'Normalizing my data without knowing population estimates
I am currently working on a deep learning project. My input is of the shape (20,938). I'm trying to normalize the data and then scale them between 0 and 1. Now my dataset contains around 53k examples.
It is computationally expensive for me to read every data point and calculate the overall min, max, mean and standard deviation.
Does anyone know a better approach ?
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|