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, I'M Currently Working With Clearml Pipelines And Would Like To Clarify Whether It'S Officially Supported To

Hi, I'm currently working with ClearML Pipelines and would like to clarify whether it's officially supported to invoke a sub-pipeline from within another pipeline , especially when running multiple variations in parallel. Is it possible to call one pipeline (e.g., sub_pipeline ) from within another pipeline (e.g., main_pipeline ) while:

  • Not manually specifying a base_task_id
  • Running multiple sub-pipelines in parallel
  • Collecting the results after all sub-pipelines finishWhen executing this, we encounter the following error: ValueError: Node 'sub_pipeline', base_task_id is empty
  
  
Posted 3 months ago
Votes Newest

Answers 2


Hi @<1603198163143888896:profile|LonelyKangaroo55> , I don't believe this is supported.

  
  
Posted 3 months ago

@PipelineDecorator.pipeline(
name="Sub Pipeline",
project="Pipelines",
version="1.0",
multi_instance_support="parallel",
)
def sub_pipeline(parameter):
print(f"Running sub-pipeline with parameter={parameter}")
return parameter * 2

@PipelineDecorator.pipeline(
name="Main Pipeline",
project="Pipelines",
version="1.0",
)
def main_pipeline():
refs = []
for p in [1, 2, 3]:
ref = sub_pipeline(parameter=p)
refs.append(ref)

PipelineDecorator.wait_for_multi_pipelines()
# Aggregate results from all sub-pipelines

  
  
Posted 3 months ago
309 Views
2 Answers
3 months ago
3 months ago
Tags