In code above( [0404_GDAAlgorithms]), for both Momentum
and NAG
, dw
and db's
are not being normalised (division with number of data points).
Is this how it is supposed to be or just a writing error?
In code above( [0404_GDAAlgorithms]), for both Momentum
and NAG
, dw
and db's
are not being normalised (division with number of data points).
Is this how it is supposed to be or just a writing error?