Hi everyone,

Need little clarification on standardisation of data

if ’ x’ is a numpy array of real number ranging form 1000,2000(taken 20 random points)

given formula for standardization = (x - x.min() ) / ( x.max() - x.min() ) so that the values range form 0 to 1;

what if we use x / x.max() even this also range between 0 to 1;

[consider : if any negative values in x can be shifted ex: -2,3,5,6 to 0,5,7,8 ]

when i plot ‘x’ values ranging form 10000 to 2000:

’ – ’ refer to the line with this formula = x / x.max()

and solid line with this formula = (x - x.min() ) / ( x.max() - x.min() )

is there any good reason we should not use this formula = x / x.max()