You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This commit was created on GitHub.com and signed with GitHub’s verified signature.
The key has expired.
Added
Web app for visualizing SelfAwareData file trees (#23).
Accessible on the command line via datatc_app <project>.
Requires additional dependencies. Install with pip install datatc[app].
SelfAwareData tracks Intermediate File Steps (#23). When you save a file to the file system and reload it, it will track that intermediate file as part of the transformation history.
New public interface: MagicDataInterface (#22). Use dtc.MagicDataInterface.save(data, 'file_path') and dtc.MagicDataInterface.load('file_path') to take advantage of easy file saving and loading outside of DataDirectory.
Changed
DataDirectory replaces DataManager as the public interface for initializing and interacting with a data directory (#21).
Instead of dm = DataManager('project') use dd = DataDirectory.load('project').
Note: DataManager is deprecated, but will still work.
Fixed
Fixed bug where an empty SAD Directory was left on the file system if SelfAwareData.save() failed during creation (#25). This bug caused a further error: DataDirectory would fail to initialize due to not being able to find a data file for the SAD Directory.
Added option to pass kwargs to SelfAwareData.load_from_file() (#24).
Importing SelfAwareData from datatc yielded a different type id than the SelfAwareData that is type checked against internally (specifically, what is type checked in DataDirectory.save(), when determining how to save a SAD object) (#23). Fixed with relative imports.