federicogarza / covidmx

Python API to get information about COVID-19 in México.

covidmx

Python API to get information about COVID-19 in México.

Requirements

more-itertools>=6.0.0
pandas>=0.25.2
Unidecode>=1.1.1
requests==2.21.0
xlrd==1.2.0
mapsmx==0.0.3
matplotlib==3.0.3
mapclassify==2.2.0
descartes==1.1.0

How to install

pip install covidmx

How to use

Dirección General de Epidemiología

The mexican Dirección General de Epidemiología has released open data about COVID-19 in México. This source contains information at the individual level such as gender, municipality and health status (smoker, obesity, etc). The package covidmx now can handle this source as default. Some variables are encoded as integers and the source also includes a data dictionary with all relevant information. When you pass clean=True (default option) returns the decoded data. You can also have access to the catalogue using return_catalogo=True and to the description of each one of the variables with return_descripcion=True. When you use some of this parameters, the API returns a tuple.

from covidmx import CovidMX

covid_dge_data = CovidMX().get_data()
raw_dge_data = CovidMX(clean=False).get_data()
covid_dge_data, catalogo_data = CovidMX(return_catalogo=True).get_data()
covid_dge_data, descripcion_data = CovidMX(return_descripcion=True).get_data()
covid_dge_data, catalogo_data, descripcion_data = CovidMX(return_catalogo=True, return_descripcion=True).get_data()

To get historical data use:

covid_dge_data = CovidMX(date='12-04-2020').get_data()

Default date format is %d-%m-%Y, but you can also use a particular format with:

covid_dge_data = CovidMX(date='2020-04-12', date_format='%Y-%m-%d').get_data()

Plot module

As of version 0.3.0, covidmx includes a module to create maps of different COVID-19 status at the national and state levels, with the possibility of including municipalities (using information of the Dirección General de Epidemiologia).

from covidmx import CovidMX

dge_plot = CovidMX().get_plot()

You can check available status and available states using:

dge_plot.available_states

array(['MÉXICO', 'CIUDAD DE MÉXICO', 'TAMAULIPAS', 'BAJA CALIFORNIA',
       'YUCATÁN', 'GUERRERO', 'BAJA CALIFORNIA SUR', 'JALISCO',
       'NUEVO LEÓN', 'SONORA', 'VERACRUZ DE IGNACIO DE LA LLAVE',
       'PUEBLA', 'CAMPECHE', 'GUANAJUATO', 'SAN LUIS POTOSÍ',
       'MICHOACÁN DE OCAMPO', 'COAHUILA DE ZARAGOZA', 'QUERÉTARO',
       'AGUASCALIENTES', 'TABASCO', 'HIDALGO', 'ZACATECAS', 'DURANGO',
       'CHIHUAHUA', 'CHIAPAS', 'SINALOA', 'QUINTANA ROO', 'MORELOS',
       'TLAXCALA', 'NAYARIT', 'OAXACA', 'COLIMA'], dtype=object)

dge_plot.available_status

['confirmados', 'negativos', 'sospechosos', 'muertos']

To plot a national map just use:

dge_plot.plot_map(status='confirmados')

If you want to include municipalities use:

dge_plot.plot_map(status='confirmados', add_municipalities=True)

You can pass a particular state filling the state argument with a valid name included in the available_states attribute:

dge_plot.plot_map(status='confirmados', state='CIUDAD DE MÉXICO', add_municipalities=True)

state='CIUDAD DE MÉXICO'	state='JALISCO'	state='MORELOS'	state='MÉXICO'

Finally you can plot another interest variable (according to available_status attribute):

dge_plot.plot_map(status='sospechosos', add_municipalities=True)

You can save your maps using save_file_name:

dge_plot.plot_map(status='sospechosos', add_municipalities=True, save_file_name='sospechosos-nacional.png')

Serendipia

Serendipia publishes daily information of the mexican Secretaría de Salud about covid in open format (.csv). This api downloads this data easily, making it useful for task automation.

from covidmx import CovidMX

latest_published_data = CovidMX(source='Serendipia').get_data()

Then CovidMX instances a Serendipia class, searches the latest published data for both confirmed and suspects individuals and finally clean the data. Nevertheless, a more specific search can be conducted (see docs for details).

raw_data = CovidMX(source='Serendipia', clean=False).get_data()
confirmed = CovidMX(source='Serendipia', kind="confirmed").get_data()
suspects = CovidMX(source='Serendipia',kind="suspects").get_data()
particular_published_date = CovidMX(source='Serendipia', date='2020-04-10', date_format='%Y-%m-%d').get_data()

Cite as

• Federico Garza Ramírez (2020). covidmx: Python API to get information about COVID-19 in México. Python package version 0.3.1. https://github.com/FedericoGarza/covidmx.

