Welcome to keras-pandas! I’d recommend that you start with the keras-pandas section & Quick Start.


Automater.Automater([data_type_dict, …]) An Automater object, allows users to rapidly build and iterate on deep learning models.
data_types.Abstract.AbstractDatatype() Interface for all future datatypes
data_types.Boolean.Boolean() Support for boolean variables, such as owns_home: [True, False, True], or existing_acct: [True, True, False].
data_types.Categorical.Categorical() Support for categorical variables, such as fruits: [‘apple’, ‘banana’, ‘coconut’], or home_ownership: [‘rent’, ‘own].
data_types.Numerical.Numerical() Support for numerical variables, such as annual_salary: [175000, 105000, 30000000], or countries_visited: [ 1, 7, 22, 183, 12].
data_types.Text.Text() Support for text variables, such as title: [‘The count of Monte Cristo’, ‘Alice in Wonderland’], or article_text: [‘Politicians in deadlock over latest international disagreement…’, ‘42 is truly the answer to life, the universe and everything according to a study by British …’]`
data_types.TimeSeries.TimeSeries() Support for time series data, such as previous_day_closes: [[123, 3, 0], [777, 42, 0]] or last_three_purchase_prices: [[222, 111, 891], [12312312, 412412, 12]]
lib Library / helper functions for Keras-Pandas
transformations SKLearn-compliant transformers, for use as part of pipelines