Add here instructions for what the user need to do to use your code. List any dependencies and/or input files that might be needed or include instructrions for how the user can use, e.g., a custom conda environment that you may have included with your code.
(for example)
- numpy: a package to create and manipulate n-dimensional arrays
- *package_blah_blah: ...
Ideally there will be something like an environment.yml file (conda) or a requirements.txt (pip) in the repository
You should also discuss why you need specific dependencies and we would always take a minimslist view: only add what you need.
Give a short description of the tests and input validations you have included in your code.
Remember: tests are checks to see that you have not added bugs to your code since it was last run whereas validation means that you check the input data to make sure it is consistent with the expectations. Generally you need both.
Describes the limitations of your code. E.g., it only works with netCDF files because of this package that is used cannot read anything else. Any other limitations also in what users cannot do with this code. Or other things that you feel you could have done better?
Talk a bit about how you or somebody could take this code to the next level. How is the code generalizable to solve more general problems?