Last updated:
0 purchases
pyholidaycalendar 0.0.6
py_holiday_calendar is a Python package containing methods to manipulate date columns of pandas dataframes
adjusting for business days and holidays.
Usage
Just run the following snippet in your python env and start using the package.
pip install -v py-holiday-calendar
Docs
You can find the docs here.
Example
import pandas as pd
from py_holiday_calendar import py_holiday_calendar as check
# Making a dummy dataframe
df = {
'day': ['1/1/2018', '4/8/2019', '12/13/2019', '1/2/2019'],
'temp': [44, 45, 87, 66],
'str': ['Brake', 'Data', 'Gimp', 'data'],
'drake': ['eminem', 'marshall', 'drogon', 'eminem'],
'day1': ['1/5/2018', '4/12/2019', '12/31/2019', '1/8/2019'],
}
df = pd.DataFrame(df)
# Explicitly conveerting the columns to datetime (with which the package works)
df['day'] = pd.to_datetime(df['day'])
df['day1'] = pd.to_datetime(df['day1'])
# Pass in the arguments to cal diff between two date columns
check.calc_diff_with_holidays(df, 'day', 'day1', 'new_date')
print(df)
# Can also pass in custom business days as well as holidays
check.calc_diff_with_holidays(df, 'day', 'day1', 'new_date', workdays=[MO, TU, WE, TH])
print(df)
# Add custom business days to the provided column
check.add_bus_days_with_holidays(df, 'day', 'temp', 1)
print(df)
To-Dos
Add remaining methods.
Add the tests.
For personal and professional use. You cannot resell or redistribute these repositories in their original state.
There are no reviews.