Pandas groupby using only year and month

Question

I have a Python program using Pandas, which reads two dataframes, obtained in the following links:

Casos-positivos-diarios-en-San-Nicolas-de-los-Garza-Promedio-movil-de-7-dias: https://datamexico.org/es/profile/geo/san-nicolas-de-los-garza#covid19-evolucion

Denuncias-segun-bien-afectado-en-San-Nicolas-de-los-GarzaClic-en-el-grafico-para-seleccionar: https://datamexico.org/es/profile/geo/san-nicolas-de-los-garza#seguridad-publica-denuncias

What I currently want to do is a groupby in the "covid" dataframe with the same dates, having a sum of these. Regardless, no method has worked out, which regularly prints an error indicating that I should be using a syntaxis for "PeriodIndex". Does anyone have a suggestion or solution? Thanks in advance.

import pandas as pd import numpy as np import matplotlib.pyplot as plt %matplotlib notebook #csv for the covid cases covid = pd.read_csv('Casos-positivos-diarios-en-San-Nicolas-de-los-Garza-Promedio-movil-de-7-dias.csv') #csv for complaints comp = pd.read_csv('Denuncias-segun-bien-afectado-en-San-Nicolas-de-los-GarzaClic-en-el-grafico-para-seleccionar.csv') #cleaning data in both dataframes #keeping only the relevant columns covid = covid[['Month','Daily Cases']] comp = comp[['Month','Affected Legal Good', 'Value']] #changing the labels from spanish to english comp['Affected Legal Good'].replace({'Patrimonio': 'Heritage', 'Familia':'Family', 'Libertad y Seguridad Sexual':'Sexual Freedom and Safety', 'Sociedad':'Society', 'Vida e Integridad Corporal':'Life and Bodily Integrity', 'Libertad Personal':'Personal Freedom', 'Otros Bienes Jurídicos Afectados (Del Fuero Común)':'Other Affected Legal Assets (Common Jurisdiction)'}, inplace=True, regex=True) #changing the month types to dates covid['Month'] = pd.to_datetime(covid['Month']) covid['Month'] = covid['Month'].dt.to_period('M') covid

Will you please be able to post the error message ? and the group by code — AmilaMGunawardana
– AmilaMGunawardana, Commented Mar 28, 2022 at 23:20
Of course, sorry for that, I lost the code but managed to find it again. covid.groupby(pd.Grouper(key='Month', axis=0, freq='M')).sum() and the error is axis must be a DatetimeIndex, but got an instance of 'PeriodIndex' — AvocadoToast
– AvocadoToast, Commented Mar 28, 2022 at 23:46

AmilaMGunawardana · Accepted Answer · 2022-03-29 16:13:10Z

You can simply usen group by statement.Timegrouper by default converts it to datetime

import pandas as pd import numpy as np import matplotlib.pyplot as plt %matplotlib notebook #csv for the covid cases covid = pd.read_csv('Casos-positivos-diarios-en-San-Nicolas-de-los-Garza-Promedio-movil-de-7-dias.csv') covid = covid.groupby(['Month'])['Daily Cases'].sum() covid = covid.reset_index() # #changing the month types to dates covid['Month'] = pd.to_datetime(covid['Month']) covid['Month'] = covid['Month'].dt.to_period('M') covid

Collectives™ on Stack Overflow

Pandas groupby using only year and month

1 Answer 1

2 Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Related