Standardization of names and data types
bjorkqvi opened this issue · comments
We need to create a more standard naming convention. Also we need to better define which datatypes are used where, and what we are passing through the interfaces. This also includes type hinting in the methods.
Times are now either in strings, numpy datetime64, or pandas DatetimeIndex. In practice the user uses only strings, which is probably what we want to continue with.
Data is now mostly stored in xarrays, which is probably what we want. However, not all properties still have method calls to access them. I think that the user shouldn't need to access the underlying xr dataset directly at any point.