James Gaboardi
James Gaboardi
Putting this here for a bookmark / to start a discussion for potential to adopt more robust computational geometry solutions. * [`scikit-geometry`](https://github.com/scikit-geometry/scikit-geometry) * [SciPy 2020 talk](https://www.youtube.com/watch?v=LsgXJTK4OsU&list=PLYx7XA2nY5GfhekhCPLpA0NImFm2wNvSD&index=7&t=0s) See also pysal/pysal#1164 (`spatialpandas`)
Following #271, and other similar corrupted install issues in the past, it may be an opportune time to rethink the current implementations of [`@requires`](https://github.com/pysal/libpysal/blob/c103f43382493bb0694e90c31d4ae0513bb9c5b9/libpysal/common.py#L97), [`simport`](https://github.com/pysal/libpysal/blob/c103f43382493bb0694e90c31d4ae0513bb9c5b9/libpysal/common.py#L53), and the various cases to...
Following along pysal/tobler#50 and [pysal/segregation#96](https://github.com/pysal/segregation/issues/96#issuecomment-585448361) that @knaaptime raised regarding a new [pyproj](https://github.com/geopandas/geopandas/pull/1292/files) release, we may want to consider an embedded utility function that that checks for differing CRS of datasets.
It appears that `numpy` is overriding `libpysal.cg.Point` objects and converting directly into `array`s when storing the points in a `numpy.array`. I came across this issue (pysal/spaghetti#378) when implementing improved support...
The `test_silhouette.py::Silhouette_Tester::test_silhouette_alist` test has been failing for quite some time. See [here](https://github.com/pysal/esda/runs/6810288316?check_suite_focus=true#step:4:295).
Follow example in [`spopt`](https://github.com/pysal/spopt/blob/e76d84efb0d688228b0cf0e8ab7436fc2f7b9f97/docs/conf.py#L283-L296) [suggested](https://github.com/pysal/libpysal/issues/417#issuecomment-869226527) by @martinfleis for getting the links back working in the rendered tutorials to the original notebooks.
[Gitly](https://gitly.org) is a new [open-source alternative to GitHub](https://github.com/vlang/gitly) written in [`V`](https://vlang.io). It is currently in pre-alpha, but may be something we should consider in the intermediate to long-term future for...
## Standard docstring formatting conventions proposal 0. Based on [`numpy`](https://numpydoc.readthedocs.io/en/latest/format.html). 1. Summary starts directly after triple quotes on first line. 2. Empty line between all sections. 3. Valid sections are...
[`spatialpandas`](https://github.com/holoviz/spatialpandas) seems to be an interesting package that functions much like `geopandas`, but without `GEOS`, `GDAL`, etc. dependencies. It may be worth keeping an eye on.