Closed martinfleis closed 4 years ago
@ljwolf brought this up at the pysal-dev mtg and the consensus was that it was probably a better fit for mapclassify than splot.
Very nice!
Cool. I had the same feeling that it is not analytical enough. Looking at your notes, I do have some questions regarding your vision of the whole ecosystem though.
@ljwolf can you elaborate a bit your points from the meeting? How do you see relationship geopandas <-> legendgram, greedy, contextily...? Not sure if I understand the whole context.
@martinfleis would be happy to chat about the ecosystem linkages and how to integrate greedy. I think mapclassify is a good fit.
@sjsrey I agree with that and once I'll find some spare moment, I'll open an issue/PR there to discuss how exactly we should do that. I just did not have a time to do that yet.
I'm trying to remember my initial questions about the ecosystem, but looking at those notes, it is clear to me now :D.
I'll try to make it for the next release of PySAL (that is July, right?).
@martinfleis great. In terms of timing, it can go into mapclassify whenever it is ready. PySAL meta is released every six months with the next release set for July.
Moved to mapclassify, so closing
Hi,
I was exploring how to port tools from QGIS for geopandas which resulted in a port of topological colouring algorithms. I was initially thinking about including it in geopandas (https://github.com/geopandas/geopandas/issues/1165) but then released in separately as
greedy
- https://github.com/martinfleis/greedy (docs on https://martinfleis.github.io/greedy/).@ljwolf now suggested that it could be integrated into splot. I am happy to do that, rather than having a tiny independent package. The issue might be the license, as the original algorithm in QGIS in under GPL. But @nyalldawson was willing to relicense if needed.
Thinking about it in the context of pysal,
_geos_sw
be even go to libpysal.Check that and let me know if it makes sense for you. I can then make a PR if we manage to resolve the issues.