welly
welly copied to clipboard
agilescientific
Refactor curves
Hate to say it, but I think curves need to become either Pandas Series or xarray datasets (which are implemented via pandas). Essentially, all my basis stuff is just trying to reimplement indexing, and that what pandas does.
However, this isn't quite as simple as just making a DataFrame. I think we must continue to allow each curve to have its own index (i.e. be on a different sample interval, or have a different range).
ah finally I found what I need. I'm looking for info what is basis actually is. Based on your comments, apparently its just some indexing. I am also trying to find if we can put its own index (depth) to any arbitrary curves.
Right, it's indexing. Just implemented in a wonky way, compared to pandas. Each curve has its own. It is computed from start and step, and has the same length as the curve. (That is, it is not stored as an array but computed on the fly.)