autonormalize 1.0.1

a library for automated table normalization

AutoNormalize

AutoNormalize is a Python library for automated datatable normalization. It allows you to build an EntitySet from a single denormalized table and generate features for machine learning using Featuretools.

Getting Started

• Install
• Demos
• API Reference

Install

pip install featuretools[autonormalize]

Uninstall

pip uninstall autonormalize

Demos

• Blog Post
• Machine Learning Demo with Featuretools
• Kaggle Liquor Sales Dataset Demo
• Demo with Editing Dependencies
• Kaggle Food Production Dataset Demo

API Reference

auto_entityset

auto_entityset(df, accuracy=0.98, index=None, name=None, time_index=None)

Creates a normalized entityset from a dataframe.

Arguments:

• df (pd.Dataframe) : the dataframe containing data

• accuracy (0 < float <= 1.00; default = 0.98) : the accuracy threshold required in order to conclude a dependency (i.e. with accuracy = 0.98, 0.98 of the rows must hold true the dependency LHS --> RHS)

• index (str, optional) : name of column that is intended index of df

• name (str, optional) : the name of created EntitySet

• time_index (str, optional) : name of time column in the dataframe.

Returns:

• entityset (ft.EntitySet) : created entity set

find_dependencies

find_dependencies(df, accuracy=0.98, index=None)

Finds dependencies within dataframe with the DFD search algorithm.

Returns:

• dependencies (Dependencies) : the dependencies found in the data within the contraints provided

normalize_dataframe

normalize_dataframe(df, dependencies)

Normalizes dataframe based on the dependencies given. Keys for the newly created DataFrames can only be columns that are strings, ints, or categories. Keys are chosen according to the priority:

1. shortest lenghts
2. has "id" in some form in the name of an attribute
3. has attribute furthest to left in the table

Returns:

• new_dfs (list[pd.DataFrame]) : list of new dataframes

make_entityset

make_entityset(df, dependencies, name=None, time_index=None)

Creates a normalized EntitySet from dataframe based on the dependencies given. Keys are chosen in the same fashion as for normalize_dataframeand a new index will be created if any key has more than a single attribute.

Returns:

• entityset (ft.EntitySet) : created EntitySet

normalize_entity

normalize_entity(es, accuracy=0.98)

Returns a new normalized EntitySet from an EntitySet with a single entity.

Arguments:

• es (ft.EntitySet) : EntitySet with a single entity to normalize

Returns:

• new_es (ft.EntitySet) : new normalized EntitySet

Feature Labs

AutoNormalize is an open source project created by Feature Labs. To see the other open source projects we're working on visit Feature Labs Open Source. If building impactful data science pipelines is important to you or your business, please get in touch.