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
Unanswered
Has Anyone Used Dynaconf With Clearml? Trying To Decide Whether To Migrate To Hydra Or Stick With Dynaconf. Would Love To Take Advantage Of Automatic Logging Of The Hyperparameters


right now I'm doing

# clear existing configs in case we're rerunning the notebook
if hydra.core.global_hydra.GlobalHydra().is_initialized():
    hydra.core.global_hydra.GlobalHydra.instance().clear()

# initialize Hydra
hydra.initialize(
    version_base=None,
    config_path=".",
    job_name="test_app",
)

# use the compose API since we're running in a notebook instead of __main__
cfg = hydra.compose(
    config_name="config", 
    overrides=[],
)

# report in ClearML Config UI
clearml.binding.hydra_bind.PatchHydra._register_omegaconf(cfg)

Which works for my purposes. Not sure if there's a good way to automate it - I'll just make it part of my team's internal library for setting up ClearML and assigning names to experiments, etc

  
  
Posted one year ago
149 Views
0 Answers
one year ago
one year ago