fusetools.ml_tools.Prep

class fusetools.ml_tools.Prep

Bases: object

Functions for preparing data for machine learning tasks.

Methods

label_encode_df	Assigns an incremental number for each string value and then converts the new number to a float so we can add the missing value (NaNs) back in for later imputation.
make_model_feature_df	Creates Pandas DataFrame with cumulatively trained combined feature names, coefficients, absolute coefficients in order of trained together.

classmethod label_encode_df(df, col, col_new)

Assigns an incremental number for each string value and then converts the new number to a float so we can add the missing value (NaNs) back in for later imputation.

Parameters

• df – Pandas DataFrame with atleast one feature to encode.

• col – Name of column to encode.

• col_new – Name of new, encoded column.

Returns

Pandas DataFrame with new, encoded column.

classmethod make_model_feature_df(df, cat, model_list)

Creates Pandas DataFrame with cumulatively trained combined feature names, coefficients, absolute coefficients in order of trained together.

Parameters

• df – Pandas DataFrame of a fitted SckitLearn estimator’s features, coefficients absolute coefficients.

• cat – Name of ScikitLearn estimator or self defines estimator type.

• model_list – List of ScikitLearn estimators.

Returns

Pandas DataFrame with cumulatively trained combined feature names, coefficients, absolute coefficients and performance placeholders in order of trained together.