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
One More Strange Behaviour Of Clearml-Agent. Two Tasks With The Same Code (Except Usage Of Task.Execute_Remotely()) Generate Different Structure Of Output Artifacts. The Code Is A Regular Training With Keras That Includes Lines:

One more strange behaviour of clearml-agent. Two tasks with the same code (except usage of task.execute_remotely()) generate different structure of output artifacts. The code is a regular training with Keras that includes lines:
` model.save(f'{ARTIFACTS_ROOT}/mnist_model.h5')

task.upload_artifact('model_weights', artifact_object=f'{ARTIFACTS_ROOT}/mnist_model.h5') `When running locally artifact 'model_weights' is generated + model "mnist_model'. When running remotely only artifact 'model_weights' is generated (no model)

  
  
Posted 3 years ago
Votes Newest

Answers 9


clearml 1.0.2. clearml-agent was 0.17 yet, now I deleted it and installed 1.0.0

  
  
Posted 3 years ago

HelpfulHare30 can you share the ClearML version?

  
  
Posted 3 years ago

TimelyPenguin76 , rechecking this situation with clearml-agent 1.0.0 now...

  
  
Posted 3 years ago

TimelyPenguin76 , sorry I didn't see this comment. No. I mean that when I run task locally (from PyCharm and without task.execute_remotely()), model is uploaded and registered. But when I do the same with task.execute_remotely() and it runs on agent model cannot be found in the task after this. I speak about the same script I sent in the second thread

  
  
Posted 3 years ago

TimelyPenguin76 , the same behaviour with clearml-agent 1.0.0

  
  
Posted 3 years ago

TimelyPenguin76 Ok, when no explicit artifact upload is done, it indeed uploads model locally, but not remotely

  
  
Posted 3 years ago

HelpfulHare30 the model.save(f'{ARTIFACTS_ROOT}/mnist_model.h5') should save the model automatically, any special reason saving it manually with the upload_artifact ? Can you try just with model.save(f'{ARTIFACTS_ROOT}/mnist_model.h5') ?

  
  
Posted 3 years ago

As I remember, I added it because it was not added automatically. But I'll recheck now...

  
  
Posted 3 years ago

didn't get it, do you get the model in the task with local path?

  
  
Posted 3 years ago