion55> , can you please share the code you're trying to run?

  
  
Posted one year ago

Hey @<1523701087100473344:profile|SuccessfulKoala55> ,

thank you for your quick answer.

the code for the optimization task is nearly similar to the clearml example: None
Instead of the parameter name "template_task_id", I use "taskIdTrainYolov5".

In my pipeline I added the following step:

pipe.add_step(
    name="hyper_parameter_optimization_search_yolov5",
    parents=['train_yolov5'],
    base_task_project='***',
    base_task_name='Hyper-Parameter Optimization Models 1 Yolo5',
    execution_queue='task',
    parameter_override={
        'General/taskIdTrainYolov5': '${train_yolov5.id}',
        'General/pipeline': '${pipeline.pipeline}',
        'General/devMode': '${pipeline.devMode}',
    }
)

As you can see, I hand over the ID of the previouse pipeline step as the value for the parameter "taskIdTrainYolov5".

Inside the optimization task I print out the arguments and I can see, the right value for "taskIdTrainYolov5" is provided.

  
  
Posted one year ago