Acknowledgments

• Max Mergenthaler
• Mario Jimenez

Release information

0.3.1 (Current version)

• 2020-06-01
• Updated new urls from serendipia source. (Thanks to Mario Jimenez.)

0.3.0

• 2020-04-26.
• Includes a plot module at state and municipality leveles.
• Includes a better handling of encodings. (Thanks to Mario Jimenez.)

0.2.5

• 2020-04-20. The Dirección General de Epidemiología:
  • Added an id column.
  • Released historical information.
  • Now the API can handle this changes.

0.2.4

• 2020-04-16. The Dirección General de Epidemiología source renamed two columns:
  • HABLA_LENGUA_INDI -> HABLA_LENGUA_INDIG (column name and description are now homologated)
  • OTRA_CON -> OTRA_COM
  • Now the API can handle this change.

0.2.3

• Now works with python3.5+.
• Using clean=True returns encoded data instead of decoded data without cleaning columns (as works in 0.2.0 and 0.2.1).

0.2.1

• Minor changes to README.

0.2.0

• Added new source: Dirección General de Epidemiología. Default source.
• Only works with python3.7+.

0.1.1

• Minor changes to README.

0.1.0

First realease.

• Only one source, Serendipia. Default source.

covidmx's Issues

Cannot read the data

I'm trying to run the covid_dge_data = CovidMX().get_data() line, but it get the runtime error Cannot read the data. Also get the following: URLError: <urlopen error [Errno 60] Operation timed out>.

I've failed to pull request at the dge.py file to change the URL_DESCRIPTION and URL_HISTORICALfor the updated ones:

URL_DESCRIPTION = 'http://epidemiologia.salud.gob.mx/gobmx/salud/datos_abiertos/diccionario_datos_covid19.zip' URL_HISTORICAL = 'http://epidemiologia.salud.gob.mx/gobmx/salud/datos_abiertos/datos_abiertos_covid19.zip'

Fails when installing in windows

I was trying to update the last release to date version 0.3.0 and I got this error when installing:

A GDAL API version must be specified. Provide a path to gdal-config using a GDAL_CONFIG environment variable or use a GDAL_VERSION environment variable.
----------------------------------------
ERROR: Command errored out with exit status 1: python setup.py egg_info Check the logs for full command output.

Error in dge.py file

When running Python to get information through the get_data method of the CovidMX object, it returns the following error:

raceback (most recent call last):
File "c:/Users/Alejandro Rebollo/Documents/Covid19MX.py", line 4, in
raw_dge_data = CovidMX(clean=False).get_data()
File "C:\Users\Alejandro Rebollo\AppData\Local\Programs\Python\Python38-32\lib\site-packages\covidmx\dge.py", line 33, in get_data
df, catalogo, descripcion = self.read_data()
File "C:\Users\Alejandro Rebollo\AppData\Local\Programs\Python\Python38-32\lib\site-packages\covidmx\dge.py", line 59, in read_data
raise RuntimeError('Cannot read data description.')
RuntimeError: Cannot read data description.

¿Como citar esta info?

Sería bueno agregar la forma correcta de citar este repo.

Error encoding='UTF-8'

covid_dge_data = CovidMX().get_data()

data = pd.read_csv(url_data, encoding='UTF-8')
73 except BaseException:
---> 74 raise RuntimeError('Cannot read the data.')
75
76 try:

RuntimeError: Cannot read the data.

