Hi,
I am studying logistic regression and gradient ascent and have seen it used with a cost function and without one. Could anyone tell me why you would use a cost function? It seems just as effective without one.
alpha = .05
h = data * weights
error = labels - sigmoid(h)...
Thanks for the help. It makes more sense to me now. I was getting confused because I was applying the chain rule and not simplifying enough, I think. For instance using your example.
d/dx = √(x^2 + 25) = (x^2 + 25)^1/2 = 1/2(x^2 + 25)^-1/2 * (2x) = x(x^2 + 25)^-1/2 =
x/(x^2 + 25)^1/2
btw...
That is the solution I copied from the text. I'm teaching myself this, so sometimes it helps to have someone be able to elaborate on the why part of what the text is showing.
To me I would think that since I'm taking the derivative with respect to x1, I take the derivative of x1 which is -4...
Homework Statement
I'm trying to follow my textbook on an application of the chain rule.
Two objects are traveling in elliptical paths given by the following parametric equation.
x1 = 4 cos t
x2 = 2 sin 2t
y1 = 2 sin t
y2 = 3 cos 2t
At what rate is the distance between the two...
Hi,
I was searching the forum about comparing continuous probability distributions and came across this post back in 2005.
"You could make two variables X(t) = value of the "true" disrtibution (expensive simulation) at point t and Y(t) = value of the alternative dist. (practical simulation)...