0
$\begingroup$

I'm analyzing a dataset that contains a feature 'street_name' with 5980 unique values. I used the LeaveOneOutEncoder class for encoding, but I noticed that the cardinality reduced a lot. There are now 340 unique values. In another feature called 'name' the cardinality increased. Before there were 309 unique values and after encoding 1239 nunque. So, I would like to know if this could be a problem when training the models. What are the effects of this cardinality changing on the model's training

Improve this question
$\endgroup$

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.