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
I Have A Question About Accessing The Parameters From Hydra Omegaconfig. I Am Trying To Follow The Hyperparam Optimization Example, And There It Is Written

I have a question about accessing the parameters from Hydra Omegaconfig.
I am trying to follow the hyperparam optimization example, and there it is written
# here we define the hyper-parameters to optimize     # Notice: The parameter name should exactly match what you see in the UI: <section_name>/<parameter>     # For Example, here we see in the base experiment a section Named: "General"     # under it a parameter named "batch_size", this becomes "General/batch_size"     # If you have argparse for example, then arguments will appear under the "Args" section,     # and you should instead pass "Args/batch_size"My question is how to access the OmegaConf params to set the hyperparams. The UI has a "OmegaConf" section, but I tried to pass "OmegaConf/model/learning_rate" and "OmegaConf/model.learning_rate" but I get
clearml.automation.optimization - WARNING - Could not find requested hyper-parameters ['OmegaConf/model.learning_rate', 'OmegaConf/model.dropout', 'OmegaConf/model.epochs'] on base taskWhat is the way to access them?

  
  
Posted 3 years ago
Votes Newest

Answers 5


YummyLion54 hi!
are you referring to PARAMETERS  section OR to the  CONFIGURATION OBJECTS

  
  
Posted 3 years ago

the OmegaConf under CONFIGURATION OBJECTS

  
  
Posted 3 years ago

will give it a try tomorrow thank you!

  
  
Posted 3 years ago

YummyLion54 , please try the following:

` from clearml import Task

base_task = Task.get_task(task_id="base task id")
cloned_task = Task.clone(source_task=base_task)
cloned_task.connect_configuration(name="OmegaConf", configuration="path/to/conf/file") `

  
  
Posted 3 years ago

CostlyOstrich36 I already set allow_omegaconf_edit to true, and I could edit the hydra conf in the UI when cloning the original task, but now I want to update it in the code for hyperparam optimization

  
  
Posted 3 years ago