Shreyas Gadgin Matha

Answered few more questions in FAQs.md

export_png() should not require external program to save as png

5

`export_png()` requires vega to be installed. This should be a feature that is built into ggvis, especially because so many users ([1](http://stackoverflow.com/questions/25999741/getting-ggvisexport-png-working), [2](http://stackoverflow.com/questions/27002277/ggvisexport-png-on-windows), [3](http://stackoverflow.com/questions/29035628/how-to-save-a-ggvis-plot-displayed-in-the-viewer-panel-of-rstudio-with-an-r-scri)) have faced difficulties in installing vg2png...

When axis labels are rotated, ggvis should automatically move the axis title to avoid overlaps with labels. This is what we would want: ``` ggplot(iris, aes(x = Species,y = Sepal.Length))...

Simple port to Python 3

Minor documentation issue in `blackmarble.raster.bm_raster`

1

## Bug Report ### Description `blackmarble.raster.bm_raster` takes parameter `gdf`, but the API reference mentions `roi` instead. #### Screenshots / Error Messages (if applicable) 

RPOP - Handling zeros in diversity / ubiquity

2

Diversity / ubiquity is often 0 when calculating with RPOP. Need better handling of these cases, since ECI / PCI show up as NaN for all countries in these cases...

Added comments to briefly explain why PCI values are not normalized with PCI mean and std deviation.

If columns called "mcp" or "eci" or whatever already exist and the column label dict is provided like this {"time": "year","loc": "exporter","prod": "commodity_code","val": "mcp",}, then there's a conflict and the...

Using knn option gives integer density outputs

1

Densities should be float