Woodwork provides you with a common DataTable object to use with Featuretools, EvalML, and general ML. A DataTable object contains the physical, logical, and semantic data types present in the data. In addition, it can store metadata about the data.
Install with pip:
python -m pip install woodworkor from the conda-forge channel on conda:
conda install -c conda-forge woodworkBelow is an example of using Woodwork. In this example, a sample dataset of order items is used to create a Woodwork DataTable, specifying the LogicalType for three of the columns.
import woodwork as ww
data = ww.demo.load_retail(nrows=100, return_dataframe=True)
dt = ww.DataTable(data, name='retail')
dt.set_types(logical_types={
'quantity': 'Double',
'customer_name': 'Categorical',
'country': 'Categorical'
})
dt Physical Type Logical Type Semantic Tag(s)
Data Column
order_id Int64 Integer [numeric]
product_id category Categorical [category]
description string NaturalLanguage []
quantity float64 Double [numeric]
order_date datetime64[ns] Datetime []
unit_price float64 Double [numeric]
customer_name category Categorical [category]
country category Categorical [category]
total float64 Double [numeric]
We now have created a Woodwork DataTable with the specified logical types assigned. For columns that did not have a specified logical type value, Woodwork has automatically inferred the logical type based on the underlying data. Additionally, Woodwork has automatically assigned semantic tags to some of the columns, based on the inferred or assigned logical type.
If we wanted to do further analysis on only the columns in this table that have a logical type of Boolean or a semantic tag of numeric we can simply select those columns and access a dataframe containing just those columns:
filtered_df = dt.select(include=['Boolean', 'numeric']).to_dataframe()
filtered_df order_id quantity unit_price total cancelled
0 536365 6.0 4.2075 25.245 False
1 536365 6.0 5.5935 33.561 False
2 536365 8.0 4.5375 36.300 False
3 536365 6.0 5.5935 33.561 False
4 536365 6.0 5.5935 33.561 False
.. ... ... ... ... ...
95 536378 6.0 4.2075 25.245 False
96 536378 120.0 0.6930 83.160 False
97 536378 24.0 0.9075 21.780 False
98 536378 24.0 0.9075 21.780 False
99 536378 24.0 0.9075 21.780 False
As you can see, Woodwork makes it easy to manage typing information for your data, and provides simple interfaces to access only the data you need based on the logical types or semantic tags. Please refer to the Woodwork documentation for more detail on working with Woodwork tables.
