learn_gd: Learn Gradient Descent

In bquast/learNN: Examples of Neural Networks

Learn Gradient Descent

Description

Learn Gradient Descent

Usage

learn_gd(X, y, alpha, hiddenSize)

Arguments

X	input data
y	output data
alpha	fraction of gradient descent
hiddenSize	size of the hidden layer

References

https://qua.st/handcoding-gradient-descent/ https://iamtrask.github.io/2015/07/27/python-network-part2/

Examples

# input dataset
X = matrix(c(0,0,1,
             0,1,1,
             1,0,1,
             1,1,1), nrow=4, byrow=TRUE)

# output dataset
y = matrix(c(0,
             1,
             1,
             0), nrow=4)

# set parameters
alpha = 0.1
hiddenSize = 32
# also try using:
# alphas = c(0.001,0.01,0.1,1,10,100,1000)
# for (alpha in alphas) {
#   print(paste("Training With Alpha", alpha))
#   learn_gd(X, y, alpha, hiddenSize)         }

# run gradient descent function
learn_gd(X, y, alpha, hiddenSize)

bquast/learNN documentation built on June 26, 2022, 2:08 a.m.