The order of labels of pandas’ categorical features as CategoricalDtype(order=True) might be used by estimators. For example:

print(houses['quality'].unique())
[poor, fair, typical, good, excellent]
Categories (4, object): [poor < fair < typical < good < excellent]

Note how order is embedded in the data type above.

I was expecting codes like these:
0 poor
1 fair
2 typical
3 good
4 excellent

And I'm sure estimators would provide more meaningful results if such order was used.

But LabelEncoder gives random integer codes, probably using data as it comes:
3 poor
1 fair
0 typical
4 good
2 excellent

Thank you in advance