RMSE or NRMSE ??

Hello every one

I have a question about RMSE in function approximation with neural networks .

I have Normalized ALL of my Data including train and test data and input and targets in [-1 1]. and I did not denormalize Data in the end including inputs and output of my model and targets and in the end I want to calculate the RMSE value between model outputs and targets.

My question is that should I Divide the RMSE obtained value by 2 or not ? because I normalized Data in [-1 1].

with best Regards.