rasterframes
rasterframes copied to clipboard
Geospatial Raster support for Spark DataFrames
Python documents have a number of places where it's used, and it's too much ceremony that we can eliminate.
There are cases where before (and after?) reading a `RasterSource` one would want to know how big it is (dimensions as well as footprint?).
Trying to read a hadoop scheme with gdal installed such as `s3a://${bucket}/${prefix}.TIF` produces an error such as below message/stack trace. This seems consistent with documented use of HDFS versus GDAL...
@vpipkt Sometime before 1.0, we should change these signatures to use the `var: Type` ordering, as that's what type hints in Python 3.7 use (along with Scala). That said, really...
Even with gdal properly configured... I seem to (on the notebook side) get silent errors with reading GDAL sources of the `gdal://vsis3` variety. Here's what i see in the `docker...
min is max double max is min double mean is NaN
So this guy only takes one param but it is documented (in reference.pymd) as if it takes 3. Not sure what the intent is. I would think we should have...
In trying to work with `pyproj` library within the rf-notebooks docker container we see error messages like this on the notebook server stdout: proj_create: Open of /opt/conda/share/proj failed proj_create: init=epsg:/init=IGNF:...