magicpandas 0.0.4

A dead simple interface for manipulating pandas dataframes

##### UNDER CONSTRUCTION #####

magicpandas

magicpandas makes working with pandas dead simple.

Main Features

• MagicDataFrame subclasses DataFrame to make existing methods more intuitive as well as add new methods
• MagicDataFrame adds verbose labels that are used by default when displaying data
• MagicDataFrame supports Django ORM
  • inspectdf uses DataFrame column types to produce a Django model class (cf. inspectdb)
  • to_django saves the DataFrame to SQL using Django ORM's bulk_update and bulk_create.

Installation

pip install magicpandas

Examples

from magicpandas import MagicDataFrame
mdf = MagicDataFrame(df)
mdf2 = mdf.drop('*e', axis=1) # df2 drops all columns ending in "e"
mdf.browse() # opens the DataFrame in MS Excel with nice formatting
mdf.browse(client='webbrowser') # opens the DataFrame as html displayed in Chrome with nice formatting
mdf.graph()  # Opens a graph in Chrome using the excellent Altair library using sensible encodings
mdf.inspect_for_django()  # prints text that corresponds to a Django model definition

Philosophy

• Convention over configuration
• Employs Altair rather than than matplotlib because it's web first and based on the Grammar of Graphics
• Employs MS Excel for data browsing
• There should be little gap between intention and implementation