GCS-Metadog is a CLI tool for m3dev/gokart.
Gokart solves reproducibility, task dependencies, constraints of good code, and ease of use for Machine Learning Pipeline.
And, gokart supports GCS as a data store for intermediate results of Tasks in the pipeline.
Then also, gokart is going to support easy searching GCS object from parameter name, task name, dependency of task.
GCS-Metadog is the interface for this purpose.
Thank you for reviewing my code, @kitagry, @hirosassa, @m3dev!
brew install TlexCypher/tap/gmd
When you want to search GCS object from metadata key only, you can use normal mode search. You can get full GCS object path list.
gmd -b ${bucket_name} -m ${metadata_key1} -m {metadata_key} .....
When you search and filter gokart GCS cache, sometimes, you want to search Task dependency. For example, you want to search some Tasks those have TaskA(parameter1=1, parameter2=True) as dependency. You can easily search with gmd by using nest mode. Parameter and Task names are case sensitive.
gmd -b ${bucket_name} -t ${dependency_task_name} -p {parameter_name1=parameter_value1} -p {parameter_name1=parameter_value1} .....
I plan to support adding multiple sets of -t and -p options in the future.