Agenda 24-09 about squidpy HOT 2 CLOSED

Notes:

• use type hints
• use scanpy-style docstrings
• write unit tests using pytest
• clean your own notebooks and keep plotting functions in the notebook
• cleanup / testing / documentation is done via issues -> assign yourself to the issue about the functions that you implemented

from squidpy.

Minutes:

• David and Hannah are working on data structure to handle tiff file
• For docstrings, let's use what Scanpy uses check contributing docs
• Use keywords (strings) for e.g. feature selection, potentially graph functions, allows for easier development
• for feature extraction: not everything, but cover maybe some more examples
• for graph: pretty much all done for graph space, still missing some spatial statistics that could be interesting
• @ALL open issues with your function that needs to be tested, together with testing, pls consider:
  • clean up notebooks explaining functions as well as a different way to use them
  • plotting functions as well in notebooks
  • clean up code
• Somebody needs to look into optimizing image feature part (maybe nacho)?
• for next week: unit tests. Make one test file for each python file in the package (whatever you implemented).

Meeting next week with unit tests !

from squidpy.