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
Hi Guys, I'M Trying To Familiarize Myself With Hyperparameter Optimization Using Clearml. It Seems Like There Is A Discrepancy Between

Hi guys,
I'm trying to familiarize myself with Hyperparameter Optimization using ClearML.
It seems like there is a discrepancy between clearml-param-search CLI tool and the HyperParameterOptimizer class; the params-override flag is apparently only available in the CLI tool and not in the HyperParameterOptimizer class.

Is there a way to override some parameters in the base task that we optimize with the class approach, say changing the number of epochs?

Edit: I'm on ClearML v1.6.2

  
  
Posted one year ago
Votes Newest

Answers 4


Hi GiganticMole91 . You could use something like
` from clearml.automation import DiscreteParameterRange

HyperParameterOptimizer(
...,
hyper_parameters=[DiscreteParameterRange("epochs", values=[100]), ...] # epochs is static, ... represent the other params
) to get the same behaviour --params-override ` provides

  
  
Posted one year ago

Hi CostlyOstrich36
I have created a base task on which I'm optimizing hyperparameters. With clearml-param-search I could use --params-override to set a static parameter, which should not be optimized, e.g. changing the number of epochs for all experiments. It seems to me that this capability is not present in HyperParameterOptimizer . Does that make sense?

From the example on https://clear.ml/docs/latest/docs/apps/clearml_param_search/ :
clearml-param-search {...} --params-override '{"name": "epochs", "value": 30}'

  
  
Posted one year ago

Yeah, that makes sense. The only drawback is that you'll get a single point that all lines will go through in the Parallel Coordinates plot when the optimization finishes 🙂

  
  
Posted one year ago

Hi GiganticMole91 ,

Can you please elaborate on what are you trying to do exactly?

Doesn't HyperParameterOptimizer change parameters out of the box?

  
  
Posted one year ago
676 Views
4 Answers
one year ago
one year ago
Tags