jupyter on home laptop

• jupyter already installed

    > pip3.7 install ipyleaflet
    Installing collected packages: traittypes, branca, ipyleaflet
    Successfully installed branca-0.4.1 ipyleaflet-0.13.3 traittypes-0.2.1

here goes:

    cd ~/gis/jup_notebook
    jupyter notebook

it prints out this, which opens in Firefox:

    http://localhost:8888/tree?token=2ead9ee867947dba3149b21ecf67993e7ff7ab3cbee2dcf3

then:

    New -> Python3

and following sample from here:

    https://ipyleaflet.readthedocs.io/en/latest/usage.html

it works FINE, and a map shows up!

read & display geopkg

import os
os.chdir("/home/wendell/cssi/feed/netcdf/from_rserver")

import geopandas as gpd
#aa = gpd.read_file("the_first.gpkg")
bb = gpd.read_file("the_second.gpkg")

first_data = GeoData(geo_dataframe = bb,
                   style={'color': 'purple', 'opacity':3, 'weight':1.9, 'dashArray':'2', 'fillOpacity':0.6},
                   hover_style={'fillColor': 'red' , 'fillOpacity': 0.2},
                   name = 'WxPolys')

m.add_layer(first_data)
m.add_control(LayersControl())
m

read & display geojson of VIL and et

retrieve from webfaction:

$ scp wendell@ilikecarrots.com:cssi/from_rserver/echo_tops/ciws/files/QuantizedVIL.20200829T193500Z.gjson .
$ scp wendell@ilikecarrots.com:cssi/from_rserver/echo_tops/ciws/files/QuantizedEchoTop.20200829T193500Z.gjson .

in jupyter:

import os
os.chdir("/home/wendell/cssi/feed/grid_testing")

import geopandas as gpd
cc = gpd.read_file("QuantizedVIL.20200829T193500Z.gjson")

from ipyleaflet import Map, Marker
center = (36.085808, -79.484851)
m = Map(center=center, zoom=15)
marker = Marker(location=center, name="Burlington")
m.add_layer(marker)
display(m)

from ipyleaflet import GeoData, LayersControl
first_data = GeoData(geo_dataframe = cc,
                   style={'color': 'purple', 'opacity':3, 'weight':1.9, 'dashArray':'2', 'fillOpacity':0.6},
                   hover_style={'fillColor': 'red' , 'fillOpacity': 0.2},
                   name = 'vil')

m.add_layer(first_data)
m.add_control(LayersControl())
m

animation

very LAME notebook: notebook

interesting class: Visualize and publish with Python

and these::

Best Libraries for Geospatial Data Visualisation in Python

1. PyViz/HoloViz (Geoviews, Datashader, HvPlot)

   • GeoViews
   • Interactive Geospatial Data Visualization with Geoviews in Python really good map on startup page
   • Datashader
   • Panel a holoviz thing

2. Folium

   • Creating interactive crime maps with Folium, 2015
   • Your Guide to Getting Started with Geospatial Analysis using Folium (with multiple case studies) lots of pop-ups and ads.

3. Plotly/Plotly Express

   • Dash is the most downloaded, trusted framework for building ML & data science web apps. ads, banners

4. kepler.gl for Jupyter User Guide

   • deck.gl a WebGL-powered framework for visual exploratory data analysis of large datasets.
   • Kepler.GL & Jupyter Notebooks : Geospatial Data Visualization with Uber’s opensource Kepler.GL

5. IpyLeaflet

6. GeoPandas

However, recent advances and additions of Contextily for base maps and IPYMPL for interactive matplotlib plots makes it straightforward to create interactive maps with Geopandas.

unknown: Basic Interactive Geospatial Analysis in Python, 2016

notes: python + movement + display (leaflet, rshiny, ???)

*  folium

very interesting: folium

notebook examples

Spatial Visualizations and Analysis in Python with Folium

a reference to HeatMapWithTime(), possibly interesting...

*  ipympl (matplotlib)

* [Leaflet TimeDimension](https://github.com/socib/Leaflet.TimeDimension)

javascript-only, and only for GeoJson, but looks promising

Leaflet TimeDimension Examples Add time dimension capabilities on a Leaflet map

one of them looks good, (example 8), but still just javascript displaying geojson

most promising: folium + TimestampedGeoJson

class folium.plugins.TimestampedGeoJson

docs page

folium/plugins/timestamped_geo_json.py

SO question/answer: https://stackoverflow.com/questions/60044777/folium-plugins-timestampedgeojson-time-not-avaliable

another SO: TimestampedGeoJson does not plot entire image #1235

which is an i/f to the leaflet (javascript) Leaflet TimeDimension

on faa laptop:

installing folium:

    $ pip3.7 install branca
    $ pip3.7 install jinja2
    $ pip3.7 install folium
    Collecting folium
      Downloading folium-0.11.0-py2.py3-none-any.whl (93 kB)
    Installing collected packages: folium
    Successfully installed folium-0.11.0

Some functionalities may require extra dependencies altair, etc. (???)

start jupyter:

    cdmarc
    cd folium
    $ jupyter notebook --no-browser

1) starting from example here:

SO question/answer

2) or better yet, this one: Scraping USA Snow Storm Data and visualizing them using Folium

    $ mkdir Data_Script
    $ mkdir unzipped
    $ mkdir shapes

run parts 1,2,3

    cd shapes

    #!/bin/bash -x
    for F in ../Data_Script/snow*.tar.gz ; do
        tar zxvf $F
    done

NOPE: missing filtered_data.csv file!!!

3) or this example:

TimestampedGeoJson does not plot entire image

worked, and was able to reproduce the incorrect diagram :-(

4) this one:

TimestampedGeoJson duration parameter causes polygons to disappear