usually a int8 tensor ranges from 0 to 255
but in float, the value range is not fixed but it’s dynamic
meaning that the levels between the min and max values are what determines floating point
can I make a fixed range between eg. 0 and 1
so that it is not dynamic