Add deaths/cases multiplier

Hola!! I am about to do the weekly multipliers (https://www.facebook.com/gapminder.org/posts/3397004387001162) a very simple yet not common metric of evaluation (I have not seen it for Mexico, yet). I will start by states but then do it per metropolitan regions or municipalities. Let me know if you have suggestions or if you think it could be useful to add this in repo.

KeyError: 'HABLA_LENGUA_INDIG'

Hi! I'm running the following command:
covid_dge_data, catalogo_data, descripcion_data = CovidMX(return_catalogo=True, return_descripcion=True).get_data()
And I am getting the following error:

KeyError Traceback (most recent call last)
in
3 #covid_dge_data, catalogo_data = CovidMX(return_catalogo=True).get_data()
4 #covid_dge_data, descripcion_data = CovidMX(return_descripcion=True).get_data()
----> 5 covid_dge_data, catalogo_data, descripcion_data = CovidMX(return_catalogo=True, return_descripcion=True).get_data()

~\AppData\Local\Continuum\anaconda3\lib\site-packages\covidmx\dge.py in get_data(self)
36 if self.clean:
37 print('Cleaning data')
---> 38 df = self.clean_data(df, catalogo, descripcion)
39
40 print('Ready!')

~\AppData\Local\Continuum\anaconda3\lib\site-packages\covidmx\dge.py in clean_data(self, df, catalogo, descripcion)
172 for col in df.columns:
173 df[col] = self.replace_values(
--> 174 df, col, desc_dict, catalogo_dict)
175
176

~\AppData\Local\Continuum\anaconda3\lib\site-packages\covidmx\dge.py in replace_values(self, data, col_name, desc_dict, catalogo_dict)
130 def replace_values(self, data, col_name, desc_dict, catalogo_dict):
131
--> 132 formato = desc_dict[col_name]
133 if 'FECHA' in col_name:
134 return pd.to_datetime(

KeyError: 'HABLA_LENGUA_INDIG'

unexpected keyword argument 'return_catalogo' / 'return_descripcion'

Hi,

when I run covid_dge_data, catalogo_data = CovidMX(return_catalogo=True).get_data() get:

Traceback (most recent call last) <ipython-input-8-9c05e955ccf6> in <module>() 3 covid_dge_data = CovidMX().get_data() 4 raw_dge_data = CovidMX(clean=False).get_data() ----> 5 covid_dge_data, catalogo_data = CovidMX(return_catalogo=True).get_data() 6 covid_dge_data, descripcion_data = CovidMX(return_descripcion=True).get_data() 7 covid_dge_data, catalogo_data, descripcion_data = CovidMX(return_catalogo=True, return_descripcion=True).get_data() /usr/local/lib/python3.6/dist-packages/covidmx/covidmx.py in CovidMX(source, **kwargs) 27 28 if source == "Serendipia": ---> 29 return Serendipia(**kwargs) TypeError: __init__() got an unexpected keyword argument 'return_catalogo'

and covid_dge_data, descripcion_data = CovidMX(return_descripcion=True).get_data() get:

Traceback (most recent call last) <ipython-input-22-1b8abfd4e7ee> in <module>() 4 #raw_dge_data = CovidMX(clean=False).get_data() 5 #covid_dge_data, catalogo_data = CovidMX(return_catalogo=True).get_data() ----> 6 covid_dge_data, descripcion_data = CovidMX(return_descripcion=True).get_data() 7 #covid_dge_data, catalogo_data, descripcion_data = CovidMX(return_catalogo=True, return_descripcion=True).get_data() /usr/local/lib/python3.6/dist-packages/covidmx/covidmx.py in CovidMX(source, **kwargs) 27 28 if source == "Serendipia": ---> 29 return Serendipia(**kwargs) TypeError: __init__() got an unexpected keyword argument 'return_descripcion'

By the way, Unidecode package is required, so could be usefull to add this line to install instructions:

pip install unidecode

Catalogos missing

When I am calling this line

covid_dge_data = CovidMX().get_data()

I have this error:

"There is no item named 'Catalogos_0412.xlsx' in the archive"

long time to read dataset from web

It is quite long the web dataset reading, for debugin or adding new functions. I have a quick fix for it garciaguevara@ca80395. If you think it is useful I could improve/clean it and do a merge request. Nice and useful repo BTW!