Calculate feature matrix can fail when comparing categorical columns with different categories using Equal and NotEqual primitives

[thehomebrewnerd]

Feature matrix calculation can fail in some cases when using Equal and NotEqual primitives to compare categorical columns with different categories. An example is included below using the attached data.

SBAcase_cleaned_train.csv

Code Sample, a copy-pastable example to reproduce your bug.

import pandas as pd
import featuretools as ft

df = pd.read_csv("SBAcase_cleaned_train.csv")

es = ft.EntitySet()

logical_types = {
    "NAICS": "Categorical",
    "NewExist": "Categorical",
}
es.add_dataframe(dataframe_name="df", dataframe=df, index="id", make_index=True, logical_types=logical_types)

ft.dfs(entityset=es, target_dataframe_name="df", trans_primitives=["equal", "not_equal"])

TypeError: Categoricals can only be compared if 'categories' are the same.