Neural Network Training (Part 3): Gradient Calculation

[101LiquidNitrogen]

calculus great im fucked

[00YURIN00]

doesn't the derivative of a constant equal zero ?

i don't get how u get to the 0.045 ... :s

[Patterion]

@00YURIN00

It actually is not a derivative of constant, but derivative of the transfer (sigmoid) function then evaluated with the 1.13 constant as parameter for x.

That is:

f=1/(1+e^(-x)) =>

f'=(e^x)/[e^(2x)+2e^(x)+1] =>

f'(1.13)=0.1845 and then

d=0.25*0.1845=0.046.

f is sigmoid function and f' is its derivative.

The 1.13 is actually 1.1278 so that's why 0.046 instead of 0.045.

[00YURIN00]

@Patterion

Hey yeah thanks,  i had allready figured it out.

Seems like a lot of neural net- teachers assume u know the derivative of the sigmoidfunction by heart or something :p

thanx !

[jasuncion1]

Really great tutorials!