Plotly express choropleth bins

Question

I am using plotly express to create a choropleth chart. I want to have a discrete colour scale using these bins: [−50,0) , [0,50), [50,100), [100,150), [150,200), [200,250), [250,300), [300,350), [350,400), [400,450), and [450,500). Below is a snippet of my dataframe.

    Country       1990   2019   Code    PGR
0   Afghanistan    12.4  38.0   AFG     206.451613
1   Albania        3.3   2.9    ALB    -12.121212
2   Algeria        25.8  43.1   DZA     67.054264

I can get the plot to display but I can't figure out how to get my colour set. My code so far is:

 fig = px.choropleth(popCodes,locations= popCodes["Code"],
                     color = popCodes["PGR"],
                     range_color = (-50, 500), 
                     hover_name = popCodes["Country"],
                     color_discrete_sequence=px.colors.sequential.Plasma)

 fig.show()

Answer 1

Not as tidy as Rob Raymond's answer but this is what I did. I'd use Rob's if I were you

popCodes['PGR'] = popCodes['PGR'].astype(float)

# Use this to bin the data and apply discrete colour codes later
conditions = [
    (popCodes['PGR'] >= -50) & (popCodes['PGR'] <0),
    (popCodes['PGR'] >= 0) & (popCodes['PGR'] <50),
    (popCodes['PGR'] >= 50) & (popCodes['PGR'] <100),
    (popCodes['PGR'] >= 100) & (popCodes['PGR'] <150),
    (popCodes['PGR'] >= 150) & (popCodes['PGR'] <200),
    (popCodes['PGR'] >= 200) & (popCodes['PGR'] <250),
    (popCodes['PGR'] >= 250) & (popCodes['PGR'] <300),
    (popCodes['PGR'] >= 300) & (popCodes['PGR'] <350),
    (popCodes['PGR'] >= 350) & (popCodes['PGR'] <400),
    (popCodes['PGR'] >= 400) & (popCodes['PGR'] <450),
    (popCodes['PGR'] >= 450) & (popCodes['PGR'] <500)
    ]
values = ['[−50,0)', '[0,50)', '[50,100)', '[100,150)', '[150,200)', 
'[150,200)', '[250,300)', '[300,350)', '[350,400)', '[400,450)', '[450,500)']
popCodes['range'] = np.select(conditions, values)




popCodes['PGR'] = popCodes['PGR'].astype(str)     # converting PGR to string gets plotly express to use discrete colours apparently
popCodes = popCodes.sort_values('range')          # sorts the dataframe by the range so I get a better legend later

fig = px.choropleth(popCodes, locations= popCodes["Code"],
                    color = popCodes["range"],                       # PGR calculated from dataframer
                    hover_name = popCodes["PGR"],                    # column to add to hover information
                    color_discrete_map={"[−50,0)": "maroon",         # discrete colours based on the range of PGR from above
                                        '[0,50)':'red',
                                        '[50,100)':'orange',
                                        '[100,150)':'yellow',
                                        '[150,200)':'lime',
                                        '[200,250)':'green',
                                        '[250,300)':'aqua',
                                        '[300,350)':'teal',
                                        '[350,400)':'blue',
                                        '[400,450)':'navy',
                                        '[450,500)':'purple'
                                       }
                   )

fig.show()

Answer 2

• it appears there is a very similar data set on kaggle. Have sourced it. Does not include PGR so calculated it as sum of yearly columns
• given you want discrete bins, simplest way is to use https://pandas.pydata.org/docs/reference/api/pandas.cut.html
• pandas intervals are not supported in plotly, so converted to string .astype(str)
• given you are using plotly express it's simpler to refer to columns by name rather than always passing a series
• complete code below

import kaggle.cli
import sys, requests
import pandas as pd
from pathlib import Path
from zipfile import ZipFile
import urllib
import plotly.express as px

# fmt: off
# download data set
url = "https://www.kaggle.com/mohaiminul101/population-growth-annual"
sys.argv = [sys.argv[0]] + f"datasets download {urllib.parse.urlparse(url).path[1:]}".split(" ")
kaggle.cli.main()
zfile = ZipFile(f'{urllib.parse.urlparse(url).path.split("/")[-1]}.zip')
dfs = {f.filename: pd.read_csv(zfile.open(f)) for f in zfile.infolist()}
# fmt: on

popCodes = dfs["world_population_growth.csv"]
popCodes["PGR"] = popCodes.select_dtypes("number").sum(axis=1)
popCodes = popCodes.sort_values("PGR")

px.choropleth(
    popCodes,
    locations="Country Code",
    color=pd.cut(popCodes["PGR"], bins=range(-50, 501, 50)).astype(str),
    hover_name="Country Name",
    hover_data={"PGR":":.1f", "Country Code":True},
    color_discrete_sequence=px.colors.sequential.Plasma,
)