[jamesblonde]

Normalization is a model-dependent transformation and happens after the feature store - needs to be consistent between training and inference pipelines.

[nikhilsimha]

Normalization is overloaded. I was referring to schema normalization (3NF etc) not feature normalization - like standard scaling etc.

[jamesblonde]

Ok, but star schema is denormalized. Snowflake is normalized.

[nikhilsimha]

To be pedantic, even in star schema - the dim tables are denormalized, fact tables are not.

I agree that my statement would be much better if used snowflake schema instead.