I have created geo-dataframe using a combination of geopandas and geoviews. Libraries I'm using are below:
import pandas as pd
import numpy as np
import geopandas as gpd
import holoviews as hv
import geoviews as gv
import matplotlib.pyplot as plt
import matplotlib
import panel as pn
from cartopy import crs
gv.extension('bokeh')
I have concatenated 3 shapefiles to build a polygon picture of UK healthcare boundaries (links to files provided if needed). Unfortunately, from what i have found the UK doesn't produce one file that combines all of those, so have had to merge the shape files from the 3 individual countries i'm interested in. The 3 shape files have a size of:
shape file 1 = 5mb (https://www.opendatani.gov.uk/dataset/department-of-health-trust-boundaries)
shape file 2 = 204kb (https://geoportal.statistics.gov.uk/datasets/5252644ec26e4bffadf9d3661eef4826_4)
shape file 3 = 22kb (https://data.gov.uk/dataset/31ab16a2-22da-40d5-b5f0-625bafd76389/local-health-boards-december-2016-ultra-generalised-clipped-boundaries-in-wales)
I have merged them all successfully to build the picture i am looking for using:
Test = gv.Polygons(Merged_Shapes, vdims=[('Data'), ('CCG_Name')], crs=crs.OSGB()).options(tools=['hover'], width=550, height=700)
Test_2 = gv.Polygons(Merged_Shapes, vdims=[('Data'), ('CCG_Name')], crs=crs.OSGB()).options(tools=['hover'], width=550, height=700)
However, I would like to include these charts in a shareable html file. The issue I'm running into, is that when I save the HTML using:
from bokeh.resources import INLINE
layout = hv.Layout(Test + Test_2)
Final_report = pn.Tabs(('Test',layout)).save('Map_test.html', resources=INLINE)
I generate a html file that displays the charts, but the size is 80mb, which is far to large, especially if I want include more polygon charts and other charts in the same html.
Does anyone know of a more efficient way, from a memory perspective, I can store my polygon charts within a HTML file for sharing?
You can make the file smaller by rasterizng or by decimating the shapes. For rasterizng you can call hv.operation.datashader.rasterize(obj), and I think there is something in Shapely or GeoPandas for simplifying the shapes.