[Noisebridge-discuss] Tomorrow night: ML Wednesday!
Josh Myer
josh at joshisanerd.com
Tue Mar 24 22:38:17 UTC 2009
Hey guys, tomorrow night is another machine learning Wednesday! This
week, we'll be taking last week's perceptron and making its output a
sigmoid. Once we've done that, we'll move on multilayer perceptrons,
which should let us build full-on neural networks.
For homework, you'll need to code up the following function:
sigmoid(beta, x) = 1/(1+e^(-beta*x))
Here's the vanilla ruby:
def sigmoid(beta, x)
beta = beta.to_f
x = x.to_f
return 1.0/(1.0+Math.exp(-beta*x))
end
Happy hacking, see you guys Wednesday!
--
Josh Myer 650.248.3796
josh at joshisanerd.com
More information about the Noisebridge-discuss
mailing list