I wish to store thousands of datasets of different sizes in a single HDF5 file. Almost all the attributes that describe the datasets are identical, except for size and another attribute. I would rather avoid repeatedly storing the same attributes with the same values. Is there a way for datasets to inherit (most of) its attributes from another dataset in the same file? Thank you.
A popular approach is to attach common attributes to the root group or the containing group, rather than replicate through individual data sets. Consuming applications would need to understand the relationship.