Create fit lines for data plots in R

cost_func.r

# Linear func for varying theta0 and theta1
h <- function(x, h0, h1){
  return (h0 + h1*x) 
}

# Cost function for fitting data
J <- function(h0, h1, m, d) {
	y = seq()
	for (i in seq(m)) {
		y[i] = (h(d[i,1], h0, h1) - d[i,2])^2 
	}
	return ((1/(2*m))*sum(y))
}

# Minimize given a range ra, samplesize m and data 2 col matrix d
minimize <- function(ra, m, d) {
	cur = 1000000000000000
	fi = 0
	fj = 0
	for (i in ra) {
		for (j in ra) {
			curj = J(i, j, m, d)
			if (curj < cur) {
				cur = curj
				fi = i
				fj = j
			}
		}
	}
  # Return the lowest cost and the theta0/theta1 pair that created it
	return (c(cur, fi, fj))
}

# Generate Y values for a sequence using H and the
# Returned/minimized theta0 and theta1 of minimize
genline <- function(min, max, theta0, theta1) {
	y = seq(max-min)
	for (i in seq(from=min, to=max)) y[i-min+1] = h(i, theta0, theta1)
	return (y)
}