Visualizing mean normalization

One of the labs mentioned that subtracting the mean from a feature centers that feature’s values around 0. I imagine that as a transformation of data points ie. shrinking the range of input feature values along the x-axis (considering there is only one feature and y is the target variable). How can I visualize dividing by standard deviation? Would that translate into shrinking the range of the input feature (x) along the y-axis?

Hello @mvrbiguv,

Mean subtraction can be imagined as shifting of data points along that feature axis, whereas dividing by standard deviation can be imagined as shrinking (if the standard deviation is larger than 1), or expanding (smaller than 1).

Try to visualize it by drawing a few points on a piece of paper, then add to the drawing the data points that have the subtraction done, and then finally add to the dawing the data points that have the division done.