You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It seems that most of this works together rather easily, given more arbitrary W procedures. With the proposed changes in wconstructor for geointerfaces, a pysal.weights object could be constructed from any GeoSeries.
I've been playing around with what's possible given this W from GeoPandas objects. esda gains instant support, as calling geodataframe[column].values yields a (n,) numpy vector, so every pysal.Moran* works using Series' .values methods. Compatibility holds using this in spatial_dynamics, too.
We still run into issue #461 for spreg, as gdf[col].values has to be reshaped for y but not X. The fact that pandas provides column vectors as (n,) might indicate a good external standard to peg on for this issue.
That said, once the new wconstructor gets tested, doing this might be better off as a (highly visible) set of example notebooks.
https://github.com/kjordahl/geopandas
The text was updated successfully, but these errors were encountered: