pandas2
pandas2 copied to clipboard
wesm
Design documents and code for the pandas 2.0 effort.
There are a number of places where we "guess" a type (e.g. `np.float64`) where there is no reasonable choice, e.g. in the CSV parser code. Many databases have the notion...
I had started building some things here because it was difficult several months ago to link to the Python bridge between NumPy/pandas land and C++-only land. Now that libarrow_python is...
@jorisvandenbossche the automated docs builds here don't have the requisite Python packages to run the IPython directives, it seems. We can revert to manual builds if necessary...
xref https://github.com/pandas-dev/pandas/issues/14052 currently we have an (implicity) numpy conversion when we access ``.values`` of a 1D (Series). This *mostly* returns a numpy array, though we do return numpy-like objects several...
Do we want to continue to use NaN? There are possible computational benefits to doing so, but we have the opportunity with pandas 2.0 to change to using bitmaps everywhere,...
Given that we seem to be on board with only supporting Python 3 (#5) in pandas 2.0, let's take full advantage of that and require [type annotations](https://docs.python.org/3/library/typing.html) for all internal...
looks ``scipy`` encapsulated lots of function type things here: https://github.com/scipy/scipy/pull/6509 we eventually will want to enable things like this as well for ufunc like things that can be processed efficiently...
This will serve as a master issue to record API changes / testing that needs segregation in pandas. Copy-on-write can help here: - [ ] view propagation: https://github.com/pandas-dev/pandas/issues/14953
xref https://github.com/pandas-dev/pandas/issues/9220 prob not going to fully support this is current pandas, its not a lot of work, but I c limited utility here, also not supported on all platforms...
I'm thinking we can come up with a plan to yield a better .append implementation that defers stitching together arrays until it's actually needed for computations. We can do this...