piper.custom.factorize¶
-
piper.custom.factorize(series: pandas.core.series.Series, categories: Optional[List] = None, ordered: int = False) → pandas.core.series.Series[source]¶ factorize / make column a categorical dtype
- Parameters
series – pd.Series object to be converted to categorical
categories – list of unique category values within pd.Series
ordered – If true, categorical is ordered.
- Returns
Returned series with categorical data type
- Return type
pd.Series
Examples
cat_order = ['Tops & Blouses', 'Beachwear', 'Footwear', 'Jeans', 'Sportswear'] %%piper sample_sales() >> assign(product=lambda x: factorize(x['product'], categories=cat_order, ordered=True)) >> group_by(['location', 'product']) >> summarise(Total=('actual_sales', 'sum')) >> unstack() >> flatten_cols(remove_prefix='Total') >> head(tablefmt='plain') location Tops & Blouses Beachwear Footwear Jeans Sportswear London 339236 388762 274674 404440 291561 Milan 523052 368373 444624 364343 319199 Paris 481787 464725 383093 178117 150222