slarrain / clchoropleth Goto Github PK

It creates a SVG Choropleth of a Chilean region.

You need a Dictionary or a Pandas.Series with the values. The index or key can be the Comuna name and the value an int. The package uses "clcomuna" to get the comuna code, and it creates bins for the values. You can specify the comuna region, even if your data has comunas for the hole country.

pip3 install clchoropleth

• clcomuna
• pandas
• bf4 (BeautifulSoup)
• Python3.3+

>>> import clchoropleth

import clchoropleth
arica = {'ARICA': 27, 'CAMARONES': 13, 'GENERAL LAGOS': 10, 'PUTRE': 8}
data = clchoropleth.prepare_data(arica)
clchoropleth.run(data, "arica.svg", "15")

This should save an SVG file similar to this:

prepare_data

def prepare_data(data, t=70, noNone=False):

Prepares the dataset. If its a dictionary, it turns it into a pandas.Series object. It checks the comuna name and uses its code instead. You can specify the threshold for the clcomuna package. It aggregates by comuna code, so, for exmaple, if you have {'arica':10, 'Arica':11, 'ARICA':12} it will return 'arica':33 as a pd.Series. It can drops the 'None's that you will get for below the threshold. Returns a pd.Series

arica = {'ARICA': 27, 'CAMARONES': 13, 'GENERAL LAGOS': 10, 'PUTRE': 8, 'arica': 33}
clchoropleth.prepare_data(arica)
>>>
    15101    60
    15102    13
    15201     8
    15202    10
    dtype: int64

select_comunas_from_region

def select_comunas_from_region (data, region):

Given the dataset (after prepare_data), returns a subset with only the comunas from the given region. 'region' must be a string.

data = {'ARICA': 27, 'CAMARONES': 13, 'GENERAL LAGOS': 10, 'PUTRE': 8, 'arica': 33, 'iquique': 123, 'santiago': 999}
data_p = prepare_data(data)
data_p
>>>
    01101    123
    13101    999
    15101     60
    15102     13
    15201      8
    15202     10
    dtype: int64
select_comunas_from_region(data_p, "15")
>>>
    15101    60
    15102    13
    15201     8
    15202    10
    dtype: int64

discretize

def discretize(data, bins=5, quantile=False):

Changes the values of the data into a bin number. You can specify a number of bins to be used and if to use 'quintile' based cut or not.

arica
>>>
    15101    60
    15102    13
    15201     8
    15202    10
    dtype: int64
discretize(arica)
>>>
    15101    4
    15102    0
    15201    0
    15202    0
    dtype: category
    Categories (5, int64): [0 < 1 < 2 < 3 < 4]
discretize(arica, 4, True)
>>>
    15101    3
    15102    2
    15201    0
    15202    1
    dtype: category
    Categories (4, int64): [0 < 1 < 2 < 3]

make_map

def make_map(data, filename, region, colors='orange'):

It creates the choropleth map and saves it to filename destination.

Args:

data: The pd.Series with the data. The index needs to be the code and the value the bin number.

filename: The filename of the map we are going to save.

region: we need to specify the region again, as a string ("09"), to open the original svg file and modify it.

colors: it can be a string, if is one of the 5 pre-set color lists: orange, blue, green, red, pruple. Or it can be a list of strings, with each string being a color. For example:['#ffffcc','#c2e699','#78c679','#31a354','#006837'] It needs to have the same length as the number of bins. For more colors: http://colorbrewer2.org/

Returns:

It saves the map. No return valule.

run

def run(data, filename, region, colors='orange', bins=5, quantile=False):

Is a wrapper function that runs all the other functions of the module, except for 'preapare_data'. Doesn't return anything.