mapclassify icon indicating copy to clipboard operation
mapclassify copied to clipboard

integrate `legendgram` into `mapclassify`?

Open jGaboardi opened this issue 1 year ago • 1 comments

xref https://github.com/pysal/legendgram/issues/21

jGaboardi avatar Jul 13 '24 22:07 jGaboardi

This was briefly discussed in https://github.com/pysal/mapclassify/pull/216 and I guess that it could simply end up as an extension of that function, no?

martinfleis avatar Jul 14 '24 18:07 martinfleis

So considering #234 & #250:

  1. Can this issue be closed?
  2. Should we begin ramping down legendgram for archival?

jGaboardi avatar Jun 13 '25 01:06 jGaboardi

yes, yes

only question is whether we want to keep both plot_legendgram and plot_histogram since we don't really need the latter anymore

knaaptime avatar Jun 13 '25 02:06 knaaptime

yes, yes

only question is whether we want to keep both plot_legendgram and plot_histogram since we don't really need the latter anymore

If we don't need, let's go ahead and dump it?

@sjsrey @martinfleis – thoughts here?

jGaboardi avatar Jun 14 '25 01:06 jGaboardi

The ci is failing, I believe this has to do with something our dependencies changing so that the image checks in legendgram are failing in most cases.

sjsrey avatar Jun 14 '25 14:06 sjsrey

The RMS tolerance is very small, so I'd bet it's simply differences in matplotlib installs/versions. I have experienced this previously when installing matplotlib via PyPI vs. Conda-forge

jGaboardi avatar Jun 14 '25 15:06 jGaboardi

yes, yes only question is whether we want to keep both plot_legendgram and plot_histogram since we don't really need the latter anymore

If we don't need, let's go ahead and dump it?

@sjsrey @martinfleis – thoughts here?

I would keep plot_histogram around.

sjsrey avatar Jun 14 '25 15:06 sjsrey

OK, so I'll go ahead and close this out and we can reopen if needed.

jGaboardi avatar Jun 18 '25 22:06 jGaboardi