Just getting started on the journey, hoping for some helpful explanations.
Specific Question
- If we normalize our DataBunch for training, why don’t we need to normalize the data we are predicting?
General Question
2. Does anyone know of some more detailed tabular examples?
Fast.ai course is such an amazing resource, but it feels like Tabular was covered really quickly, and lots of questions were deferred to be answered later, but I’m not sure when they will actually revisited.