Seth Goodman

Finally had some time to come back to this - stripped out the "latitude correction" stuff I had accidentally merged into PR branch - kept the `like` arg in the...

triggering CI rebuild

@davidabek1 - vectorizing for exact intersections will not have any flaws in terms of accuracy, but will be more costly in terms of computations. For a few small geometries, like...

i would track each type and then let the user decide how they are returned. group them by default with an option to return them separately

looks like the non overlapping areas get lumped into nodata due to the the boundless options (`boundless_array` func in rasterstats for arrays, and equivalent steps in rasterio for rasters) which...

this might work: if the user wants no_overlap count returned, they need to provide a second, unused "fake" nodata value. this value is passed to rasterio as the raster's nodata...

PR #146 adds 'nan' stat option

rasterio issue on shapes extending beyond raster extent https://github.com/mapbox/rasterio/issues/995

@perrygeo almost done with a no overlap stat proof of concept that works with array rasters. but i think there may be some limitations to getting it working with rasters...

I think @jacobwhall inadvertently broke this failing example by fixing the OSM feature, haha. One of us will find another feature that fails and update shortly