I have trained my model and I have deployed the model which uses category inputs and 2 numeric normalized fields as inputs. When I run the validation on the sample of the data, I can see that the predictions are coming through. In the inference section where for the input form, in the deployed webapp form and and for the deployed curl command I am able to change the category inputs but I am unable to change the numeric fields (year and a negative number). Both of these fields have been normalised when the dataset was configured before training the model.
What input should I look to provide, the actual numeric, the numeric in " " or the normalised value.