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
Playing Around With Hpo For First Time. I Am Giving This As Hyperparameter:

Playing around with HPO for first time.

I am giving this as hyperparameter:

DiscreteParameterRange("General/num_iterations", values=[10, 20, 30]),
Is it expected that the jobs are repeated?

  
  
Posted 2 years ago
Votes Newest

Answers 27


You mean the job with the exact same arguments ?
do you have other arguments you are passing ?
Are you using Optuna / HBOB ?

  
  
Posted 2 years ago

Optuna

  
  
Posted 2 years ago

You mean the job with the exact same arguments ?

Yes

  
  
Posted 2 years ago

It completed after the max_job limit (10)

  
  
Posted 2 years ago

Ok, just my ignorance then? ๐Ÿ™‚

  
  
Posted 2 years ago

The "Optimizer task" will continue to run as long as there are sub-Tasks it launched.
Is anything else running/pending ?

  
  
Posted 2 years ago

The Optimizer task is taking a lot of time to complete. Is it doing something here:

  
  
Posted 2 years ago

Tried context provider for Task?

I guess that would only make sense inside notebooks ?!

  
  
Posted 2 years ago

No, all of them completed!

  
  
Posted 2 years ago

how do you see things being used as the most normal way?

  
  
Posted 2 years ago

just that the task itself is still Running state

  
  
Posted 2 years ago

Yeah, Curious - is a lot of clearml usecases not geared for notebooks?

  
  
Posted 2 years ago

Something like:

with Task() as t: #train

  
  
Posted 2 years ago

Yeah, Curious - is a lot of clearml usecases not geared for notebooks?

That is somewhat correct, notebooks are not actually used with a lot of deep-learning projects as they require entire repository to support.
I guess generally speaking the workflow is, "test your code" (i.e. small scale with limited data), then clone and enqueue for remote execution.
That said, I think it will be great to expand the support.
TrickySheep9 I like the idea of context for Tasks, can you expand on how this is used as part of a (jupyter) workflow ?

  
  
Posted 2 years ago

no task.close()

  
  
Posted 2 years ago

I am running from noebook and cell has returned

  
  
Posted 2 years ago

Notice Optuna will do TPE & hyper band Bayesian optimization to find the best combination

  
  
Posted 2 years ago

So this is optuna ๐Ÿ™‚ the idea is it will test which parameters have potential (with early stopping), then launch a subset of the selected parameters

  
  
Posted 2 years ago

I am running from noebook and cell has returned

Well the Task will close when you shut down the notebook ๐Ÿ™‚

  
  
Posted 2 years ago

Ok, just my ignorance then?ย 

LOL, no it is just that with a single discrete parameter the strategy makes less sense ๐Ÿ™‚

  
  
Posted 2 years ago

The job itself doesnโ€™t have any other param

  
  
Posted 2 years ago

It completed after the max_job limit (10)

Yep this is optuna "testing the water"

  
  
Posted 2 years ago

Tried context provider for Task?

  
  
Posted 2 years ago

Ah ok thereโ€™s only optimizer.stop in the example

  
  
Posted 2 years ago

image

  
  
Posted 2 years ago

Only one param, just playing around

  
  
Posted 2 years ago
670 Views
27 Answers
2 years ago
3 months ago
Tags