OptHoldoutSize (version 0.1.0.1)

plot.optholdoutsize_emul: Plot estimated cost function using emulation (semiparametric)

Description

Plot estimated cost function, when semiparametric (emulation) method is used for estimation.

Draws posterior mean of cost function as a line and indicates minimum. Also draws mean +/- 3 SE.

Assumes a power-law form of k2 unless parameter k2 is set otherwise.

Usage

# S3 method for optholdoutsize_emul
plot(x, ..., k2form = powerlaw)

Value

No return value; draws a plot only.

Arguments

x

Object of type optholdoutsize_emul

...

Other arguments passed to plot()

k2form

Function governing expected cost to an individual sample given a predictive score fitted to n samples. Must take two arguments: n (number of samples) and theta (parameters). Defaults to a power-law form powerlaw(n,c(a,b,c))=a n^(-b) + c.

Examples

# Simple example

# Parameters
N=100000;
k1=0.3
A=8000; B=1.5; C=0.15; theta=c(A,B,C)

# True mean function
k2_true=function(n) powerlaw(n,theta)

# Values of n for which cost has been estimated
np=50 # this many points
nset=round(runif(np,1,N))
var_k2=runif(np,0.001,0.002)
k2=rnorm(np,mean=k2_true(nset),sd=sqrt(var_k2))

# Compute OHS
res1=optimal_holdout_size_emulation(nset,k2,var_k2,N,k1)

# Plot
plot(res1)